The resection of tumors within the primary motor cortex is a constant challenge. Although tractography may help in preoperative planning, it has limited application. While it can give valuable information on subcortical fibers, it is less accurate in the cortical layer of the brain. A 38-year-old patient presented with paresis of the right hand and focal epileptic seizures due to a tumor in the left precentral gyrus. Transcranial magnetic stimulation was not applicable due to seizures, so microsurgical resection was performed with preoperative tractography and intraoperative direct electrical stimulation. A histopathological assessment revealed a diagnosis of glioblastoma. Postoperative magnetic resonance imaging (MRI) showed complete resection. The paresis dissolved completely during follow-up. Surgery within the precentral gyrus is of high risk and requires multimodal functional planning. If interpreted with vigilance and consciousness of the underlying physical premises, tractography can provide helpful information within its limitations, which is especially subcortically. However, it may also help in the identification of functional cortex columns of the brain in the presence of a tumor.

In the brain, neuronal cells located in different functional regions communicate through a dense structural network of axons known as the white matter (WM) tissue. Bundles of axons that share similar pathways characterize the WM anatomy, which can be investigated in-vivo thanks to the recent advances of magnetic resonance (MR) techniques. Diffusion MR imaging combined with tractography pipelines allows for a virtual reconstruction of the whole WM anatomy of in-vivo brains, namely the tractogram. It consists of millions of WM fibers as 3D polylines, each approximating thousands of axons. From the analysis of a tractogram, neuroanatomists can characterize well-known white matter structures and detect anatomically non-plausible fibers, which are artifacts of the tractography and often constitute a large portion of it. The accurate characterization of tractograms is pivotal for several clinical and neuroscientific applications. However, such characterization is a complex and time-consuming process that is difficult to be automatized as it requires properly encoding well-known anatomical priors. In this thesis, we propose to investigate the encoding of anatomical priors with a supervised deep learning framework. The ultimate goal is to reduce the presence of artifactual fibers to enable a more accurate automatic process of WM characterization. We devise the problem by distinguishing between volumetric and non-volumetric representations of white matter structures. In the first case, we learn the segmentation of the WM regions that represent relevant anatomical waypoints not yet classified by WM atlases. We investigate using Convolutional Neural Networks (CNNs) to exploit the volumetric representation of such priors. In the second case, the goal is to learn from the 3D polyline representation of fibers where the typical CNN models are not suitable. We introduce the novelty of using Geometric Deep Learning (GDL) models designed to process data having an irregular representation. The working assumption is that the geometrical properties of fibers are informative for the detection of tractogram artifacts. As a first contribution, we present StemSeg that extends the use of CNNs to detect the WM portion representing the waypoints of all the fibers for a specific bundle. This anatomical landmark, called stem, can be critical for extracting that bundle. We provide the results of an empirical analysis focused on the Inferior Fronto-Occipital Fasciculus (IFOF). The effective segmentation of the stem improves the final segmentation of the IFOF, outperforming with a significant gap the reference state of the art. As a second and major contribution, we present Verifyber, a supervised tractogram filtering approach based on GDL, distinguishing between anatomically plausible and non-plausible fibers. The proposed model is designed to learn anatomical features directly from the fiber represented as a 3D points sequence. The extended empirical analysis on healthy and clinical subjects reveals multiple benefits of Verifyber: high filtering accuracy, low inference time, flexibility to different plausibility definitions, and good generalization. Overall, this thesis constitutes a step toward characterizing white matter using deep learning. It provides effective ways of encoding anatomical priors and an original deep learning model designed for fiber.

No description available.

One particular challenge of brain connectomics deals with inferring differences in the brain due to diseases such as Alzheimer's. More specifically, structural connectomics aims at investigating the connectivity between regions in the brain based on the distribution of neuronal fibers. The first step in generating structural connectomes is to perform tractography reconstruction on diffusion MRI (dMRI) data, to extract the most likely pathways of neural fibers. However, current tractography reconstruction algorithms suffer from having high sensitivity and low specificity. Thus, the following steps  of creating, analyzing and deriving graphs metrics from connectivity maps based on tractography impair the reliable assessment of structural connectivity. A promising method to improve tractography and subsequent structural connectomes is to apply tractogram filtering methods. In this study, the impact of tractogram filtering on structural connectomics and derived graph measures of subjects with mild cognitive impairment (MCI), specifically using spherical-deconvolution informed filtering of tractograms (SIFT), is experimentally examined. Moreover, the study also aims at inferring the effects of tractogram filtering in machine-learning based classification of the aforementioned structural connectomes. The pipeline in this experimental setup uses registration tools from FSL, tractography tools from MRTrix3Tissue as well as Keras for classification. The results from the given experiments show, that graph measures such as nodestrength and betweenness centrality are altered for the individual nodes. This leads to new connectomes with nodes, which are more important after tractogram filtering. This effect was also seen in connectomes weighted by fractional anisotropy (FA), mean diffusivity (MD) and radial diffusivity (RD). Moreover, structural connectomes based on filtered tractograms yield a higher classification performance. The best classification performance was reached with 88.65% on raw connectomes. Limiting factors in this experimental setup are identified as the small number of subjects at hand and computation time and the errors introduced by image registration and tractography parameterization.

Diffusion tensor imaging (DTI) is a specialist MRI modality that can identify microstructural changes or abnormalities in the brain. It can also be used to show fibre tract pathways. Both of these features were used in this thesis. Firstly, standard imaging analysis techniques were used to study the effects of mild, repetitive closed head injury on a group of professional boxers. Such data is extremely rare, so the findings of regions of brain abnormalities in the boxers are important, adding to the body of knowledge about more severe traumatic brain injury. The author developed a novel multivariate analysis technique which was used on the same data. This new technique proved to be more sensitive than the standard univariate methods commonly used. An important part of diagnosing and monitoring brain damage involves the use of biomarkers. A novel investigation of whether diffusion parameters obtained from DTI data could serve as bio-markers of cognitive impairment in Parkinson's disease was conducted. This also involved developing a multivariate approach, which displayed increased sensitivity compared with any of the component parameters used singly, and suggested these diffusion measures could be robust bio-markers of cognitive impairment. Fibre tract connectivity between regions of the brain is also a potentially valuable measure for diagnosis and monitoring brain integrity. The feasibility of this was investigated in a multi-modal MRI study. Functional MRI (fMRI) identifies regions of activation associated with a particular task. DTI can then find the pathway of the fibre bundles connecting these regions. The feasibility of using regional connectivity to interrogate brain integrity was investigated using a single healthy volunteer. Fibre pathways between regions activated and deactivated by a working memory paradigm were determined. Though the results are only preliminary, they suggest that this line of research should be continued.

L’imagerie par résonance magnétique de diffusion (IRMd) et de l’IRM fonctionnelle (IRMf) permettent d’explorer la connectivité cérébrale de façon in vivo. Avec l’IRMd, l’architecture du cerveau est inférée en observant la diffusion des molécules d’eau le long des faisceaux de matière blanche. La reconstitution virtuelle de ces fibres est appelée tractographie et représente encore un défi dans ce domaine. Avec l’IRMf, la connectivité fonctionnelle entre deux régions cérébrales est obtenue en examinant la corrélation spatiotemporelle des basses fréquences présentes dans le signal. Effectuer ces analyses sur l’entièreté des voxels du cerveau est très coûteux en termes de temps de calcul et nécessite des connaissances anatomiques précises à chaque individu. Bien qu’il y ait eu d’énormes progrès dans la sophistication des techniques d’imagerie pour traiter les maladies cérébrales, l’infrastructure informatique pour soutenir celles-ci est encore au niveau de l’Âge de pierre, entravant ainsi à leur déploiement en salle d’opération. Il est donc impératif de développer de nouveaux outils informatiques pouvant gérer la complexité de ces données dans un temps efficace. Cette thèse vise à réorienter le paradigme standard d’imagerie cérébrale qui généralise l’information entre individus vers une approche individualisée. Pour ce faire, nous avons 1) quantifié la variabilité présente dans les données d’IRM. Puis, nous avons 2) développé des outils neuro-informatiques permettant d’explorer la connectivité cérébrale au niveau individuel. Ces outils ont permis entre autres 3) d’améliorer la reconstruction virtuelle des radiations optiques, procurant ainsi une information plus complète aux neurochirurgiens. À terme, les méthodes proposées dans ce mémoire fourniront de l’aide aux chirurgiens afin d’améliorer le pronostic d’un patient. / Abstract : Combining diffusion Magnetic Resonance Imaging (dMRI) and functional MRI (fMRI) permits a unique way of exploring brain connectivity in vivo. With dMRI, information about the structural architecture of the brain can be obtained by probing the diffusion of water molecules in and around the white matter (WM) fiber pathways. The process of virtually reconstructing these pathways is called tractography and still represents a difficult challenge in the field. With fMRI, functional connectivity is derived by examining the spatio-temporal correlations in the low frequency bracket of the blood-oxygen-level dependent (BOLD) signal. However, this process can be computationally expensive and requires anatomical knowledge. This thesis aims at shifting the standard brain imaging paradigm of generalizing information across individuals towards a subject-specific approach. Indeed, valuable information is discarded when assuming constant parameters across subjects. From a neurosurgical perspective, capturing the idiosyncrasies of individuals is paramount and requires a highlyspecialized set of mathematical tools. There have been huge advances in the sophistication of brain imaging techniques to treat brain diseases, but computational infrastructure to support the guidance of such treatment has lagged behind, hindering accessibility to their robust deployment. It is therefore imperative to develop a set of new mathematical and computational tools that can handle the complexity of these data in a time efficient manner. Here, applied cutting edge computational methods to improve scientific visualization of brain imaging data in a subject-specific fashion. Ultimately, the methods proposed here will allow surgeons to make a far more informed decision on patient outcome.

Extracting structure from initially incomprehensible streams of events is fundamental to a range of human abilities: from navigating in a new environment to learning a language. These skills rely on our ability to extract spatial and temporal regularities, often with minimal explicit feedback, that is known as statistical learning. Despite the importance of statistical learning for making perceptual decisions, we know surprisingly little about the brain circuits and how they change when learning temporal regularities. In my thesis, I combine behavioural measurements, Diffusion Tensor Imaging (DTI) and resting-state fMRI (rs-fMRI) to investigate the structural and functional circuits that are involved in statistical learning of temporal structures. In particular, I compare structural connectivity as measured by DTI and functional connectivity as measured by rs-fMRI before vs. after training to investigate learning-dependent changes in human brain pathways. Further, I combine the two imaging modalities using graph theory and regression analyses to identify key predictors of individual learning performance. Using a prediction task in the context of sequence learning without explicit feedback, I demonstrate that individuals adapt to the environment’s statistics as they change over time from simple repetition to probabilistic combinations. Importantly, I show that learning of temporal structures relates to decision strategy that varies among individuals between two prototypical distributions: matching the exact sequence statistics or selecting the most probable outcome in a given context (i.e. maximising). Further, combining DTI and rs-fMRI, I show that learning-dependent plasticity in dissociable cortico-striatal circuits relates to decision strategy. In particular, matching relates to connectivity between visual cortex, hippocampus and caudate, while maximisation relates to connectivity between frontal and motor cortices and striatum. These findings have potential translational applications, as alternate brain routes may be re-trained to support learning ability when specific pathways (e.g. memory-related circuits) are compromised by age or disease.

Diffusion of water molecules in biological tissues depends on several microstructural properties. Therefore, diffusion Magnetic Resonance Imaging (dMRI) is a useful tool to infer and study microstructural brain changes in the context of human development, ageing and neuropathology. In this thesis, the state-of-the-art of advanced dMRI techniques is explored and strategies to overcome or reduce its pitfalls are developed and validated. Firstly, it is shown that PCA denoising and Gibbs artefact suppression algorithms provide an optimal compromise between increased precision of diffusion measures and the loss of tissue's diffusion non-Gaussian information. Secondly, the spatial information provided by the diffusion kurtosis imaging (DKI) technique is explored and used to resolve crossing fibres and generalize diffusion measures to cases not limited to well-aligned white matter fibres. Thirdly, as an alternative to diffusion microstructural modelling techniques such as the neurite orientation dispersion and density imaging (NODDI), it is shown that spherical deconvolution techniques can be used to characterize fibre crossing and dispersion simultaneously. Fourthly, free water volume fraction estimates provided by the free water diffusion tensor imaging (fwDTI) are shown to be useful to detect and remove voxels corrupted by cerebrospinal fluid (CSF) partial volume effects. Finally, dMRI techniques are applied to the diffusion data from the large collaborative Cambridge Centre for Ageing and Neuroscience (CamCAN) study. From these data, the inference provided by diffusion anisotropy measures on maturation and degeneration processes is shown to be biased by age-related changes of fibre organization. Inconsistencies of previous NODDI ageing studies are also revealed to be associated with the different age ranges covered. The CamCAN data is also processed using a novel non-Gaussian diffusion characterization technique which is invariant to different fibre configurations. Results show that this technique can provide indices specific to axonal water fraction which can be linked to age-related fibre density changes.

Le système visuel humain est complexe par son organisation anatomique et par son fonctionnement incomplètement élucidé. Il est fonctionnellement divisé en deux systèmes. Le premier système est destiné à la vision consciente communément appelée voie visuelle principale ou en anglais « image forming visual pathways ». Le second, appelé système secondaire ou accessoire, n’apporte pas d’information visuelle consciente, il est dit « non image forming visual pathway ». Ce dernier apporte à notre cerveau une information sur l’environnement telle que la sensation jour/nuit. Ses fonctions sont sous-tendues par l’afflux d’informations rétiniennes non visuelles sur des structures de l’hypothalamus comme le noyau supra-chiasmatique. Les deux systèmes visuels ont un substratum anatomique complexe faisant intervenir de nombreuses structures anatomiques au sein des différents étages du cerveau cortical et sous-cortical comme les noyaux gris centraux dits « Basal Ganglias » (BG). Le système visuel secondaire intervient aussi comme une structure de contrôle des mouvements oculomoteurs tels que la poursuite ou les saccades nécessaires pour explorer notre environnement. Ainsi les saccades oculomotrices sont sous le contrôle modulateur des BG. De ce fait l’étude des saccades apparait comme un très bon modèle pour explorer le fonctionnement du système extrapyramidal au cours des maladies neuro-dégénératives. Les connaissances actuelles sur ce système de contrôle des saccades proviennent essentiellement des études sur le primate non humain et sur des observations cliniques chez l’homme au cours de pathologies dégénératives ou toxiques des BG. L’observation des structures anatomiques, en particulier du réseau de la substance blanche cérébrale qui supporte les connections axonales, n’est pas accessible à l’imagerie clinique de routine. Pour décrire et étudier ces réseaux de connections, la notion de connectomique a été introduite il y a un dizaine d’années. Dans ce travail, nous nous sommes donné l’objectif de décrire le connectome des saccades oculomotrices sur un plan structurel. Nous avons exploré les structures sous-corticales intervenant dans le contrôle des saccades comme les BG, le colliculus supérieur et le pulvinar. Pour ce faire, nous avons utilisé l’imagerie IRM structurelle en diffuseur de tension (DTI) chez deux groupes de patients présentant une maladie neuro-dégénérative : un groupe souffrant de maladie de Parkinson chez qui une atteinte des BG et une dysfonction des saccades sont reconnues, et un groupe de trembleurs essentiels reconnu pour ne pas présenter de dysfonction des saccades et chez qui les BG sont épargnés. Le résultat de ce travail a permis pour la première fois une description in vivo du connectome des saccades chez l’Homme. Il a de plus montré des différences dans la structure du connectome dans les deux groupes de patients. Une meilleure connaissance de ce connectome pourrait permettre de mieux comprendre certains troubles oculomoteurs et aussi de suivre l’évolution de certaines maladies neurodegeneratives. / Visual system is complex by its anatomy and its function. Neuro-anatomists have been interested in understanding the link between the visual pathways and the brain for centuries. Classical brain fixation and dissection methods were used to describe the visual pathways identifiable macroscopically. Non–image visual pathway, particularly the part involves in saccadic eye movements network in human is still not mastered. Our current knowledge in SCM is based on animal studies, anatomic dissection and brain histopathology examination of specimens from patients with clinical basal ganglia (BG) disorders. Saccadic eye movements (SCM) are under the control of the basal ganglia (BG) and SCM circuitry within the BG represents a good model for studying pathology in the extra-pyramidal system. The diagnosis of Parkinson’s disease (PD), which affects SEM and its distinction from non-dopaminergic, essential tremor (ET) where SEM are not impaired can be challenging and still relies on clinical observations. Diffusion tensor imaging and fiber tractography (DTI-FT), a new MRI technology, can be used to evaluate the presence and integrity of white matter tracts using directional diffusion patterns of water. The purpose of this study is to use DTI-FT to analyse SEM networks within BG and compare the SEM neural pathways or connectome of patients clinically diagnosed with PD and ET. To date, there are no studies, using DTI-FT for the extensive exploration of non-image visual pathways and SCM circuits, notably the deep brain connections. For this goal, we introduced the concept of SCM connectomes, derived from the general concept of connectome. Our study used structural MRI to identify nuclei and fascicles of the SCM connectome in PD and ET patients; imageries were acquired in routine clinical conditions fitted for DBS surgery. We found a reduction of the fiber number in two fascicles of the connectome in PDcompared to ET group.

Der menschliche inferiore Parietallappen (IPC) gehört zum Assoziationskortex und spielt eine wichtige Rolle bei der Integration von somatosensorischen (taktilen), visuellen und akustischen Reizen. Bisher gibt es keine eindeutigen Informationen über den strukturellen Aufbau dieser Hirnregion. Parzellierungen anhand der Zytoarchitektur reichen von zwei (Brodmann 1909) bis sieben Subareale (Caspers et al. 2006). Homologien zwischen dem IPC des Menschen und Makaken-Affen sind weitestgehend unbekannt. In der vorliegenden Arbeit wurden der Aufbau und die Konnektivitäten des menschlichen IPC genauer untersucht. Dazu führte man eine konnektivitätsbasierte Parzellierung des IPC an 20 Probanden durch. Als Methode kam Diffusions-Tensor-Imaging (DTI) kombiniert mit probabilistischer Traktogra-phie zum Einsatz. Der IPC konnte anhand der Konnektivitäten in drei Subareale (IPCa, IPCm, IPCp) parzelliert werden. Diese besitzen in beiden Hemisphären eine ähnliche Größe und eine rostro-kaudale Anordnung. Die Parzellierung ist vergleichbar mit der des Makaken-IPC, bei dem ebenfalls eine Unterteilung in drei Areale (PF, PFG, PG) und eine rostro-kaudale Anordnung nachgewiesen werden konnte. Jedes Subareal des menschlichen IPC besitzt ein individuelles Konnektivitätsmuster. Beim Menschen als auch beim Makaken gibt es starke Verbindungen zum lateralen prämotorischen Kortex und zum superioren Parietallappen. Diese Gemeinsamkeiten lassen darauf schließen, dass strukturelle Eigenschaften im Laufe der Evolution erhalten geblieben sind. Allerdings sind beim Menschen auch Neuentwicklungen nachweisbar. Dazu gehören die deutlich hervortretenden Verbindungen zum Temporallappen. Möglicherweise haben sich diese erst während der Evolution entwickelt und sind beim Menschen als Teil des perisylvischen Sprachnetzwerkes an der Sprachbildung beteiligt. / The human inferior parietal cortex convexity (IPCC) is an important association area, which integrates auditory, visual and somatosensory information. However, the structural organization of the IPCC is a controversial issue. For example, cytoarchitectonic parcellations reported in the literature range from two to seven areas. Moreover, anatomical descriptions of the human IPCC are often based on experiments in the macaque monkey. In this study we used diffusion-weighted magnetic resonance imaging (dMRI) combined with probabilistic tractography to quantify the connectivity of the human IPCC, and used this information to parcellate this cortex area. This provides a new structural map of the human IPCC, comprising three sub-areas (IPCa, IPCm, IPCp) of comparable size, in a rostro-caudal arrangement in the left and right hemisphere. Each sub-area is characterized by a connectivity fingerprint and the parcellation is similar to the subdivision reported for the macaque IPCC (rostro-caudal areas areas PF, PFG, and PG). However, the present study also reliably demonstrates new structural features in the connectivity pattern of the human IPCC, which are not known to exist in the macaque. This study quantifies inter-subject variability by providing a population representation of the sub-area arrangement, and demonstrates substantial lateralization of the connectivity patterns of IPCC.