Medical image computing

Medical image computing (MIC) is an interdisciplinary field at the intersection of computer science, information engineering, electrical engineering, physics, mathematics and medicine. This field develops computational and mathematical methods for solving problems pertaining to medical images and their use for biomedical research and clinical care. The main goal of MIC is to extract clinically relevant information or knowledge from medical images. While closely related to the field of medical imaging, MIC focuses on the computational analysis of the images, not their acquisition. The methods can be grouped into several broad categories: , image registration, image-based physiological modeling, and others. Data forms Medical image computing typically operates on uniformly sampled data with regular x-y-z spatial spacing (images in 2D and volumes in 3D, generically referred to as images). At each sample point, data is commonly represented in integral form such as signed and u

Unsupervised multiscale image segmentation

Ana Dimitrijevic

In the early stages of visual information processing one of the most fundamental and complex task is segmentation. This process is used to divide images or image sequences into meaningful regions or objects. Such a representation is definitely needed to go from the first rough description of the scene (i.e. light intensity) up to the semantic level, where one is able to discriminate shapes and other high-level visual primitives. The biological visual system performs this extremely difficult task with such ease that is still a major challenge for computer vision to reach this level of speed and accuracy. It is by now widely believed that a method inspired by biological vision has the best chances of accomplishing this goal. Therefore, our main goal in this thesis is to step on biologically inspired schemes to propose new and efficient techniques to solve the image segmentation problem. The starting point and key concept in this research, is the multiresolution paradigm. Indeed, the fact that the Human Visual System (HVS) processes the information in a multiresolution fashion is now well established. Several techniques have been introduced to model this behaviour. Scale-space is one of them and we will thus focus on such a formulation of the segmentation problem. Such a representation is composed by a stack of successive versions of the original data set at coarser scales. It is assumed that, the bigger the scale, the less information referred to local characteristics of the input data will appear. We also impose that general information applying to large scales will last through scale. Taking that into account, it is reasonable to think that local and high resolution scale information can be related to general and low resolution information. This will enable us to extract image structures. We thus propose to link the multiresolution approach to a computational model based on the analysis of the entire range of significant information across scales. The inherent multiscale image structure of the non-linear scale-space guides a coarse to fine segmentation using different strategies. The first one might be termed extrinsic because the non-linear scale-space is analyzed by an external process of linking information through scales. This leads to a completely adaptive scheme that can then be coupled to a labeling procedure in order to define a genuine segmentation. The second one could be labeled intrinsic because it adapts to the image through its inner properties and incorporate this in the labeling step of the main algorithm. These two approaches are particular examples of multiscale image segmentation. We also apply the same scale-space concept to define a segmentation variational formulation that leads to a region merging segmentation. In order to illustrate the proposed segmentation schemes, we apply our algorithms to gray-scale, noisy, color and texture images. Finally, it should be mentioned that the main problem addressed in this thesis, general image segmentation, plays an important role in the image understanding process, especially those that involve natural scenes rich in color and texture. The potential applications of the algorithms are thus numerous: medical image analysis, interactive multimedia developments, object based coding and compression, surveillance applications, image content based browsing and retrieval, etc.

EPFL2006

Determining patterns of post-stroke motor recovery through longitudinal multimodal MRI: A step towards patient stratification

Julia Brügger

Motivated by the need for a better understanding of post-stroke recovery and new biomarkers to improve stroke patient stratification and outcomes, this thesis investigated structure-function coupling and its role in post-stroke recovery. Furthermore, in order to increase data comparability between sessions and centers (a critical challenge in contemporary clinical research), this thesis assessed the reproducibility of microstructure-informed structural connectivity measures in a multi-center dataset.Stroke is one of the major sources of permanent impairment, frequently of motor origin. However the clinical picture is very heterogeneous: patients show divergent courses of recovery and the underlying mechanisms are still unclear. In addition, current treatments still have limited success and are restricted to a 'one-fits-all' approach, not considering the individual patient's phenotype. As efforts to stratify patients based on structural or functional biomarkers are still needed, we chose to investigate the potential of a multimodal biomarker, the Structural Decoupling Index (SDI) (Preti & Van De Ville, 2019b), a new metric assessing the structure-function coupling strength in the brain. The first part (Studies 1 and 2) of this thesis investigates the potential of the SDI as a biomarker. In Study 1, the goals were to evaluate the feasibility of applying the SDI on an individual level in healthy older adults and to investigate the effect of an acute stroke on it. Consistent with the literature, we found a network gradient of SDI in healthy older adults, from high coupling in lower-level sensory areas, to low coupling in higher-level cognitive areas. This confirmed the applicability of SDI on an individual level. Furthermore, we showed that stroke impacts the SDI, with a higher decoupling on the ipsi- compared to the contralesional side, and with network-specific effects in RH stroke patients. In Study 2, the goal was to see whether the SDI evolved over time post-stroke and whether it links to behavior. The longitudinal analysis revealed differential network effects of stroke at T1 and T3. Furthermore, we showed that impairment in cognitive and psychological behavioral domains significantly correlates with variations in SDI in a number of key areas including motor regions (e.g., primary motor cortex) and that the brain pattern associated to behavior changes between T1 and T2. Surprisingly, motor performance did not explain variability in key motor areas (e.g., primary motor cortex). The link between post-stroke behavioral performance and SDI underlines its potential clinical relevance.Driven by the quest for more reproducible analysis pipelines in the context of longitudinal and clinical studies, the second part of this thesis (Study 3) addressed the subject-specific reproducibility and repeatability of microstructure-informed tractography. Through a multi-center study, we demonstrated its high reproducibility and subject-specificity, and we found evidence for increased biological accuracy. My thesis made a contribution by showing that the SDI is sensitive to pathophysiological changes that occur following a stroke and that it links to clinically relevant behavioral measures. In addition, it confirms the reproducibility and subject-specificity of microstructure-informed tractography. Together, my findings pave the way towards patient stratification and more personalized treatments in stroke rehabilitation.

EPFL2022

Unsupervised multiscale image segmentation

Ana Dimitrijevic

EPFL2006

Determining patterns of post-stroke motor recovery through longitudinal multimodal MRI: A step towards patient stratification

Julia Brügger

EPFL2022

Unsupervised multiscale image segmentation

Interface graphique pour la visualisation d'images médicales

Determining patterns of post-stroke motor recovery through longitudinal multimodal MRI: A step towards patient stratification

Unsupervised multiscale image segmentation

Determining patterns of post-stroke motor recovery through longitudinal multimodal MRI: A step towards patient stratification

Interface graphique pour la visualisation d'images médicales