New Page 1

Automatic ED/ES phase detection for LVA XA
	The left ventricle can be analyzed using clinical imaging modalities (e.g. X-ray Angiogram). After a series of images is recorded, the different heart phases can be viewed. Two important phases are the end-diastolic (ED) phase, in which the heart has the highest blood volume, and the end-systolic (ES), in which the heart is has the lowest blood volume. There are two ways to find ED/ES phases: • Analysis of the ECG signal. • Analysis of the image series. Because the ECG signal is not always available the focus of this thesis should be at the analysis of the image series. Several problems have to be solved for an accurate ED/ES detection based on the images: • Determine a region of interest, only the left ventricle should be taken into account. • The best ED/ES phase has to be found in a series of ED/ES phases. • X-ray scanners dynamically adjust their intensity based upon the measured signals. Because of the use of contrast fluids, the intensity will change over time. The thesis subject for the student is to develop an automatic detection of the best end-diastolic / end-systolic phase based on an X-ray image-run in C/C++ programming language. In detail the following tasks need to be performed: • Planning of the activities. • Literature research; o Which methods are available, o What are the pros/cons of each method in relation to the application for use. • Development automatic ED/ES phsse detection algorithm; o Several development tools can be used for prototyping, such as Mathematica and/or Matlab, o Documentation of the algorithm, o Implementation of the algorithm in C/C++ into development environment, o Evaluation of the developed algorithm.
Last modified: February 07, 2010	Status is Open


Cardiac Flow and Function – Master projects
	This project proposal describes a large field of research. Together with the student a challenging and well-defined topic within the field can be identified that matches the student’s interests and skills. Problem statement An interventional cardiologist diagnosis the coronary tree and potentially treats stenotic segments during Percutaneous Coronary Interventions (PCI). Especially vulnerable plaques and flow limiting lesions need to be treated. Image analysis methods for assessment of cardiac flow could aid the cardiologist in diagnosing the patient. Identified lesions are treated using balloon-angioplasty and stenting. In order to place the right stent at the right position, automated analysis of the shape and motion of the coronary tree is needed to aid the cardiologist during treatment. Data from other modalities can also be combined with the data from the X-ray system. Fusion of multimodality data is useful as each modality has unique characteristics to image certain features related to the coronary tree. Goals Project proposal 1: • Find/optimize/invent image analysis methods for assessment of cardiac flow Project proposal 2: • Find/optimize/invent automated analyses of the shape and motion of the coronary tree, including automatic labeling of the vessel segments Project proposal 3: • Multimodality fusion of, e.g. intravascular ultrasound with 2D/3D/4D X-ray acquisitions, this would involve registration and segmentation techniques Tasks • Combined clinical and technical literature study • Implement (automated) image analysis methods for image data (based on C-arm X-ray systems) • Optimize selected methods • Evaluate the methods using clinical data and/or phantom data • Write a report and give a final presentation Student profile • Master student Technical University: Biomedical Engineering, Electrical Engineering, Applied Physics • Prior experience in software development (preferably C, C++, C#) or advanced Matlab/Mathematica user • Prior knowledge on cardiovascular disease, cardiovascular imaging, image analysis • Analytical, systematic, independent • Fluent English in both speaking and writing • If the desired prior knowledge is not available, the student should be eager and able to acquire this knowledge efficiently at the start of the project. Application procedure Send your CV and related achievements, period in which you are available and specific research interests to Gert Schoonenberg, CV Innovation, Philips Healthcare (Gert.Schoonenberg@philips.com) and Bart ter Haar Romeny, BIOMIM, BMT, TU/e (B.M.terhaarRomeny@tue.nl).
Last modified: February 03, 2010	Status is Open


Bundle identification for studies of neonates at risk for neurodevelopmental disorders
	Severe premature birth may lead to a disturbed cognition. Approximately 15-25% of preterm children with a gestational age below 30 weeks or a birth weight below 1000 grams have an abnormal cognitive development. Development support at an early age can help to improve cognitive development. However, nowadays it is not possible to distinguish neonates that need this development support from neonates that do not need it. The Universitair Medisch Centrum Utrecht has started a study of preterms using MRI. The main aim of this study is to answer the question whether MRI-observed abnormalities in the brain structures after premature birth can predict abnormalities in cognitive development. Diffusion Tensor Imaging is the only non-invasive MRI technique that allows invivo reconstruction and visualization of white matter structure. DTI data has been acquired of neonates at term equivalent age (i.e., when they were expected to be born). Quantitative comparison of white matter structure needs to be performed to be able to distinguish between preterm neonates with high-risk and low-risk of neurodevelopment disorders. At 2 years of age a study takes place to identify pre terms with and without disorders. Right now the scanned children are getting this evaluation results. A first manual analysis of the DTI data has been done [1]. In this study fiber bundles had to be identified manually, after unsuccessfully trying to apply clustering techniques developed for adult brains by Moberts [2] (see figure 2). It proved to be very difficult to define the parameters for clustering in a stable way. The disadvantage of manual definition is that it is time consuming and suffers from user bias. Therefore, we would like to develop new techniques that are able to identify the main fiber bundles automatically in the neonates scans. In literature atlas based clustering methods that allow the identification of bundles for adult brains have been presented [3]. However, these have not been applied to neonates and the question is whether similar techniques could be used for this data, or to which extend this would be possible. DTI data in neonates suffers from extra constraints such as noise and low anisotropy that will limit the effectiveness of these methods. The BMIA group is developing a tool, DTITool[4], that allows the visualization and analysis of DTI data. The new algorithms developed in this project should be part of this tool. This tool is developed based on VTK and Qt libraries and using C++ as Task • Literature research of automatic bundle detection for fiber tracking • Analysis and design of techniques that facilitate the quantitative analysis of preterm data. • Implementation of the techniques in the DTITool • Evaluation of the results with a study on the preterm neonates given the developed techniques. [1] Gijs Hoskam, Master Thesis Quantification of DTI in preterm children Bmia.bmt.tue.nl/Education/Master [2] Bart Moberts, Master Thesis Hierarchical Visualization using Fiber Clustering Bmia.bmt.tue.nl/Education/Master [3] Lauren O'Donnell, C.-F. Westin High-Dimensional White Matter Atlas Generation and Group Analysis Ninth International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI'06), 243-251 (2006) [4] http://bmia.bmt.tue.nl/Research/MVIAV/DTI
Last modified: January 24, 2010	Status is Ongoing


Voxel Classification in High Angular Resolution Diffusion Imaging
	The limitations of DTI (Diffusion Tensor Imaging) for resolving voxels in the white matter of the brain with complex fiber structure (i.e. kissing, crossing, diverging fibers) is already known due to the simple Gaussian model for the Probability Density Function of the water molecules motion in the brain. HARDI (High Angular Resolution Diffusion Imaging ) is a novelty that goes beyond the simple DTI model, and gives more insight to the local information within a voxel. One of the biggest drawbacks of HARDI (w.r.t visualization and interactivity) is the multidimensionality of the data and relatively poor (glyph based) and quite slow visualization. On the other hand, DTI has proved quite accurate in depicting the voxels with single fibre orientation, (approximately 1/3 of the white matter of the brain) and has already established visualization techniques which are quite fast and robust. Then why not combining the best of these two methods?! The proposal of this project would be classification of the voxels in the white matter of the brain, depending on the order of high-dimensionality (order of Spherical Harmonics or High Order Tensors) to depict the voxel compositions. There has been previous work on classification of voxels with SH and HOT approaches, so the student will have to make an overview of existing techniques, find the most suitable (for the methods that we are working on like QBall and Diffusion Orientation Transform) and make a decision when to use DTI and when HARDI . That would rapidly improve the speed of the visualization and help in fibertracking in HARDI (which is a novelty and seems quite challenging), by restricting the probabilistic decision making, only in voxels with more complex structure, whereas using simple deterministic fibertracking where DTI is enough. Other possibility/extension: Apply developed segmentation techniques (used for DTI till now) to this combined DTI/HARDI framework, e.g define combined similarity measures, i.e. using DTI and/or HARDI information depending on the classification. Knowledge: Mathematics (Spherical Harmonics, Fourier Transform and similar), basic programming skills in C++.
Last modified: November 12, 2009	Status is Ongoing


Optimal Acquisition Schemes in High Angular Diffusion Imaging
	The recent challenge in diffusion imaging is to find acquisition schemes and analysis approaches that can represent non-Gaussian diffusion profiles in a clinically feasible measurement time. Many High Angular Resolution Diffusion Imaging Techniques have been proposed, but only a small part of them can be feasible in this kind of setting. Previous work contains investigation of the effect of b-value and the number of gradient vector directions on Q-ball imaging and the Diffusion Orientation Transform (DOT) in a structured way using computational simulations, hardware crossing-fiber diffusion phantoms, and in-vivo brain scans. In this project similar work has to be continued, adding the state-of-the-art Spherical Deconvolution as well as Fiber- ODF, as a subject of comparison. Further analysis should be done in finding the optimal regularization scheme and range of parameters for the different HARDI techniques. The work will be mainly consisting of implementing the new HARDI techniques Spherical Deconvolution and Fiber ODF, and do statistical analysis and comparison as in Optimal Acquisition Schemes in High Angular Resolution Diffusion Imaging - V. Prckovska et al. Knowledge of optimal acquisition schemes for HARDI can improve the utility of diffusion weighted MR imaging in the clinical setting for the diagnosis of white matter diseases and presurgical planning. Knowledge: Mathematics (Spherical Harmonics, Fourier Transform and similar), basic programming skills in C++, Mathematica
Last modified: November 12, 2009	Status is Finished


Improve, Assess, Validate and Bench Mark Our Stereotactic Frame Registration Method
	For the project description in PDF with figures please click here.
Last modified: November 11, 2009	Status is Open


Motion Extraction Based on Multiscale Anchor Point Movement and a Soft Constraint for Greyscale Conservation
	The problem of extracting the motion in every frame of an image sequence is usually not solvable using the information present in the image sequence alone. For example, if a single-colored, textureless disc rotates, no movement can be observed. This problem also manifests itself if a rectangle is moving sideways. In that case, no motion can be seen if we only look at a small region around the upper edge of the rectangle (where we can not see the sides). The human visual system analyses the image sequence at different scales, and can therefore correctly estimate the movement at the upper edge of the rectangle. This is possible because it regards the upper edge as a part of the entire rectangle. The most wide-spread approach in traditional motion estimation, or optic flow, is based on the use of local derivatives of the image sequence to estimate movement. Since only the information of a small region region is used to estimate the flow at a certain position, the above mentioned problem is not solved. An existing method, which uses the motion of so-called toppoints in an image sequence, does not have to cope with this shortcoming, because toppoints exist at different scales in an image. Therefore their motion describes the movement of image structures with different scales, similar to the approach of the human visual system. We propose an optic flow estimation method that combines these two approaches into one, using the benefits of one to overcome the drawbacks of the other. Perturbation theory is used for the implementation and we perform an evaluation using test results from three image sequences. During this evaluation a problem in the motion estimation is encountered that leads to a great inaccuracy. Closer inspection reveals that this problem is caused by an erroneous relation between toppoint motion and the flow field, on which the existing toppoint method is based. Some alternatives to this erroneous relation are presented, which unfortunately do not appear to be the correct solution. It is shown, however, that if this problem can be overcome, our approach is expected to provide performance comparable to other state-of-the-art optic flow techniques.
Last modified: November 03, 2009	Status is Finished


Automatic Analysis of 2D MR Quantitative Flow Images
	Problem statement With cardiac-triggered magnetic resonance imaging it is possible to measure the velocity of the blood flow in the human body as a function of time in the cardiac cycle. The specific MR scanning method is often denoted as magnetic resonance quantitative flow imaging (MR Qflow). MR Qflow results in two types of image series, namely an anatomical and a phase-contrast series. The Philips ViewForum Clinical Workstation contains functionality for the semi-automatic quantitative analysis of MR Qflow image data. The analysis application has been optimized primarily for cross-sections through the larger blood vessels in the human body. These tubular-shaped vessels can be rather easily detected. In the current analysis application, the user has to manually delineate a contour around the blood in one of the phases. The contours in the images acquired at the other phases are then automatically derived from this initial manual contour by a mechanism called contour propagation. The current application has a number of drawbacks. The purpose of this master thesis project is to find/invent, implement and evaluate methods that eliminate these drawbacks. The end product should be a prototype software application that can efficiently, robustly and accurately quantify MR Qflow images and that can perform specific flow calculations and comparisons for a number of specific cardiovascular diseases. Goals The goals are: • find/invent methods for the fully automatic segmentation (contouring) of MR Qflow images • find/invent methods for the automatic quantitative analysis of MR Qflow images • If user interaction is needed during segmentation and/or quantification: design task guidance for this interaction, • implement and optimize the segmentation and quantification methods and their task guidance in a software prototype analysis application, • design/implement the means to perform flow comparison for a number of clinically relevant Qflow examinations, • create a gold standard for the evaluation of the prototype software application (both using synthetic images and user-drawn contours in realistic patient images), • evaluate the prototype using the gold standard and report the results. Tasks 1. Learn the programming environment (C, EasyScil) 2. Learn the specific properties of MR Qflow imaging data 3. Study earlier investigated/published segmentation and quantification methods (incl. task guidance) 4. Implement the most promising analysis method(s) into a software prototype 5. Optimize the implemented methods (choose best parameter values) 6. Create gold standard viability images: - Develop software to generate synthetic functional images (with known parameter values) - Manually create a gold standard on realistic patient viability images 7. Evaluate the prototype on gold standard image data (synthetic and realistic) 8. Write a report, give a final presentation Student profile 1. Master student Technical University Biomedical Engineering, Electrical Engineering, Applied Physics 2. Prior experience in software development (preferably C) 3. Prior knowledge on cardiovascular disease, cardiovascular imaging, image analysis 4. Analytical, systematic, independent, perseverance 5. Fluent English speaking and writing If the desired prior knowledge is not available, the student should be eager and able to acquire this knowledge efficiently at the start of the project. Application procedure - send your CV and a transcript of your achieved marks to Marcel Breeuwer (Marcel.Breeuwer@philips.com). - he will decide whether you will be invited for an interview/evaluation - if invited, you will give a 30-minute presentation about your work, experience and ambitions - you will get explanations/demos of their work, mostly concerning the possible master project of course - you will be interviewed by some members of Marcel Breeuwer's group - they will judge you on your CV, marks, presentation, interviews and the period during which you are available
Last modified: October 29, 2009	Status is Open


Automated segmentation of cardiovascular structures using 4D MRI flow measurements
	Background : Measurements from a wide variety of modalities are used to diagnose cardiovascular diseases. Magnetic Resonance Imaging (MRI) provides a significant share; especially for more complex cases. Current procedures in clinical practice are based on anatomic MRI scans, while more advanced MRI scanning protocols exist that provide quantitative blood flow information. Clinical research takes a large interest in finding correlations between blood flow and the progression of vascular diseases. Until recently computational fluid dynamics (CFD) simulations delivered the main source of information. Nowadays also electrocardiogram (ECG) triggered quantitative flow MRI data can be acquired; providing an actual measurement of the blood flow, as opposed to a simulation of the blood flow. The flow measurements provide a large data set, which is hard to interpret on a slice-by-slice basis. Mental reconstruction of vascular structures can be challenging, even for a skilled radiologist. The flow measurements add an unsteady flow field to these complex structures, resulting in a case which becomes exceptionally hard to read. Novel visualization techniques could support clinical research to obtain better understanding of the acquired data. Problem Statement: Visualization of anatomical structures very often requires a reliable segmentation of the data set, especially in a 3D environment. A segmentation, commonly represented by a labeled data set, provides valuable information of structures concealed within the data set. Anatomical structures can be depicted separately and landmarks can be derived automatically from the segmentation. The majority of the segmentation algorithms, such as region growing; level-sets; watersheds and multi-scale segmentation, perform well on good quality morphologic scans, for instance acquired by computed tomography (CT). MRI measured images are more challenging to segment, since the acquisition process implicates more noise and the contrast varies significantly between different acquisition sequences. In case of a 4D flow MRI acquired data, segmentation becomes even more challenging. For each phase of the cardiac cycle, anatomy images and flow (phase and magnitude) images are reconstructed. The acquisition of the data needs to be executed fast, in order to obtain a reliable representation of the blood flow throughout a heart beat. As a result of the fast acquisition, reconstructed anatomical images have very low contrast and a relatively low resolution. Together with the fact that the anatomical structures move over time, the images become extremely hard to segment. Goal: Next to the morphologic data, the blood flow field also provides substantial information of the morphological structures. The magnitude of the blood flow velocities gives a good indication of the location of the inner walls of the structures of interest. Therefore, we would like to investigate a novel segmentation algorithm of the cardiovascular structures, incorporating both blood flow information and morphologic data. Execution: Literature Study Investigate existing literature on segmentation. This includes both common techniques for scalar data, as well as flow based segmentation techniques. To the best of our knowledge, there are limited publications that take both velocity and intensity into account. For example, there is the work by Person et al. [1] and recently the work by Kainmuller et al. [2]. Design of an algorithm Based on the findings from the literature study, a new algorithm should be proposed to segment the 4D MRI flow. Implementation in C++ / VTK Implementation should be executed in the C++ language, with support of the visualization toolkit (VTK). Evaluation Small case study of several data sets, in order to evaluate the quality of the resulting segmentation. Contact: Roy van Pelt (r.f.p.v.pelt@tue.nl) or Anna Vilanova (a.vilanova@tue.nl) Working location can be either Eindhoven or Best References: [1] "Phase Contrast MRI Segmentation Using Velocity and Intensity", Markus Persson, Jan Erik Solem, Karin Markenroth, Jonas Svensson, and Anders Heyden in Proceedings of Scale Space (2005) [2] "Level set segmentation of the heart from 4D phase contrast MRI", Dagmar Kainmuller, Roland Unterhinninghofen, Sebastian Ley and Rüdiger Dillmann in Proceedings of the SPIE, Volume 6914, pp. 691414-691414-8 (2008)
Last modified: October 29, 2009	Status is Ongoing


Dose painting by numbers for lung tumors
	Introduction: In current radiotherapy lung (and other) tumors are irradiated to homogeneous dose levels. However, there is growing evidence that a recurrence of a lung tumor (after irradiation) is more likely to occur in tumor tissue that showed high FDG uptake in a PET scan (taken before irradiation). Or reversely, to increase tumor control, it seems beneficial to boost the dose to tumor voxels that having a high FDG uptake. Modifying the dose distribution in a patient based on the voxel values of a PET scan is called dose painting by numbers. Aim of the project: Facilitating the clinical introduction of dose painting for lung tumours by: 1. Developing a sensible relationship between the voxel intensities and required voxel dose based on biological models 2. Developing and implementing a new software tools for creating dose painted IMRT plans 3. Developing an adequate scoring system for ranking different treatment plans, how do we score a treatment plan optmized on both physical and biological objectives. 4. Comparing dose painted plans with conventional plans for 10 patients with an advanced staged lung tumor. Research collaboration: The project is part of a research collaboration between the Catharina Hospital Eindhoven (CZE), The Netherlands and Philips Medical Systems, Madison, USA. For more information please contact G. Meijer, PhD (gert.meijer@cze.nl). See: http://bmia.bmt.tue.nl/people/BRomeny/Projects/DosePainingByNumbers.pdf
Last modified: October 23, 2009	Status is Open


Non-invasive analysis of cardiac mechanics by means of ultrasound imaging
	The SinMod method, recently developed by Prof. Theo Arts, is a tracking method based on sine wave modelling for the extraction of myocardial motion from magnetic resonance tagging (MRIT) images of the heart (see figure, top). The measured displacement field is used for subsequent estimation of myocardial strain, rotation and torsion. SinMod is less sensitive to artifacts than the standard method of harmonic phase analysis (HARP). When applying SinMod on synthetically generated images, an interesting result is that it performs well not only in images with a grid pattern such as MRIT scans, but also in images with a speckle pattern. Therefore, the next step is to investigate its applicability in the field of cardiac ultrasound (see figure, bottom). In this context, we are in search of one or two master students, interested in image processing and/or cardiac mechanics, willing to cooperate with our group. The project can result in a master thesis for the student(s), but can also be intended as a 10-weeks internship. Students interested in a master thesis are preferred. Contacts and supervisors: Alessandro Rossi, MSc - alessandro.rossi@bf.unimaas.nl - tel. 043-3881662 Tammo Delhaas, MD, PhD - t.delhaas@fys.unimaas.nl - Tel. 043-3881078
Last modified: October 18, 2009	Status is Open


AutoMerge Initialization
	The neurosurgery department of the university hospital in Maastricht is working with the Polestar N20, Medtronic's Intraoperative MR scanner. To make pre-operative patient data (such as MRI scans, fMRI, DTI scans etc.) available during surgery in the correct configuration, our group has worked on an image registration algorithm to align pre-operative data with intra-operative Polestar scans. We would like to continue this work in a new master student project. The goals of this project are: I. Improving the performance of PoleStar Registration Initialization algorithm During the first phase, our research group did a good job in developing an automatic initialization algorithm for pre-OP MR to PoleStar registration. However, the performance is not yet within the commercially acceptable range. We would like the algorithm to be further improved in multiple areas: Currently the success rate is about 60%, we would desire a success rate of close to 90%. (note that newer PoleStar images have a better image quality compared to what you used for this project) Only T1W images were tested in the first phase, we would like the test to be extended to T2W images as well. Of lower priority but still of interest is optimization of the algorithm to run in real-time (1 min or less) II. Possibility of using a variation of the algorithm for SLAB data We would be interested to see if the same algorithm or a variation of it can initialize the SLAB images to be registered to full MR/CT data using AutoMerge. For more information please contact Bram Platel
Last modified: September 27, 2009	Status is Finished


Locating and Visualizing the Optic Radiation for Epilepsy Surgery
	Background Temporal lobe epilepsy (TLE) is the most frequent form of epilepsy. In about 70% of cases TLE is associated with neuronal cell loss in the hippocampus. Anterior temporal lobe resection (ATLR) is a well-established and effective means of treatment. However, this procedure commonly results in a partial loss of the patientís visual field, typically the upper quadrant on the opposite side. In 25 to 45% of cases the visual field loss can be severe enough to prevent the patient from driving a car. Visual field loss is mainly caused by damage to the fibers in the frontal part of the optic radiation, also known as Meyer's loop. Damage is very likely for two reasons: (1) the fibers are close to or even inside the resection area and (2) on conventional MRI the optic radiation is hardly distinguishable from the surrounding white matter. Furthermore, there exists a large variation in the anatomy and extent of Meyer's loop between different patients. Project description The first step in reducing the number of visual field deficits in patients undergoing temporal lobe resection, is to visualize the optic radiation, including Meyer's loop. This will improve the surgeonís ability to choose an angle of approach or resection boundary that minimizes the risk of damage to the fibers. The main goal of this project: to reliably reconstruct and visualize the fibers of the optic radiation using Diffusion Tensor Imaging and fiber tracking. Tasks Do literature study. Investigate which fiber tracking tools exist and which one is most suitable. Investigate (combinations of) seed regions that result in a reproducible and complete reconstruction of the optic radiation, including Meyer's loop. Investigate different fiber tracking algorithms (deterministic, probabilistic) and choose the algorithm that produces the most reliable results. Investigate the sensitivity of the fiber tracking algorithm for user input parameters and how these affect the output result. Investigate what data preprocessing steps are needed to optimize the fiber tracking process (noise reduction, distortion correction, etc.). Investigate what other imaging modalities could be useful to supplement the fiber tracking and visualization process (ocnventional MRI, f-MRI, etc.). Perform small user study with group of volunteers to compare fiber tracking results of the optic radiation. (Optional) Investigate how uncertainties in the DTI processing and visualization pipeline (due to noise, low resolution, partial volume effects, etc) affect the output results. Contact and work location Ralph Brecheisen (r.brecheisen@tue.nl) or Bram Platel (b.platel@tue.nl). Place of work can be either Eindhoven or Maastricht.
Last modified: September 27, 2009	Status is Ongoing


Deep Brain Stimulation for Parkinsons’s Disease: Assessing STN Connectivity using Functional and Diffusion-weighted MRI of Healthy Volunteers
	Background Parkinson’s patients have motor symptoms, for example tremors. Long-term drug treatment can result in even more severe motor side-effects, such as dyskinesia. An alternative therapy is Deep Brain Stimulation (DBS); the implantation of electrodes in the subthalamic nucleus (STN), an area that displays burst activity in Parkinson’s patients. The electrical stimulation results in significant improvements of the patient’s locomotion. However, in about half of cases, changes also occur in the behavior or intellectual capabilities of the patient, causing side-effects such as depression or mania. The three functions of the STN (motor, cognitive, emotional) are thought to be located in three separate areas. It is assumed that the side-effects will be less if the electrode is inserted more accurately, e.g. only in the motor part of the STN. Problem statement To achieve this, it is important to localize the STN and its motor part more precisely. Data acquired by anatomical, diffusion-weighted (visualizing the nerve tracts) and functional MRI (visualizing the activity), may facilitate this. In 2008, we have acquired MRI data of 12 healthy volunteers. We have performed this volunteer study to avoid burdening Parkinson’s patients and to acquire data that are suitable for scientific research. In a previous master project, the anatomical and diffusion-weighted MRI data were co-registered to the Talairach brain atlas using existing software. The connections of the STN were investigated using DTI fiber tracking. Goal of the master project Proceed with the results of the previous master project and try to segment the different functional parts of the STN based on the diffusion-weighted and functional MRI data. Various clustering methods can be used to perform this segmentation, using different types of input information: i.e. MRI data within the STN itself, fiber tracking results (anatomical connectivity information), correlation in fMRI activity (functional connectivity information). This will probably be too much work for a master student, so work will be shared with Ellen Brunenberg (involved PhD student) according to preference. Contact information The project will be performed in the Image-guided Surgery part of the Biomedical Image Analysis group, supervised by Ellen Brunenberg (PhD student) and Bram Platel (assistant professor). We will collaborate closely with the Radiology department in Maastricht (Walter Backes/Maarten Vaessen). Working location can be either Eindhoven or Maastricht. Please contact Ellen Brunenberg (e.j.l.brunenberg@tue.nl) or Bram Platel (b.platel@tue.nl) for more information. More project details (proposed approach, student profile, and references) can be found in the long description pdf, that can be accessed by clicking on "details" below.
Last modified: September 25, 2009	Status is Open


Diffusion Tensor Imaging study of neonates at risk for neurodevelopmental disorders
	Severe premature birth may lead to a disturbed cognition. Approximately 15-25% of preterm children with a gestational age below 30 weeks or a birth weight bellow 1000 grams have an abnormal cognitive development. Development support at an early age can help to improve cognitive development. However, nowadays it is not possible to distinguish neonates that need this development support from neonates that do not need it. The goal of this project is to investigate new tools for the analysis of Diffusion Tensor Imaging (DTI) data. In order to be able to answer the question whether DTI observed abnormalities in the brain white matter structure after premature birth can predict abnormalities in cognitive development. Picture: White matter tracks of a full-term newborn with normal MRI at day 6 obtained using diffusion tensor imaging [Pul et al. 2006]
Last modified: September 23, 2009	Status is Finished


High-temporal resolution dynamic contrast-enhanced breast MRI
	Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is crucial in the work up of findings in the breast that cannot be assessed from X-ray or ultrasound images, and in the screening of the high-risk population. The high-risk population consists of women that have a personal or familial history of breast cancer, or that are known to be carriers of certain gene mutations (for instance BRCA1 and BRCA2). DCE-MRI is a very sensitive technique (most tumors are detected), but the specificity is low (too many lesions are suspect of malignancy, resulting in unnecessary biopsies). Computer-aided diagnosis could be a powerful tool to improve the specificity of DCE-MRI of the breast. During a DCE-MRI breast exam, a low-molecular-weight contrast agent, commonly gadolinium-based, is injected. The contrast agent causes enhancement in the tissue by shortening the longitudinal relaxation time (T1). The uptake of contrast agent is therefore measured with a series of T1-weighted images over time. Due to the process of tumor angiogenesis, contrast-agent uptake is often more rapid and higher in tumor tissue compared to normal tissue. The DCE-MR image series serves to assess the tissue on a morphological basis (spatial patterns) and kinetic basis (temporal/uptake patterns); the latter being the topic of this master project. Kinetic analysis, as it is currently performed in clinical practice, is based on descriptive curve-shape analysis. This means that curve properties like time-to-peak and maximum enhancement are calculated. Newer analysis methods are however aiming to describe curve-shapes in terms of the physiological processes underlying the observed uptake of contrast agent. Physiology-based modeling is gaining attention, but it still has to prove its added value in the field of clinical diagnostics. One of the issues in the field of clinical diagnostics is that commonly the temporal resolution of the T1-weighted dynamic image series is low; whereas physiology-based modeling requires data with high temporal resolution. Due to this mismatch, it is hard to assess the usefulness of physiology-based modeling. However, without demonstrating the possible advantages of acquiring data at higher temporal resolution, the clinical practice will not change. The purpose of this project is therefore to explore the possible impact of physiology-based modeling on diagnostics: does physiology-based kinetic modeling of high-temporal resolution data lead to useful kinetic features? To enable this research, several data sets (~ 20) were acquired at high temporal resolution.
Last modified: September 04, 2009	Status is Finished


Master project at Maastro Physics, Maastricht
	MAASTRO is the Maastricht Radiation Oncology Centre in Maastricht. The MAASTRO PHYSICS group is also occupied with the education of Master students Medical Engineering of the Department of Biomedical Engineering of the Eindhoven University of Technology. The education is embedded within MAASTRO SCHOOL. See MAASTRO SCHOOL (link below) for further information. If you are interested in doing a research project at MAASTRO PHYSICS please contact Geert Bosmans (geert.bosmans@maastro.nl).
Last modified: August 24, 2009	Status is Open


Fully Automatic Blood Vessel Branch Labeling
	Volume representations of blood vessels acquired by 3D rotational angiography are very suitable for diagnosing a stenosis or an aneurysm. For optimal treatment, the shape parameters of the diseased vessel parts are needed by physicians. Therefore, a fully-automatic extraction of this shape from such a volume representation has been developed. The demo program v3d_main has been developed to test the various algorithms, such as the segmentation algorithm, the wave propagation algorithm and the thinning algorithm. This paper first discusses and analyses the blood branch labeling acceleration algorithms, and proposes two methods for improvement. The first one is called the surface wave propagation method which restricts the wave moving only along the blood vessel surface. This method is applied to detect the extremities of the vessel voxel structures. The second one combines a thinning algorithm with the surface propagation to extract the center lines and the bifurcations of the blood vessels, which also gives the vessel voxels a unique number (label) per vessel branch. Proper validation results are given in this report. The result shows that the two methods can substantially decrease the computation time and keep the labeling accuracy. However, whether the branch labeling result, generated by surface wave propagation based on a vessel graph, are suitable for computer assisted diagnosis has not been investigated yet.
Last modified: August 05, 2009	Status is Finished


Hippocampus Region Segmentation for Alzheimer’s Disease Detection
	This thesis provides an automatic method for the segmentation of the hippocampus region and the detection of Alzheimer’s disease based on the size of the regions. The hippocampus is known to shrink in time due to cell death, and it is linked with increased memory loss, which is the primary symptom of Alzheimer’s disease. First, the hippocampus region has been analyzed by a region-of-interest mask defined for the coronal views and was determined according to the Medial Temporal lobe Atrophy (MTA). To compare this template-based method, we do also perform the segmentation of the hippocampus based on several low level segmentation methods, such as, region growing, edge-based segmentation and active contours. However, none of these methods achieves the desired result because of the complex appearing of regions surrounding the hippocampus and the low image contrast in the hippocampus region. As a result, we propose to use an active shape model to represent the shape of hippocampus based on eigenvectors extracted from training shapes. This model allows for more freedom of shape changes, such as, the affine transformation. Mutual information is used to evaluate the segmentation label. After full segmentation of the hippocampus, more information, such as, pixels of CSF surrounding the hippocampus could be extracted using local region growing. With respect to sensitivity, specificity and ROC curve, features extracted from the segmentation of the hippocampus region lead to the best result. Moreover besides features extracted from baseline data, CSF changes within a period (roughly one year) is also helpful for computer diagnosis.
Last modified: August 05, 2009	Status is Finished


Deep Brain Stimulation for Parkinson's Disease: Finding the STN on anatomical, functional, and diffusion-weighted MRI of healthy volunteers
	Deep Brain Stimulation is the implantation of electrodes in the subthalamic nucleus (STN), a brain area that displays strong changes in Parkinson’s Disease patients. Stimulation results in significant improvements of the patient’s locomotion. However, in half of the cases, changes also occur in the behavior or intellectual capabilities of the operated patient, such as a depression or mania. It is assumed that the side-effects will be less if you insert the electrode in a more accurate way, meaning only in the motor part of the STN. To achieve this goal, it is important to localize the STN and its motor part. We think that data acquired by anatomical MRI, diffusion-weighted MRI and functional MRI, can help to visualize the STN and its motor part. In September and October 2008, we have acquired anatomical, diffusion-weighted and functional data of 12 healthy volunteers. The goal of this project is to find features in anatomical, diffusion-weighted and functional MRI images that can be combined to find the subthalamic nucleus.
Last modified: July 21, 2009	Status is Finished


Enhancement of HARDI-images for Localization of Deep Brain Stimulation Targets
	Deep brain stimulation (DBS) of specific brain regions has provided remarkable therapeutic benefits for otherwise treatment-resistant neurological disorders. To date more than 40,000 people worldwide have been treated with the deep brain stimulation therapy for essential tremor and advanced Parkinson’s disease. For many of these patients DBS results in a long-term improvement in motor function and dyskinesia, however, a significant number of patients (>55%) suffers from side effects after DBS. These side effects range from cognitive problems to depression and mania. Existing planning methods for DBS surgery are not able to accurately locate the exact position of the stimulation area. Incorrect targeting can lead to stimulation of other brain areas, causing the mentioned side effects. Therefore we consider exact localization of the stimulation targets for DBS based on a new clinical imaging modality, High Angular Resolution Diffusion Tensor Imaging (HARDI). HARDI provides a detailed probability profile of the diffusion of water molecules for each voxel, i.e a HARDI-image is a probability density on the joined domain of 3D-positions and orientations. HARDI can resolve complex structures such as nerve fiber crossings and bifurcations that are abundant around the stimulation targets for DBS. So far usefulness of HARDI in a clinical setting has been limited due to the low signal-to-noise ratio in these images when acquired within a clinical scan time (<15min). Therefore we exploit our recently introduced concept of “orientation scores”, its underlying group structure to enhance the HARDI data such that fiber junctions are maintained, while the high frequency noise in the joined domain of positions and orientations will be reduced. Direct goals within this master project are development and implementation of left-invariant diffusions on HARDI-images (which naturally deal with crossing fibers) within Mathematica (or C++). Here we distinguish between two approaches: 1. kernel operators algorithms (SE(3)-convolutions) 2. finite difference scheme algorithms The first approach is well-suited for parallelization, whereas the second approach is well-suited for non-linear adaptive diffusions. For a detailed description (inc. literature suggestions) see here: http://bmia.bmt.tue.nl/people/BRomeny/Projects/HARDI2009.html Profile: Good/Excellent BME-student (or EE-student) with good programming skills and a high interest in the application of more advanced Mathematics. Graduate Professor: Bart ter Haar Romeny, B.M.terhaarRomeny@tue.nl +31-402475537 Daily Supervisors: Remco Duits, R.Duits@tue.nl, +31-402472859 or +31-402473037, WHoog 2.110, HG 8.93 and Bram Platel, B.Platel@tue.nl, +31-402475514 Whoog 2.101
Last modified: July 03, 2009	Status is Open


Quantification of collagen fibers based on orientation scores: a real-time hardware accelerated approach
	This Master Thesis project, performed at the Biomedical Image Analysis group (Eindhoven University of Technology, department of Biomedical Engineering), is focused on the practical application of orientation scores using dedicated hardware resources. We will show the potential of orientation scores in real time quantification of collagen fibers in 2D slices taken from the fibrous tissues around the bone under strong hardware restrictions. The quantification is needed for mechanical studies where some of the relevant features are the orientation of the collagen fibers, the coherence of the fibrous structure and the curliness of the fibers. With coherence is meant in this context the presence of local organized patterns in the tissue like groups of aligned fibers, the pattern that we find in the periosteum around the bone, or the orthogonal organizations that appear on cardiac tissues.
Last modified: July 01, 2009	Status is Finished


GPU-based glyph rendering for HARDI
	Diffusion Tensor Imaging (DTI) is relatively recent MR imaging modality, used for describing the underlying structure of tissues such as white matter of the brain, muscle and bone marrow. Each voxel of the dataset is assigned a second-order tensor, used to describe the local water diffusion. Despite its ability for visualizing and characterizing white matter connections, DTI has some important limitations – it can only determine a single fiber orientation at any given location in the brain. This is clearly inadequate in regions (voxels) with complex white matter architecture, where different axonal pathways cross each other. In order to better describe the complexity of water motion in a voxel, a novel approach, High Angular Resolution Diffusion Imaging (HARDI), has been proposed by Tuch et al. [1]. At the cost of longer acquisition times, the sphere is sampled in N (far more than used in DTI, typically: 70-200) discrete gradient directions, and the apparent diffusion coefficient (ADC) profile along each direction is computed [2]. Hence, at each voxel, we have a discrete spherical function with *no a priori* assumption about the nature of the diffusion process within the voxel. The visualization of such a large amount of data is a challenging problem. One approach is to show the ADC or probability density function (PDFs) for all the voxels in a 2D region of interest (ROI) that is defined by the user. Current visualizations use glyphs or iconic representations of the ADCs/PDFs by mapping the orientations on a sphere and then deforming the sphere according to the ADCs/PDFs. Thus, the sphere will deform such that orientations with large ADCs/PDFs will show up as spikes on the surface of the sphere. There are several problems with this approach. First, a geometry and topology of points must be created that must be deformed. This is a complicated and slow process. Also, the rendering process for many glyphs is slow, and without making this even slower it is not possible to show smooth glyphs (with large amount of vertices on the surface). In order to improve the visual quality and the rendering speed of the glyphs, we propose to apply GPU-based rendering. For this, the HARDI data must be loaded in the GPU memory. Then, for each of the points that define the positions of the glyphs, a glyph must be rendered. This can be done, for example, by using geometry and fragment shaders. The geometry shader can create a bounding box around the to-be-created glyph and as a result, all pixels on the screen where the glyph would be projected will be rendered by a fragment shader. This fragment shader can then apply a ray-casting approach to determine, for each pixel, the intersection of the view ray and the glyph and using that, compute the colour of the pixel by applying lighting calculations. The student will need to evaluate the feasibility of the proposed method and can also develop a different method for the rendering of the glyphs. Tasks: Study HARDI and GPU programming using GLSL. Implement a method for rendering HARDI glyphs on GPU. References:** D.S. Tuch, T.G. Reese, M.R. Wiegell, N. Makris, J.W. Belliveau, and V.J. Wedeen. High angular resolution diffusion imaging reveals intravoxel white matter fiber heterogeneity. Magnetic Resonance in Medicine, 48(4):577–582, October 2002. D.S. Tuch. Diffusion MRI of complex tissue structure. PhD thesis, Division of Health Sciences and Technology, Massachusetts Institute of Technology, 2002.
Last modified: June 18, 2009	Status is Finished


Cerebrale connectiviteit in epilepsie
	Project beschrijving afstudeerstage BMT (BME of ME) Universitair Medisch Centrum Maastricht (MUMC+) Afdeling Radiologie Achtergrond. Patiënten met chronische epilepsie ontwikkelen vaak cognitieve problemen die soms erger zijn dan de aanvallen zelf. Deze problemen variëren van geheugen problemen en mentale traagheid tot globale cognitieve achteruitgang. Pogingen om deze problemen te verklaren aan de hand van epilepsie gerelateerde factoren (aantal aanvallen etc.) bieden tot nu toe geen uitsluitsel. Nieuwe ontwikkelingen op het gebied van Magnetic Resonance Imaging (MRI) geven ons de mogelijkheid om een neuronaal substraat voor cognitieve problemen te traceren en eventuele weefselschade te diagnosticeren. Projectomschrijving. In dit project proberen wij een relatie aan te tonen tussen cognitief presteren (en achteruitgang) en functionele netwerken in het menselijke brein. Functionele MRI (fMRI) geeft een indirecte maat voor neurale activiteit in het brein. Een functionele connectie tussen twee hersengebieden bestaat wanneer deze in de tijd gecorreleerde activiteit vertonen. Aangezien veel hogere cognitieve processen netwerk functies zijn (d.w.z. essentiele betrokkenheid van een aantal hersengebieden) is het interessant om het probleem van cognitieve achteruitgang aan te pakken vanuit een netwerk perspectief. Het specifieke doel van dit project is om het topologische patroon van functionele inter-connecties te kwantificeren. Dit is mogelijk door gebruik te maken van het zogenaamde “small-world” model. Hierbij wordt een netwerk zoals het brein samengevat in een aantal te berekenen parameters die informatie geven over de structuur van het netwerk. Het project is onderdeel van een intensieve en lang bestaande samenwerking met het Epilepsie Instituut Kempenhaeghe (prof.dr. A. Aldenkamp) te Heeze. Taken. Aan de hand van bestaande data sets zal eerst uitgezocht worden hoe de cerebrale connectiviteit verschilt tussen patiënten met epilepsie en gezonde vrijwilligers. In de loop van het project zal de afstudeerder gaan participeren in nieuwe MRI beeldacquisitie, nadat een intern MRI rijbewijs is behaald. De belangrijkste taken zijn literatuuronderzoek, uitvoeren van MRI scans en data analyse/verwerking. De student dient over voldoende kennis van Matlab te beschikken en inzicht in computationeel intensieve data analyses (zie bijgevoegde fig. over DTI connectiviteit). Goede contactuele eigenschappen is een voorwaarde in een zorginstelling. Contact. De student zal worden begeleid door drs. Maarten Vaessen (promovendus Radiologie/MUMC+) en dr. ir. Walter Backes (Klinisch Fysicus Radiologie/MUMC+ en eindverantwoordelijke voor de afstudeeropdracht). Neem voor meer informatie contact op met Maarten Vaessen (mjvaessen@yahoo.com, 043-3876984) of Walter Backes (w.backes@mumc.nl of 043-3876948).
Last modified: June 08, 2009	Status is Open


3DRA Segmentation & Measurements
	An important imaging modality within Philips Medical System’s X-ray product line is 3D Rotational Angiography (3DRA). In this acquisition mode, an X-ray detector is rotated around a patient, while acquiring a few hundred images of e.g. the vessels in the patients head, see Figure. This makes it possible to visualize geometrically complicated pathologies like cerebral aneurysms. Treatment of such diseases, e.g. by stenting or coiling, an aneurysm is thus supported by more accurate information about the location, the shape and the dimensions of the diseased area. The size of the aneurysm is measured in order to determine the number of coils, or the size of the stent to be used in the subsequent treatment. Other applications of 3DRA in which such questions appear to be useful are: aneurysm labeling and virtual stenting. A problem of today’s threshold-based measurement methods is that the result is quite dependent on the parameters of the underlying algorithm, and an often heard question then not surprisingly is: “How accurate is this measurement?” *A long description can be found in a pdf* by clicking in details**
Last modified: May 20, 2009	Status is Open


Deep Brain Stimulation for Parkinson's Disease: Establishing the STN neuroanatomy using diffusion-weighted MRI of rat brains
	Deep Brain Stimulation is the implantation of electrodes in the subthalamic nucleus (STN), a brain area that displays strong changes in Parkinson’s Disease patients. Stimulation results in significant improvements of the patient’s locomotion. However, in half of the cases, changes also occur in the behavior or intellectual capabilities of the operated patient, such as a depression or mania. It is assumed that the side-effects will be less if you insert the electrode in a more accurate way, meaning only in the motor part of the STN. To achieve this goal, it is important to localize the STN and its motor part. We think that data acquired by diffusion-weighted MRI can help to visualize the STN and its motor part. We have acquired anatomical and diffusion-weighted data of post mortem rat brains. The goal of this project is to examine the STN and its environment using clustering algorithms.
Last modified: March 12, 2009	Status is Ongoing


Illustrative Rendering for Brain Diffusion Tensor Imaging Data
	Diffusion Tensor Imaging is a Magnetic Resonance (MR) technique that allows the measurement of water diffusion in tissue. This technique allows reconstructing the complex white matter structure of the brain. If the fiber structure of the whole brain is visualized, it is very difficult to obtain useful information due to the clutter (see Figure 1). At the BMIA group at the TUe, methods have been developed to help the visualization of this data (i.e., DTITool). Clustering and segmentation methods for the visualization of DTI brain data have been (and are being) researched [1][2][3] at BMIA. However, there are no visualization techniques that allow an easy and insightful visualization of the results; and that allow the combination of several levels of detail in one image. It is known that illustration techniques can help in visualizing clutter information. In the BMIA group, there has also been the development of GPU based illustrative techniques for volume rendering [4] (see Figure 2). The goal of this project is to extend these techniques for the purpose of DTI visualization of the brain white matter. Illustrative techniques are also relevant for showing several levels of detail at the same time avoiding cluttering and helping in showing the relation between each level. It is assumed that the levels of detail are given. The challenge is the visualization of those. It should also be considered the combination of anatomical data ( i.e., T1 weighted MR) with DTI data. User interaction and interactivity of the visualization is of great importance for the project. Therefore, GPU based techniques will need to be exploited. [1] http://bmia.bmt.tue.nl/Research/MVIAV/DTI [2] P.R. Rodrigues, A. Vilanova, T. Twellmann, and B.M. ter Haar Romeny. Adaptive distance learning scheme for diffusion tensor imaging using kernel target alignment . In MICCAI Workshop on Computational Diffusion MRI (CDMRI), pages 148-158, 2008. [3] B. Moberts, A. Vilanova, and J.J. van Wijk. Evaluation of fiber clustering methods for diffusion tensor imaging . In IEEE Visualization Conference Proceedings, pages 65-72, Oct 2005. [4] http://bmia.bmt.tue.nl/Research/MVIAV/IVR
Last modified: February 11, 2009	Status is Ongoing


Implementation and Assessment of Different Geodesic Connectivity Measurements for Diffusion Tensor Images
	The aim of this project is to implement a white matter tractography method based on anisotropic wavefront propagation in diffusion tensor images. The next task is to assess and analyze the results of different connectivity measurements in order to further develop the DTItool. Once we have selected a measurement that can give more coherent results, we intend to apply the methodology described above to fiber tracking in rat or human brain diffusion tensor images. We are looking for ways to improve deep brain stimulation procedures. Our main interest is a more specific targeting of the motor part of the subthalamic nucleus, stimulation of which enhances motor function in Parkinson’s Disease patients. We assume that an investigation of the connectivity of the subhalamic nucleus with other brain structures like the motor cortex, globus pallidus, and substantia nigra can help to achieve this goal. Because the tracts of interest are smaller and probably more entangled than major bundles like the corpus callosum and the pyramidal tracts, we need sophisticated tractography techniques.
Last modified: January 23, 2009	Status is Finished


Semi-automatic muscle segmentation using Diffusion Tensor Imaging and T1-Weighted datasets of the human lower arm
	Predicting functional human movement becomes more and more important. Often, minimal invasive surgery techniques are used to maintain as much functionality of the operated tissues as possible.The patient-specific questions and the anticipated options or strategies during surgery may be simulated on beforehand with Computer Assisted Surgical Planning (CASP) software. Only when this CASP-software is based on functional, patient-specific musculoskeletal models will it be possible to predict the functional outcome of surgery. DTI (diffusion tensor imaging) has proven to be able to determine muscle geometry. The aim of this project is to extend the DTI-Tool (developed at the BMIA group) with a fiber segmentation algorithm based on a minimal distance of a fiber end (DTI) to a tendon plate (T1-W)
Last modified: January 23, 2009	Status is Finished


Intramuscular lipid storage in health and disease: 3D confocal laser scanning microscopy
	Numerous studies have reported a strong correlation between intramuscular lipid content and insulin resistance. However, the proposed relationship between intramuscular lipid accumulation and skeletal muscle insulin resistance is not unambiguous, as trained athletes have been shown to be markedly insulin sensitive despite elevated lipid storage in muscle tissue. Divergent metabolic events are responsible for the greater muscle lipid content in the endurance-trained versus insulin-resistant state. The greater lipid storage in the trained athlete represents an adaptive response to endurance training, allowing a greater contribution of the muscle lipid pool as a substrate source during exercise. In contrast, elevated lipid stores in the obese and/or type 2 diabetes patient seem to be secondary to a structural imbalance between plasma free fatty acid availability, lipid storage and oxidation. It seems evident to assume that there are many structural differences in the characteristics of the lipid droplets present inside skeletal muscle fibers of trained endurance athletes and sedentary, obese type 2 diabetes patients. By using 3D confocal laser scanning microscopy on human skeletal muscle tissue, this project will aim to generate more insight in the (patho)physiological characteristics and volume of the many lipid droplets present in skeletal muscle fibers. This project is a continuation of the MSc thesis of Edwin ter Voert. A software package is available to segment the droplets, for 3D visualization and for a database for the properties of the droplets. This project will focus on the analysis of the physiological properties.
Last modified: November 16, 2008	Status is Open


Left atrium segmentation for electrophysiology III
	Atrial Fibrillation (AF) is a disorder characterized by quivering atria, beating inefficiently and preventing them to be completely emptied. AF puts people at risk of blood clots and stroke. By means of ablation, i.e. surgically removing faulty electrical path ways in the heart by applying heat to the conductive tissue, patients with AF can be helped. During ablation interventions, image guidance is mostly obtained from fluoroscopy. Navigation by fluoroscopy is difficult, and therefore alternative methods for guidance, e.g., by means of augmented reality will be investigated. Such augmented reality requires presegmentation of, e.g., the Left Atrium (LA) and subsequent mapping of the segmentation result onto the live fluoroscopy imagery during the intervention. First results have been obtained with a model-guided technique, first segmenting the endocardium followed by the epicardium. However, many subjects with respect to this research are still open.
Last modified: November 03, 2008	Status is Open


Acquisition of phantom cardiac left atrium images using CT
	Atrial Fibrillation (AF) is a disorder characterized by quivering atria, beating inefficiently and preventing them to be completely emptied. AF puts people at risk of blood clots and stroke. By means of ablation, i.e. surgically removing faulty electrical path ways in the heart by applying heat to the conductive tissue, patients with AF can be helped. During ablation interventions, image guidance is mostly obtained from fluoroscopy. Navigation by fluoroscopy is difficult, and therefore alternative methods for guidance, e.g., by means of augmented reality will be investigated. Such augmented reality requires presegmentation of, e.g., the Left Atrium (LA) and subsequent mapping of the segmentation result onto the live fluoroscopy imagery during the intervention. The segmentation method will have to be tested in a controlled environment. To improve segmentation results in a difficult segmentation task (which this project is about) one can directly work on the segmentation algorithm or try to improve the input (i.e. the images) first. This project aims at acquisition of ex-vivo pig hearts with either a CT scanner or rotational angiography. These images should have optimal quality for the task at hand: left atrium segmentation. The project is performed in close collaboration with Philips Healthcare. At Philips Healthcare, a lot of expertise concerning roentgen image acquisition is available. This project will only start after the guidelines have been set up in the internship preceding this project. Preferrably, the same student performs both the internship and the final master project. Probably, the best candidate is a Medical Engineering student, but the project is not strictly limited to the ME programme.
Last modified: November 03, 2008	Status is Open


Left atrium segmentation for electrophysiology II
	Atrial Fibrillation (AF) is a disorder characterized by quivering atria, beating inefficiently and preventing them to be completely emptied. AF puts people at risk of blood clots and stroke. By means of ablation, i.e. surgically removing faulty electrical path ways in the heart by applying heat to the conductive tissue, patients with AF can be helped. During ablation interventions, image guidance is mostly obtained from fluoroscopy. Navigation by fluoroscopy is difficult, and therefore alternative methods for guidance, e.g., by means of augmented reality will be investigated. Such augmented reality requires presegmentation of, e.g., the Left Atrium (LA) and subsequent mapping of the segmentation result onto the live fluoroscopy imagery during the intervention.
Last modified: November 03, 2008	Status is Ongoing


Quality Assessment of Cardiac Tracking Using Optical Flow
	Optical flow is a method widely used for the analysis of moving objects in image scenes. In our group, we apply multiscale optical flow to track the motion of the cardiac Left Ventricle (LV). Knowledge of the detailed motion of the LV can provide insight in the condition of the heart, and some aspects of the motion (e.g. the rotation) are early indicators of pathology. It is very important to track the LV very accurately, and so it is important to improve the analysis quality such that "what you get, is what it is". One of the aspects that is important for the quality of the motion "solution" in our method is proper scale selection. This means that the scale at which the images are analysed has to be selected properly. Currently, scale selection is performed by assessment of the condition number of the system of equations that has to be solved for every pixel, in order to obtain a motion estimation. This means that we go for the most stable solution, but that does not mean that we get the most accurate solution. This project is about investigation of the relation between condition number and the quality of the solution. A regularisation of the selected condition number (read: selected combination of scales) is the goal of this project. A second topic is to investigate an alternative method of motion estimation, called DENSE. DENSE is a scanning protocol on the MR scanners, but we try to mimic DENSE "in-silico".
Last modified: October 21, 2008	Status is Ongoing


Left atrium segmentation for electrophysiology
	Atrial Fibrillation (AF) is a disorder characterized by quivering atria, beating inefficiently and preventing them to be completely emptied. AF puts people at risk of blood clots and stroke. By means of ablation, i.e. surgically removing faulty electrical path ways in the heart by applying heat to the conductive tissue, patients with AF can be helped. During ablation interventions, image guidance is mostly obtained from fluoroscopy. Navigation by fluoroscopy is difficult, and therefore alternative methods for guidance, e.g., by means of augmented reality will be investigated. Such augmented reality requires presegmentation of, e.g., the Left Atrium (LA) and subsequent mapping of the segmentation result onto the live fluoroscopy imagery during the intervention.
Last modified: October 21, 2008	Status is Finished


Development and validation of a diagnostic tool to grade visual impairment - methodology and usability
	Cerebral Visual Impairment (CVI) is a neurological disorder caused by brain damage to the optical path located posterior to the optic chiasm. Due to this brain damage patients have deficits in aspects of visual perception and processing of visual information. Depending on the affected cortical areas this may lead to poor object and face recognition, reduced perception of depth and orientation problems. At present, the diagnosis of CVI is based on neurological and psychological examinations and observations. These tests are time consuming, subjective and require cooperation of an impaired child. Because of this, a new objective diagnostic tool is needed. At the department of Neuroscience, Erasmus MC, Rotterdam, a new diagnostic tool is proposed. This new tool is based on the fact that eye and/or head responses are automatically induced towards visual feature when it is detected, so called preferential looking. A range of visual stimuli were displayed on a monitor with an automated eye-tracking system for objective gaze measurements. Each trail contains 30 attractors to attrack the attention of the subjects, 4 motion and form coherence sequences to measure the global processing of the dorsal and ventral streams, 16 competitive and non competitive dots to exam an inhibitory effect of competitive tasks, 4 dot saccades to measure visual perception and latencies and 4 smooth pursuits to test the smooth pursuit system. We tested 120 children (0-10 years) without and 13 children (2-4 years) with cerebral pathology. Based on the visual stimuli applied to both groups, 2 new diagnostic markers were defined. Latency was defined as the time between the onset of a visual stimulus projection, e.g. attractor, competive dot, and the first reflexive eye movement towards the quadrant in which this stimulus was displayed. In addition, gaze fixation was tested as diagnostic marker. Gaze fixation area was equal to a fitted ellipse area containing the gaze data in the correct quadrant. Impaired children showed significantly delayed latency time, larger fixation area and shorter looking time to the form coherence compared to healthy children (p Automated preferential looking can be applied in both healthy and disabled children. Latency and fixation area can be promising diagnostic markers to asses function loss in children with CVI.
Last modified: October 20, 2008	Status is Finished


Contour tracking in depth estimation for 3D displays
	Problem Description In the past we have witnessed the transition from black and white TV towards color TV; the next step is the transition towards 3D-TV. Nowadays more and more 3D content becomes available in a format that contains two viewpoints and is suited for watching with so called stereo glasses. Within Philips, multiview displays have been developed that can offer a real 3D viewing experience without the use of such special glasses. See the YouTube movie: http://nl.youtube.com/watch?v=HF-PMN3aK8g. The image format proposed for these displays are video sequences with 3D content; i.e. pixels in the video stream do not only contain color information, but depth information as well. The goal of this project is to develop algorithms for conversion of 2D video into 3D. The motion in a video sequence contains a lot of information about the different objects and their relative depths. We will therefore use the optical flow in the sequence to segment objects and assign accurate depths throughout the sequence. The segmentation will be based on a level sets approach for computing the object contours. Time and deliverables The planned duration of the project is nine months. Deliverables and activities are listed in the following table. Due date Activity/deliverable Week 2 Project description Month 5 Intermediate presentation of preliminary results Month 7 Software of working prototype Month 9 Thesis and final presentation of end result About Philips Research Founded in Eindhoven, The Netherlands, in 1914, Philips Research as part of Royal Philips Electronics has expanded the scale and scope of its activities to become one of the world's major private research organizations. With laboratories in different countries (The Netherlands, England, Germany, China and the United States) and staffed by around 2,500 people, our common vision is to create technologies that will lead to products for improving people's lives. Application Students with a major in computer science, biomedical engineering or related area in the final stage of master level studies are invited to apply. Some basic knowledge in the area of digital image synthesis (computer graphics) is a prerequisite. Starting date: to be agreed mutually. For more information about the project and applications, please contact: Dr Patrick Vandewalle Philips Research High Tech Campus 36 5656 AE Eindhoven Netherlands e-mail: patrick.vandewalle@philips.com
Last modified: August 15, 2008	Status is Open


IMRT Pre-treatment Verification, Quality vs. Time Consumption
	A recent technological advancement in radiotherapy that providesthe possibility to increase tumor control and reduce side effects simultaneously is Intensity Modulated Radiation Therapy (IMRT).IMRT not only uses different beam positions and shapes, but alsovaries the field intensity across the beam (a beam then consistsof segments), in that way the dose is conformed to the tumor. The possibilities for delivery errors increase, because the therapybecomes more complicated. This makes pre-treatmentverification, to ensure that the delivered dose agrees with thecalculated dose, very important. A quick, accurate and reliablemethod for this pre-treatment verification will be developed.
Last modified: March 28, 2008	Status is Finished


Computed Assisted Detection of Lymph Node Metastasis of Prostate Cancer in Contrast Enhanced MRI
	Prostate cancer is one of the most common cancers in men and a major cause of death. In the Netherlands, 8.000 new cases were diagnosed in 2004 and 2.400 men died that year of prostate cancer. Because prostate cancer grows slowly, it can, generally speaking, be effectively treated when it is detected at an early stage. However, physical problems usually related to urination can occur when the cancer has had an opportunity to grow. From the new diagnosed cases, 75% of the men in question were 65 years or older. In diagnoses and treatment, the determination of lymph node involvement is an important issue. The lymph nodes in the pelvic region give information about the spreading of the cancerous cells. When cancerous cells travel through the lymphatic system the lymph nodes in the pelvic system are the first to be infected. Presently, imaging techniques are becoming more important in the detection of metastases in lymph nodes. Especially the usage of different MRI techniques in combination with new contrast agents based on ultra small paramagnetic iron oxide particles (USPIO-particles, Sinerem) have improved the sensitivity and specificity of the detection of lymph node metastases. Detection of lymph nodes metastases on MR images is a tedious and time consuming process. Even for an experienced radiologist it can take up to 45 minutes per patient. Development of a computer algorithm that helps the radiologist in finding these lymph nodes would increase productivity and may even help to improve diagnosis. In this thesis such a CAD-algorithm is proposed.
Last modified: March 28, 2008	Status is Finished


Segmentation Techniques for the Visible Mouse
	In this report three segmentation techniques have been implemented to segment visible mouse data/magnetic resonance images of a mouse. The Magnetic Resonance Lab (MRL) in Eindhoven initiated a project to scan a mouse and its organs. By visualizing the organs of the mouse on the computer a 3D database of the mouse is created. This database can serve as a reference for researchers that are performing experiments on mice. Before visualization the organs must be segmented.
Last modified: March 28, 2008	Status is Finished


Investigation of Leaf Positioning Accuracy of Two Types of Siemens MLCs Making Use of an EPID
	With the introduction of IMRT, the position and reproducibility of the leaves become more important, especially for abutting segments. A calibration error is a systematic error in leaf positioning. The goals are: • the development of a leaf verification method applicable to a Siemens linac using a CCD-camera based EPID. • the verification of the leaf positions of two different types of Siemens MLCs and to check whether the MLCs and the calibration method using the light field are accurate enough for IMRT.
Last modified: March 28, 2008	Status is Finished


Correction of Cupping Artifacts in Megavoltage Cone-Beam CT
	Megavoltage Cone Beam CT (MV CBCT) can be used for 3D imaging of the patient anatomy on the treatment table. To use MV CBCT images for dose calculation purposes, reliable electron density (ED) distributions are needed. Patient scatter, beam hardening and softening effects result in cupping artifacts in MV CBCT images. A method based on transmission images is presented to correct for these effects without using prior knowledge of the object’s geometry.
Last modified: March 28, 2008	Status is Finished


Qualitative Determination of Errors Causing Portal Dose Differences Using Gamma Evaluation Parameters
	Purpose: To investigate the feasibility of a qualitative automated error detection method based on parameters from a gamma evaluation which can be used to evaluate deviations in portal dose distributions. Methods and materials: The accuracy of the evaluation method is investigated by performing a simulation study. Portal dose images (PDIs) are generated and perturbations due to machine output variation, set-up errors and organ motion are imposed to the images; a gamma evaluation is applied. In nearly homogeneous regions of a PDI, dose differences are calculated; in regions with dose variations, displacements are determined. A perturbed image is corrected for machine output variations before displacements are determined. Because a portal dose distribution is not invariant to set-up errors and organ motion, portal dose images are also simulated using a twodimensional (2-D) portal dose prediction model based on pencil beam scatter kernels. Using this model, changes in PDIs due to patient perturbations could be simulated. Finally, the error qualification method is tested on clinical PDIs. Results: A distinction can be made between errors caused by dose differences and displacement of patient or organs. Changes in portal dose distribution caused by patient or organ displacements do not influence the sensitivity of the error qualification method. Accuracy is dependent on the treatment site. Conclusion: The method that is developed can be useful in clinical practice for qualitative automated error detection using electronic portal dosimetry.
Last modified: March 28, 2008	Status is Finished


Tumor Delineation Based on Time Activity Curve Differences Assessed with Dynamic FDG-PET
	Dynamic PET can be used to obtain time activity curves (TACs) describing the activity profile of a tracer for each independent voxel in time. The uptake kinetics of the tracer is dependent on tracer specific properties as also on the biologic characteristics of underlying tissue. In this study, it was hypothesized that by analyzing TACs it is possible to differentiate the tumor from surrounding tissues better than from static PET measurements or by manual tumor delineation.
Last modified: March 28, 2008	Status is Finished


Dosimetric Calibration of an Amorphous Silicon Electronic Portal Imaging Device
	Electronic Portal Imaging Devices (EPIDs) are not only suited for patient set-up verification and detection of organ motion but can also be used for dosimetric verification of (complex) treatment techniques or in-vivodosimetry. The aim of this work was to investigate the dosimetric properties of a new commercially available amorphous silicon (aSi) EPID and the feasibility of a calibration procedure to obtain relative full-scatter portal dose distributions.
Last modified: March 28, 2008	Status is Finished


Fetal Volume Measurements with 3D Ultrasound
	Maxima Medical Center is specialized in perinatal and neonatal care. In the departments of Gynaecology and Obstetrics of this hospital there are various research projects to improve medical care. In these departments a new ultrasound modality is used: Kretz Voluson 730, which gives a possibility to perform 3D/4D ultrasound investigations. With 3D-ultrasound all three projections of the examined fetus together with the volume representation can be displayed. It is also possible to image the movements of a fetus in time (4D ultrasound; x,y,z,t). With 3D ultrasound the volume (weight) of an fetus can be determined. Volume (weight) is an important measurement to judge the development and condition of the fetus. On the Kretz the volume of the fetus can be determined manually. This operation is complicated and time-consuming. The measurement error of this volume calculation operation is unknown. The aim of this study is to develop an automatic volume calculation of a fetus based on 3D ultrasound volume scans.
Last modified: March 28, 2008	Status is Finished


Development of an Image Guided Neurosurgery Application
	The goal of the project is to combine segmentation, registration, DTI fiber tracking, geometric measurements and medical tool tracking, with our multimodal visualization environment. Mevislab will be used as a development environment because it contains a large amount of algorithms as directly available modules. Medtronic will provide a Stealth Station and the necessary API Stealth Link to access the data of the Stealth Station.
Last modified: March 10, 2008	Status is Ongoing


Preoperative Visualization Techniques for Brain Tumor Resection
	Around 17.500 people in the U.S. die from primary nervous-system tumors each year. One of the most common treatments for brain tumors is resection. This surgical procedure requires very careful planning and execution. To assist the neurosurgeon with the planning of such delicate procedures a visualization tool should be designed that integrates datasets from various modalities, such as MR, CT, fMRI and DTI (diffusion tensor images) and displays them in an intuitive way. An important requirement is that the design of the tool should be flexible enough to integrate navigational tool tracking information when used for surgical guidance in the operating room itself.
Last modified: January 08, 2008	Status is Finished


GPU Based VolumeFlies for Illustrative Volume Rendering
	Illustrations are able to transmit and concentrate the attention of the reader (user) to what is important and not to details that are not necessarily relevant. In the last years, inspired by these illustrations the so called illustrative volume rendering has emerged. The goal of this project is to extent Stef Busking work (see finish projects in illustrative Rendering) by implementing a hardware based VolumeFlies framework, using the flexibility of nowadays Graphics Cards. GPU based particle systems for other purposes, like flow visualization, already exist. It is also of interest to improve the specific implementations of the VolumeFlies framework modules.
Last modified: December 10, 2007	Status is Finished


Computer-aided analysis of non-mass-like enhancements in breast MR
	Dynamic-contrast enhanced MR imaging has become a valuable tool for detection, diagnosis and management of breast cancer, exposing cancerous tissue as areas of mass- or non-mass-like enhancement. Non-mass-like enhancements are often related to early forms of breast cancer and are therewith diagnostic criteria of particular importance. Unlike mass-like enhancements describing compact regions of suspiciously enhancing tissue with distinct morphologic and dynamic characteristics, non-mass-like enhancements are complex distribution patterns of suspiciously enhancing tissue interspersed by areas of normal tissue. The aim of this project is to analyze the clinical requirements and to develop, implement and evaluate methods that help radiologists to detect, delineate and/or assess areas of non-mass-like enhancements. Candidates should be interested in mathematics, image processing and/or pattern recognition, but also in the biomedical aspects of breast cancer diagnosis. Experiences in programming with C/C++ are of advantage, but not a prerequisite. The project is part of a larger collaboration between the TU/e and Philips Medical Systems, Best in the field of computer-aided diagnosis in breast MRI.
Last modified: October 04, 2007	Status is Finished


Perceptual Grouping - Curvature Enhanced Closure of Elongated Structures
	In biomedical image analysis one often needs to detect elongated structures like blood vessels, contours, or catheters. Noise and occlusions may interrupt elongated structures. Earlier work focussed on filling gaps in contours using the framework of orientation scores and G-convolution. The aim of this project is to extent this work by making the G-convolution adaptive with respect to local estimates of the curvature.
Last modified: August 27, 2007	Status is Finished


Intensity and Feature Based 3D Rigid Registration of Pre- and Intra-Operative MR Brain Scans
	In hospitals, image guided surgery is done on data made before an operation. During neurosurgery, the shape of a patient’s brain can change due to brain shift or tumor resection. Because of these tissue deformations, navigation and planning on real time MR data instead of pre-operative data can be of great help during neurosurgery. The intra-operative Polestar N20 MRI scanner (0.15 Tesla) allows the surgeon to navigate on real-time data using the Medtronic Stealth Station and the Polestar N20. The advantages of the Polestar are the low cost and there is no need for an expensive change of instruments and operating room. The compromise here is the image quality, the low field results in a low resolution and low SNR. The aim of this project is to register in 3D the pre-operative data and the intra-operative data made by the Polestar N20 to make high resolution information available for navigational purposes during surgery.
Last modified: August 27, 2007	Status is Finished


Intensity and Gradient Based Rigid 3D Registration of Pre and Intra-Operative MRI Brain Scans
	The University Hospital Maastricht AZM has acquired an open MRI scanner (Medtronic/Odin Polestar N20) for neurosurgery applications. This is the first low-field open MRI scanner in the Netherlands. The TU/e and azM are scheduling a series of collaborations, in order to improve the use and navigation issues in clinical practice (resection surgery and deep brain stimulation). The low magnetic field of the polestar makes it usable in the operation room but it inherently involves a lower signal to noise ratio. Surgical planning and inter operative navigation could be improved if the more detailed images from the high field MRI scanners could be warped onto the images from the polestar. The registration method should be able to deal with the low signal to noise ratio in the data from the polestar. Field inhomogeneities cause deformations in the polestar images. The deformations in the center of the image appeared to be quite small, but increase towards the edges. This has to be taken into account during registration./div>
Last modified: August 27, 2007	Status is Finished


Atrial Fibrillation treatment guidance by automated contrast-enhanced masking
	The project targets the clinical application area of atrial fibrillation (AF) treatment. AF is an arrhythmic condition of the atria (chambers) of the heart that results in reduced heart function. AF treatment consists of stopping the arrhythmia by electrically isolating part of the atrium from the source of unwanted activations. The most common place of treatment is the left atrium (LA) and the pulmonary veins attached to the LA. Treatment consists of isolation of the sites by means of tissue ablation: applying RF energy from a catheter inside the atrium, forming scar tissue (ablation lesion). Although electrophysiologists usually have a sound conception of the actual catheter position, it can be rather complex to retain an anatomical overview of ablation positions during the procedure, especially since the inner layer of the heart (endocard) and the pulmonary vein branches are not visible on conventional X-ray. Therefore, during the procedure the physician occasionally inserts several ‘puffs’ of contrast agent into the pulmonary vein branch(es) to guide catheter placement and retrieve an (intraprocedural) update on possible anatomy deformation. The purpose of the study is to automatically segment the contrast enhanced cardiac chamber and use the segmented mask as an overlay for the live fluoroscopy image stream. Possibly, by use of a cardiac model (external or internal provided CT dataset) segmentation results can be improved for low concentration contrast agent while compensating for cardiac/patient motion (i.e. registration to the correct cardiac phase). Literature survey on segmentation algorithms Implementation of (model-based) segmentation and mask overlay Verification study Report & Presentation
Last modified: August 23, 2007	Status is Open


Vascular remodeling analysis by Two Photon Laser Scanning Microscopy
	In the western world, vascular disorders form a major medical problem. To increase the knowledge of the underlying mechanisms of, for example, atherosclerosis, extensive research is performed, to find causes of these disorders. At the university of Maastricht, a special type of microscopy is used, called two photon laser scanning microscopy (TPLSM). Using the TPLSM, three dimensional images can be extracted from viable arteries. To describe the processes occurring in the large arteries as a reaction on changing circumstances, two methods are proposed. The first method focuses on the estimation of the radius of imaged arteries. The second method can be used for the counting of cells within the vessel wall. The estimation of the radius of an artery is complicated by the fact that it is often only possible to image a small part of the vessel wall. To get reliable results, a new method has been devised, based on the Hough transform. In this study, it is shown that the proposed method is more accurate than other methods proposed in literature. Using the proposed method, an accuracy expressed in the standard deviation of the error, of 2.5 % can be achieved, against 10 % that is obtained using the often used least squares method. The second subject that is covered by this work, is the counting of cells. A modelbased approach has been taken, to fit ellipsoids on sets of potential edge points. Despite of the need for further research, the preliminary results show to be promising. Virtually all nuclei are detected, while after merging over-segmented nuclei, few errors can be identified.
Last modified: August 23, 2007	Status is Finished


MR-based quantitative analysis of myocardial contraction
	The problem Cardiac resynchronization therapy (CRT) consists of the placement of a pacemaker with one or more electrodes (stimulation points) to resynchronize the contraction of the myocardium (heart muscle) in patients with ventricular dysynchrony. It helps the lower chambers of the heart (left and right ventricles) to beat together again. In order to reach the best electrode placement, the myocardial contractile patterns should be known as a function of the location on the myocardium and areas with similar abnormal contraction should be identified. State of the art CRT is an established therapy for patients with advanced heart failure. However, about onethird of the patients lack a favorable response to this therapy. Incorrect selection of the pacing (stimulation) location(s) is considered to be one of the reasons. The contraction of the myocardium is most frequently assessed using cardiac ultra-sound imaging. More recently, cardiac magnetic resonance (CMR) imaging has proven to be a powerful technique for the visualization and quantification of right- and left-ventricular myocardial contraction (the pictures below show various time moment in the cardiac cycle). With the most recent version of the PMS ViewForum Cardiac MR Analysis software product, only the time moment of maximum contraction (largest myocardial wall thickness) can be measured and visualized as function of the location on the myocardium. The actual local contraction (wall thickness) patterns and the correlation between patterns at different locations are not analyzed. Aim The aim is to invent, implement and test algorithms for the analysis of local myocardial contractile patterns and for the identification of areas with similar contractile behavior. Areas with abnormal contraction can then be regarded as candidate areas for stimulation by a pacemaker. Learn the programming environment (C, C++, EasyScil) Study the literature on CRT (procedure, planning methods, …) Study the literature on Cardiac MR for myocardial motion analysis Invent/select candidate myocardial contraction analysis methods Implement the selected candidate methods into a demonstrator (EasyScil) Evaluate the implemented methods on image data from heart failure patients Write a report, prepare and give a final presentation
Last modified: August 15, 2007	Status is Finished


Sparse Active Shape Modeling
	Active Shape Models (ASMs) are statistical shape models extended with a matching component. Thus, by matching an ASM to a (medical) dataset, image segmentation can be achieved. Shapes are usually described as a set of points obtained by sampling objects. Therefore, ASMs require a point-based update, i.e. an update for each point present in the shape description, to find a new instance of the ASM that matches the data (better). It can happen that at certain positions, such updates cannot be obtained from the data, but still a segmentation has to be obtained. This requires modifications to the standard ASM, to enable matching of an ASM based on a densely sampled shape class to a sparse(r) representation of the shape class in an unseen subject. This project aims at investigating three different methods for doing so, and compares the accuracy of the segmentation results obtained by the different methods.
Last modified: August 15, 2007	Status is Finished


Can T2-weighted images help in the classification of breast lesions?
	MRI of the breast makes use of the fact that tumors show angiogenesis. This can be visualized by using a dynamic, contrast-enhanced T1-weighted scan. The rate of uptake and wash-out of contrast medium is an important sign of malignancy. However, some lesions (most notable early stages of cancer) cannot be identified by the speed of contrast uptake and wash-out alone. Presently, the shape of the enhancing area identifies them to radiologists. A T2 weighted scan is also part of a breast MRI examination. The usefulness of this scan in the classification of malignant and benign lesions is not fully clear. There are indications that these images can improve in classification of breast lesions. The aim of the assignment is to come up with algorithms that measure properties of (enhancing) breast MR lesions in T2 weighted images, which would improve classification. This work will be part of a research project to develop a full CAD system.
Last modified: August 08, 2007	Status is Finished


Automatic Electrode Trajectory Planning for Deep Brain Stimulation
	To alleviate tremors, akinesia, and rigidity in Parkinson's Disease patients, chronical deep brain stimulation can be performed. Subthalamic nucleus (STN) is one of the common targets. This nucleus lies just above the brainstem, in the basal ganglia. Navigation towards this core is not without risks, and therefore requires difficult and time-consuming pre-operative planning. This trajectory needs to meet several strict requirements. We propose to develop an automatic trajectory finding algorithm which meets all the requirements of the trajectory. Afterwards, the resulting trajectory should be visualized in 3D, together with the segmented structures, to enable the neurosurgeon to examine the trajectory and correct it if necessary.
Last modified: May 23, 2007	Status is Finished


Detection of electrode positions for combined EEG/fMRI studies
	Background Electro-encephalography (EEG) is the recording of the potential differences from electrodes attached to the head. The generators of EEG consist of current sources that are activated by interacting neurons in the brain. Functional Magnetic Resonance Imaging (fMRI) is a fast scanning technique with which every 2 to 3 s a complete image of the brain is made. By making a statistical comparison between fMRI images made during a task and during rest, fMRI can be used to detect which parts of the brain are involved during the execution of the task under study. Recently it has become possible to record EEG and fMRI simultaneously. This combination of technologies is particularly promising for patients with epilepsy. In principle, EEG is used to detect the presence of epileptic activity, and fMRI is used to make the contrast between scans coinciding with such activity and scans made in the absence of epileptic activity. The contrast image indicates where in the brain the epilepsy is generated. Problem The analysis of simultaneous EEG/fMRI data is based on a classification of the EEG data into pathologic and non-pathologic epochs. This can be done by visual inspection of the EEG, but also by applying so-called source localization techniques. In the latter approach, dipole models are fitted to the observed EEG patterns. To make such advanced methods possible, one needs the 3-D coordinates of the EEG electrodes. These coordinates can in principle be obtained by measuring the electrode positions using a 3D pointer device (e.g. a Polhemus). However, such a procedure is time consuming (half an hour) and also of limited accuracy because the Polhemus coordinates need to be converted to the coordinate system of the MRI. Goal A fortunate coincidence of EEG/fMRI measurements is that the electrodes leave very clear artifacts on the MR images, see the project figure. The goal of this project is to develop a method to detect the electrode positions directly from the MRI artifacts using image processing and feature detection algorithms. The algorithms used for electrode detection should be as automatic as possible. Furthermore, the algorithm should not only detect the electrode position as a list of 3D coordinates, but it should also indicate to which electrode label each detected electrode belongs. One source of data for the electrode position is the anatomical MR scan. In addition to that, one has a list of electrode labels and their approximate relative positions. This template has been determined for subjects that underwent EEG/fMRI and for whom 3D positions were determined with a Polhemus. Because the electrodes are attached to a nylon cap, it can be assumed the relative electrode positions do not vary more over subjects than one to two cm. Required A student Biomedical Technology who: is in the last year of his/her master study; has knowledge of MatLab, C++ or another programming language; has interest in clinical studies. Supervision The trainee will be based at the department Onderzoek & Ontwikkeling of Epilepsiecentrum Kempenhaeghe and participate in the research project ‘Combining EEG and fMRI for successful delineation of the epileptogenic zone relative to a lesion’. This research project will be performed in close cooperation with the VU medical centre in Amsterdam and University Medical Centre in Utrecht. Direct supervisors are Dr. Pauly Ossenblok, medical physicist of Epileptiecentrum Kempenheaghe (e-mail: OssenblokP@Kempenhaeghe.nl) and Dr. Jan C. de Munck, theoretical physicist at VU medical centre (email: JC.deMunck@VUmc). Supervisor within the Biomedical Imaging group of the department of Biomedical Engineering will be Bram Platel.
Last modified: April 10, 2007	Status is Open


Cardiac microvessel analysis from cryomicrotomy
	The cardiac muscle (myocardium) contains a numerous amount (i.e. millions) of bloodvessels that range in diameter from a few micrometers to several millimeters. To quantify parameters describing the microvessel tree, like bifurcation level, distance between bifurcations, vessel diameter, a.o., first the complete vascular tree has to be extracted from a dataset. For a proper resolution, a digital camera with an in-plane resolution of approximately 40 micrometers and a cryomicrotome with the ability to cut slices of similar thickness are used to image goat hearts (ex-vivo) in 3D. Artifacts ocurring due to an assymetric point distribution function are repaired first. Multiscale image analysis is developed and used to extract the blood vessels from the data sets, which are typically in the order of 2000x2000x2000 voxels. Finally, a skeleton from the vascular tree is built from which information describing the tree can be obtained, and which is further used as basis to find the tiny vessels in a more accurate sense, to quantify, e.g., vessel diameter.
Last modified: February 25, 2007	Status is Finished


VolumeFlies - a Smart-Particle-inspired Framework for Illustrative Volume Rendering
	Volume rendering is a well established technique that is used for the visualization of medical volume data. Most of the techniques are based in the generation of images based on an approximation of a realistic physical model. However, if we look at the anatomic books, they still use manual illustration instead of photographs or other more realistic means. This project wants to explore illustration techniques for volume rendering and combine them with the more common volume rendering techniques to achieve enhanced images.
Last modified: January 11, 2007	Status is Finished


Improved Pharmacokinetic Analysis of Dynamic Contrast Enhanced MR Breast Images
	MR is becoming an important modality in the diagnosis of breast cancer. The main problem here is not the sensitivity but the specificity of contrast-enhanced MRI of the breast: it will enhance most lesions but also a lot of other tissue. State of the art is to make a dynamic contrast-enhanced T1 weighted scan with 4-6 stacks of 30+ images at time intervals of 90-120 seconds. As faster scanning protocols become available, it becomes possible to apply more advanced analysis to the contrast uptake of various lesions. In this project, we want to investigate this.
Last modified: January 11, 2007	Status is Finished


Image Enhancement in Quality Control of X-Ray Grids
	X-ray grids are an indispensable component in any diagnostic X-ray system. This study aims at an increase of the resolution of the digital detector, especially with super-resolution methods. These techniques use sub-pixel shifts of the detector and combine multiple shots of the same object in one better image.
Last modified: January 11, 2007	Status is Finished


Quantification of Collagen Orientation in 3D Engineered Tissue
	Analysis of 3D 2-photon microscopy images of remodeling collagen fibres in artifical heartvalve tissue.
Last modified: January 11, 2007	Status is Finished


Perceptual Grouping - The Closure of Gaps within Elongated Structures in Medical Images
	In biomedical image analysis one often needs to detect elongated structures like blood vessels, contours, or catheters. Noise and occlusions may interrupt longated structures . Thus, standard line detection algorithms may just capture parts of an elongated structure rather than the complete line or contour. The aim of this project is the development of an algorithm that detects elongated segments in orientation scores and fills the gaps in between with stochastic completion fields suited to the medical task at hand.
Last modified: January 11, 2007	Status is Finished


Visualization of DTI in a VR Environment
	MR diffusion tensor imaging (DTI) measures the diffusion of water molecules in tissue. The diffusion is expressed by a second-order tensor. This tensor is an indicator of the underlying structure, for example, the fibers in the brain. The task of this project will be to transfer the existing DTI tool to the Virtual Reality set-up available at the BIOMIM image-analysis lab.
Last modified: January 11, 2007	Status is Finished


The Effects of Filtering on Visualization and Detection of Colonic Polyps in Ultra Low Dose Multi-Detector CT Data
	To make screening of the colon possible with CT, the dose has to be reduced substantially. The goal of this project is to reduce the noise in low-dose CT and to evaluate the quality of the segmentation, visualization and detection.
Last modified: January 11, 2007	Status is Finished


Methods for 3D Orientation Analysis and their Application to the Study of Arterial Remodeling Microscopy
	Two-photon fluorescence microscopy is a relatively new imaging modality. The formation of new tissue during the formation of atherosclerotic plaque of mice is investigated. Goal of the project is to set up an image-analysis environment. Focus will be on shape, motion and textures analysis and 3D visualization of cell structures.
Last modified: January 11, 2007	Status is Finished


Semi-Automatic Quantification of the Internal Elastic Lamina Fenestrae in Remodeling Arteries
	Two-photon fluorescence microscopy is a relatively new imaging modality. The formation of new tissue during the formation of atherosclerotic plaque of mice is investigated. Goal of the project is to set up an image-analysis environment. Focus will be on shape, motion and textures analysis and 3D visualization of cell structures.
Last modified: January 11, 2007	Status is Finished


Hierarchical Visualization using Fiber Clustering
	Diffusion Tensor Imaging (DTI) is a Magnetic Resonance Imaging (MRI) technique for measuring diffusion in biological tissue. DTI data is difficult to visualize because of the high amount of information available in each sample point. A prominent DTI visualization technique is fiber tracking. The fiber tracking algorithm creates streamlines (fibers) that correspond to the major white matter fiber bundles in the brain. Individual structures are virtually indistinguishable and it is very difficult to extract any useful information. To overcome this problem, we use a clustering algorithm to organize the fibers into groups that are meaningful and anatomically correct.
Last modified: January 11, 2007	Status is Finished


Cardiac MR Segmentation using Active Contours
	Design and implementation of a dedicated tool for segmentation of cardiac structures from MR data using Active Contours and knowledge about anatomy and movement of cardiac structures.
Last modified: January 11, 2007	Status is Finished


Pulmonary Artery and Vein Separation in Multi-Detector Chest CT
	Computer aided detection of pulmonary emboli require automatic and appropriate segmentation of the pulmonary vessel-tree. In this master project, methods will be developed to detect the major pulmonary vessels. In a next step, arteries will be separated from veins.
Last modified: January 11, 2007	Status is Finished


Optic Flow from Dynamic Anchor Point Attributes
	Optic flow describes the apparent motion that is present in an image sequence. We show the feasibility of obtaining optic flow from dynamic properties of a sparse set of so called anchor points. The advantage of approaching the optic flow estimation problem using these anchor points is that in these points the notorious aperture problem does not manifest itself. The proposed optic flow estimation method succeeds in finding a dense vector field that approaches the optic flow field from a sparse set of inherent multi scale anchor points. As opposed to classical optic flow estimation schemes the proposed method accounts for an explicit scale component of the vector field, which could encode a hitherto unknown dynamic property.
Last modified: January 11, 2007	Status is Finished


Context-Enhanced Detection of Electrophysiology Catheters in Noisy Fluoroscopy Images
	Cardiac Electrophysiology procedures are performed under continuous X-ray fluoroscopy surveillance. The goal of the project is to develop image analysis techniques to detect Electrophysiology catheters in the fluoroscopic images. The idea is to improve current methods to detect thin lines (like the catheters) by taking a larger spatial context into account, using the assumptions that catheter-lines are continuous and exhibit finite curvature.
Last modified: January 11, 2007	Status is Finished


Geometrical Methods in Diffusion - Tensor Regularization
	A diffusion tensor image (DTI) is a tensor valued magnetic resonance image that captures the apparent local diffusivity of water molecules in each spatial direction. In this project the goal is twofold, (i) to devise a mathematically well-founded regularization scheme, and (ii) to define elongated structures.
Last modified: January 11, 2007	Status is Finished


Atherosclerotic Plaque Classification
	Atherosclerosis is the formation of plaque in the larger bloodvessels. It can lead to serious obstruction of the bloodflow, and rupture may give rise to infarcts in the brain. Understanding the nature of the constituents is essential for diagnosis and treatment planning. The goal of this project is to automatically find the classification of the different tissues in and around the plaque, by means of non-invasive MRI techniques.
Last modified: January 11, 2007	Status is Finished


Image Processing Functionality for Computer-Assisted Diagnosis (CAD) and Quantification (CAQ) of Lungs Imaged with CT
	In this project, we will investigate various possibilities for extending the image processing functionality for CT lung diagnosis: lung detection, lung segmentation, bronchia tracking and vessel tracking.
Last modified: January 11, 2007	Status is Finished


Segmentation of the Visible Mouse Dataset
	The Visible Mouse is a project of the MR Laboratory to acquire high resolution datasets of mice by means of high-field MRI. The mouse is the new laboratory for molecular imaging and experimental radiology. In this project a toolkit is developed in Mathematica, based on deformable contours (2D) and surfaces (3D), to segment the 3D datasets of the Visible Mouse. The high level flexible toolkit can incorporate physical models and statistics into the equations for snakes and level sets.
Last modified: January 11, 2007	Status is Finished


3D Visualization of Diffusion Tensor Imaging (Berenschot)
	MR diffusion tensor imaging measures the diffusion of water molecules in tissue. The diffusion is expressed by a tensor. This tensor is an indicator of the underlying structure, for example the fibers in the brain. This projects consist in investigating 3D visualization techniques that better allow the interpretation of this tensor data.
Last modified: January 11, 2007	Status is Finished


Multi-Scale Hierarchical Segmentation
	For many image processing tasks, hierarchical and topological methods are re-quired. New approaches to image analysis involving the deep structure of images are promising, but many questions about points in scale space are still unanswered. In this thesis, a second order reconstruction algorithm for multi-scale points is presented in order to extract information about these points. It is tested with random points and spatially equidistant points in scale space. Us-ing the equidistant points, an optimal distance between reconstruction points is measured, taking the limited machine precision into account. For this measurement, the condition number of the correlation matrix is used. The algorithm is also tested with multi-scale critical points and multi-scale top points. Two possible applications for the reconstruction from multi-scale top points are discussed: data compression for images using reconstructions and content based image retrieval using coefficients of the reconstruction algorithm. The results of both feasibility studies are promising.
Last modified: January 11, 2007	Status is Finished


Reconstruction from Top-Points in Deep-Structure Image Analysis
	For many image processing tasks, hierarchical and topological methods are required. New approaches to image analysis involving the deep structure of images are promising, but many questions about points in scale space are still unanswered. In this thesis, a second order reconstruction algorithm for multiscale points is presented in order to extract information about these points. It is tested with random points and spatially equidistant points in scale space. Using the equidistant points, an optimal distance between reconstruction points is measured, taking the limited machine precision into account. For this measurement, the condition number of the correlation matrix is used. The algorithm is also tested with multiscale critical points and multiscale top points. Two possible applications for the reconstruction from multiscale top points are discussed: data compression for images using reconstructions and content based image retrieval using coefficients of the reconstruction algorithm. The results of both feasibility studies are promising.
Last modified: January 11, 2007	Status is Finished


Multi-Scale Optic Flow Analysis for Magnetic Resonance Tagging
	One main task from the MR tagging application is to extract information about deformation of tissue. This information can be described in terms of the optic flow field. The optic flow field (or some people call it the optical flow field), is a vector field that describes an apparent movement of pixels in an image sequence. Our optic flow model is based on the biological visual system, which uses the Gaussian scale-space representation.
Last modified: January 11, 2007	Status is Finished


Quantification of Collagen Architecture in 3D Engineered Tissue
	The aim of this project is to quantify the collagen architecture in 3D engineered tissues, which includes the following steps: investigation of the optimal image settings for 2-photon microscopy, validation of the algorithms for a large set of data, quantification of 3D cellular orientation and quantification of the diameter of collagen fibers.
Last modified: November 21, 2006	Status is Open


Multiscale Analysis of Deletions and Amplifications in Human DNA-Microarray Signals
	The task of the master student is to develop a computerized approach for detection of deletions (sinks) and amplifications (blobs) in DNA-microarray signals using a multiscale approach. The underlying signal is nonuniformly sampled and the array measurements occasionally overlap (BAC-microarrays). Eventually, the output of the algorithm is a list with deletions and amplifications including the chromosomal (base pair) begin and end points of each deletion/amplification.
Last modified: November 21, 2006	Status is Ongoing