Attention deficit hyperactivity disorder is one of the most common neurodevelopmental disorders, affecting around 7% of the worldwide population in their everyday life. It has been suggested that individuals with ADHD differ in gray matter volume from typically developing controls. However, findings on in which brain areas these differences are located, as well as how gray matter volume is affected by stimulant medication, remain inconclusive. Therefore, this systematic review aimed to investigate any potential differences in gray matter volume in medication-naïve individuals with ADHD compared to controls, focusing on studies using voxel-based morphometry applied to MRI-imaging data A keyword search in the databases Web of Science, Scopus and Medline EBSCO resulted in 349 studies, of which seven met the inclusion criteria and were included in the review. The results included a total of 169 participants diagnosed with ADHD and 148 typically developing controls. Findings suggested decreased cerebellar gray matter volume, potential gender-wise volume differences in the anterior cingulate cortex, and a decrease in caudate gray matter volume, specifically in adults with ADHD. Some limitations include small sample sizes, possible effects of age on gray matter volume, and the overall heterogeneous nature of the disorder. The present review agrees that individuals with ADHD exhibit differences in gray matter volume, but also highlights the importance of expanding research on medication-naïve subjects, to be able to draw more robust scientific conclusions about the neural correlates of ADHD in the future.

Recently, Attention Deficit Hyperactive Disorder (ADHD) is getting a lot of attention mainly for two reasons. First, it is one of the most commonly found childhood behavioral disorders. Around 5-10% of the children all over the world are diagnosed with ADHD. Second, the root cause of the problem is still unknown and therefore no biological measure exists to diagnose ADHD. Instead, doctors need to diagnose it based on the clinical symptoms, such as inattention, impulsivity and hyperactivity, which are all subjective. Functional Magnetic Resonance Imaging (fMRI) data has become a popular tool to understand the functioning of the brain such as identifying the brain regions responsible for different cognitive tasks or analyzing the statistical differences of the brain functioning between the diseased and control subjects. ADHD is also being studied using the fMRI data. In this dissertation we aim to solve the problem of automatic diagnosis of the ADHD subjects using their resting state fMRI (rs-fMRI) data. As a core step of our approach, we model the functions of a brain as a connectivity network, which is expected to capture the information about how synchronous different brain regions are in terms of their functional activities. The network is constructed by representing different brain regions as the nodes where any two nodes of the network are connected by an edge if the correlation of the activity patterns of the two nodes is higher than some threshold. The brain regions, represented as the nodes of the network, can be selected at different granularities e.g. single voxels or cluster of functionally homogeneous voxels. The topological differences of the constructed networks of the ADHD and control group of subjects are then exploited in the classification approach. We have developed a simple method employing the Bag-of-Words (BoW) framework for the classification of the ADHD subjects. We represent each node in the network by a 4-D feature vector: node degree and 3-D location. The 4-D vectors of all the network nodes of the training data are then grouped in a number of clusters using K-means; where each such cluster is termed as a word. Finally, each subject is represented by a histogram (bag) of such words. The Support Vector Machine (SVM) classifier is used for the detection of the ADHD subjects using their histogram representation. The method is able to achieve 64% classification accuracy. The above simple approach has several shortcomings. First, there is a loss of spatial information while constructing the histogram because it only counts the occurrences of words ignoring the spatial positions. Second, features from the whole brain are used for classification, but some of the brain regions may not contain any useful information and may only increase the feature dimensions and noise of the system. Third, in our study we used only one network feature, the degree of a node which measures the connectivity of the node, while other complex network features may be useful for solving the proposed problem. In order to address the above shortcomings, we hypothesize that only a subset of the nodes of the network possesses important information for the classification of the ADHD subjects. To identify the important nodes of the network we have developed a novel algorithm. The algorithm generates different random subset of nodes each time extracting the features from a subset to compute the feature vector and perform classification. The subsets are then ranked based on the classification accuracy and the occurrences of each node in the top ranked subsets are measured. Our algorithm selects the highly occurring nodes for the final classification. Furthermore, along with the node degree, we employ three more node features: network cycles, the varying distance degree and the edge weight sum. We concatenate the features of the selected nodes in a fixed order to preserve the relative spatial information. Experimental validation suggests that the use of the features from the nodes selected using our algorithm indeed help to improve the classification accuracy. Also, our finding is in concordance with the existing literature as the brain regions identified by our algorithms are independently found by many other studies on the ADHD. We achieved a classification accuracy of 69.59% using this approach. However, since this method represents each voxel as a node of the network which makes the number of nodes of the network several thousands. As a result, the network construction step becomes computationally very expensive. Another limitation of the approach is that the network features, which are computed for each node of the network, captures only the local structures while ignore the global structure of the network. Next, in order to capture the global structure of the networks, we use the Multi-Dimensional Scaling (MDS) technique to project all the subjects from an unknown network-space to a low dimensional space based on their inter-network distance measures. For the purpose of computing distance between two networks, we represent each node by a set of attributes such as the node degree, the average power, the physical location, the neighbor node degrees, and the average powers of the neighbor nodes. The nodes of the two networks are then mapped in such a way that for all pair of nodes, the sum of the attribute distances, which is the inter-network distance, is minimized. To reduce the network computation cost, we enforce that the maximum relevant information is preserved with minimum redundancy. To achieve this, the nodes of the network are constructed with clusters of highly active voxels while the activity levels of the voxels are measured based on the average power of their corresponding fMRI time-series. Our method shows promise as we achieve impressive classification accuracies (73.55%) on the ADHD-200 data set. Our results also reveal that the detection rates are higher when classification is performed separately on the male and female groups of subjects. So far, we have only used the fMRI data for solving the ADHD diagnosis problem. Finally, we investigated the answers of the following questions. Do the structural brain images contain useful information related to the ADHD diagnosis problem? Can the classification accuracy of the automatic diagnosis system be improved combining the information of the structural and functional brain data? Towards that end, we developed a new method to combine the information of structural and functional brain images in a late fusion framework. For structural data we input the gray matter (GM) brain images to a Convolutional Neural Network (CNN). The output of the CNN is a feature vector per subject which is used to train the SVM classifier. For the functional data we compute the average power of each voxel based on its fMRI time series. The average power of the fMRI time series of a voxel measures the activity level of the voxel. We found significant differences in the voxel power distribution patterns of the ADHD and control groups of subjects. The Local binary pattern (LBP) texture feature is used on the voxel power map to capture these differences. We achieved 74.23% accuracy using GM features, 77.30% using LBP features and 79.14% using combined information. In summary this dissertation demonstrated that the structural and functional brain imaging data are useful for the automatic detection of the ADHD subjects as we achieve impressive classification accuracies on the ADHD-200 data set. Our study also helps to identify the brain regions which are useful for ADHD subject classification. These findings can help in understanding the pathophysiology of the problem. Finally, we expect that our approaches will contribute towards the development of a biological measure for the diagnosis of the ADHD subjects.

L’épilepsie insulaire (ÉI) est une forme rare d’épilepsie focale qui, en raison des défis liés à son diagnostic, est difficilement cernable. De plus, la prise en charge des patients avec ÉI s’avère complexifiée par le fait que cette pathologie est fréquemment résistante aux médicaments anti-crises. Pour ces cas médico-réfractaires, la chirurgie insulaire est une option viable. Cela dit, les patients subissant une telle intervention développent fréquemment des déficits neurologiques postopératoires; heureusement, la grande majorité de ceux-ci récupèrent complètement et rapidement. Or, le mécanisme sous-tendant ce singulier rétablissement fonctionnel demeure à ce jour mal compris. Deux modalités modernes d’IRM structurelle, soit l’analyse d’épaisseur corticale et la tractographie, ont permis, dans les dernières années, de décrire les altérations architecturales caractéristiques et potentiellement diagnostiques de divers types d’épilepsie ainsi que de caractériser les remodelages plastiques qui suivent la chirurgie de l’épilepsie extra-insulaire. Cependant, à ce jour, aucune étude ne s’est encore penchée sur le cas de l’ÉI. De ce fait, les études qui constituent cette thèse exploitent l’IRM structurelle afin, d’une part, de dépeindre les altérations d’épaisseur du cortex et de connectivité de matière blanche associées à l’ÉI et, d’autre part, de définir les réarrangements de connectivité subséquents à la chirurgie insulaire pour contrôle épileptique. Les deux premières études de cette thèse ont révélé que l’ÉI était associée à un pattern majoritairement ipsilatéral d’atrophie corticale et d’hyperconnectivité impliquant principalement des sous-régions insulaires et des régions connectées à l’insula. De manière intéressante, la topologie de ces changements correspondait, au moins en partie, à celle du réseau épileptique de l’ÉI. Ensuite, la troisième étude visait à décrire, par le biais d’une méta-analyse, l’histoire naturelle postopératoire des patients subissant une chirurgie pour ÉI. Cette analyse a, entre autres, confirmé que cette chirurgie était efficace (66.7% de disparition des crises) et qu’elle était fréquemment accompagnée de complications neurologiques (42.5%) qui, dans la plupart des cas, étaient transitoires (78.7% des complications) et récupéraient entièrement dans les trois mois postopératoires (91.6% des complications transitoires). Finalement, la quatrième étude a révélé que la chirurgie pour ÉI était suivie d’altérations de connectivité diffuses et bilatérales. Notamment, les connexions présentant une augmentation de connectivité concernaient particulièrement des régions localisées soit près de la cavité chirurgicale ou dans l’hémisphère controlatéral à l’intervention. De plus, la majorité de ces renforcements structurels se sont produits dans les six premiers mois suivant la chirurgie, un délai comparable à celui durant lequel la majeure partie de la récupération fonctionnelle postopératoire a été observée dans notre méta-analyse. En somme, nos résultats suggèrent que les altérations morphologiques en lien avec l’ÉI peuvent correspondre à son réseau épileptique sous-jacent. La topologie de ces changements pourrait constituer un biomarqueur structurel diagnostique qui aiderait à la reconnaissance de l’ÉI et, concomitamment, favoriserait possiblement un traitement chirurgical plus adapté et plus efficace. De plus, les augmentations de connectivité postopératoires pourraient correspondre à des réponses neuroplastiques permettant de prendre en charge les fonctions altérées par la chirurgie. Nos constats ont ainsi contribué à la caractérisation des mécanismes étayant la singulière récupération fonctionnelle accompagnant la chirurgie pour ÉI. À plus grande échelle, nos travaux offrent un aperçu du potentiel de l’IRM structurelle à assister au diagnostic de l’épilepsie focale ainsi qu’à participer à la description des changements plastiques subséquents à une résection neurochirurgicale. / Insular epilepsy (IE) is a rare type of focal epilepsy that is difficult to diagnose. In addition to the challenging nature of IE detection, management of patients with this condition is complicated by the tendency of insular seizures to be resistant to anti-seizure medications. For such medically refractory cases, insular surgery constitutes a viable and long-lasting therapeutic option. That said, patients who undergo an insular resection for seizure control frequently develop postoperative neurological deficits; fortunately, most of these impairments recover fully and rapidly. While this favorable postoperative course contributes to improving the outcome of IE surgery, the mechanism underlying the functional recovery remains unknown. Two contemporary structural MRI modalities, namely cortical thickness analysis and tractography, have recently been used to describe characteristic structural alterations of focal epilepsies and to elucidate the postoperative plastic remodeling associated with surgery for extra-insular epilepsy. While these analyses added to our understanding of several localization-related epilepsies, none specifically studied IE. In this thesis, we exploit structural MRI techniques to, first, depict the alterations of cortical thickness and white matter connectivity in IE and, second, define the progressive rearrangements that follow insular surgery for epilepsy. The first two studies of the current thesis showed that IE is associated with a primarily ipsilateral pattern of cortical thinning and hyperconnectivity that mainly involves insular subregions and insula-connected regions. Interestingly, the topology of these changes corresponded, at least in part, to the epileptic network of IE. Furthermore, the third study aimed to describe, via a meta-analysis, the postoperative outcome of patients undergoing surgery for IE. Among other findings, the analysis revealed that insular surgery was effective (66.7% seizure freedom rate) but was associated with a significant risk of neurological complications (42.5%) which, in most cases, were transient (78.7% of all complications) and recovered fully within three months (91.6% of transient complications). Finally, the fourth study showed that surgery for IE was followed by a diffuse pattern of bilateral structural connectivity changes. Notably, connections exhibiting an increase in connectivity were specifically located near the surgical cavity and in the contralateral healthy hemisphere. In addition, the majority of the structural strengthening occurred in the first six months following surgery, a time course that is consistent with the short delay during which most of the postoperative functional recovery was observed in our meta-analysis. Our results suggest that the morphological alterations in IE may reflect its underlying epileptic network. The topology of these changes may constitute a structural biomarker that could help diagnose IE more readily and, concomitantly, potentially enable a more targeted and more effective surgical treatment. Moreover, the postoperative increases in connectivity may be compatible with compensatory neuroplastic responses, a process that arose to recoup the functions of the injured insular cortex. Our findings have therefore contributed to the characterization of the driving process that supports the striking functional recovery seen following surgery for IE. On a larger scale, our work provides insights into the potential of structural MRI to assist in the diagnosis of focal epilepsy and to describe plastic changes following neurosurgical resections.

Spike timing-dependent plasticity (STDP) has been suggested as one of the key mechanism underlying learning and memory. Due to its importance, timing-dependent plasticity studies have been approached in the living human brain by means of non-invasive brain stimulation (NIBS) protocols such as paired associative stimulation (PAS). However, contrary to STDP studies at a cellular level, functional plasticity induction in the human brain implies the interaction among target cortical networks and investigates plasticity mechanisms at a systems level. This thesis comprises of two independent studies that aim at understanding the importance of considering broad cortical networks when predicting the outcome of timing-dependent associative plasticity induction in the human brain. In the first study we developed a new protocol (ipsilateral PAS (ipsiPAS)) that required timing- and regional-specific information transfer across hemispheres for the induction of timing-dependent plasticity within M1 (see chapter 3). In the second study, we tested the influence of individual brain structure, as measured with voxel-based cortical thickness, on a standard PAS protocol (see chapter 4). In summary, we observed that the near-synchronous associativity taking place within M1 is not the only determinant influencing the outcome of PAS protocols. Rather, the online interaction of the cortical networks integrating information during a PAS intervention determines the outcome of the pairing of inputs in M1.

Spike timing-dependent plasticity (STDP) has been suggested as one of the key mechanism underlying learning and memory. Due to its importance, timing-dependent plasticity studies have been approached in the living human brain by means of non-invasive brain stimulation (NIBS) protocols such as paired associative stimulation (PAS). However, contrary to STDP studies at a cellular level, functional plasticity induction in the human brain implies the interaction among target cortical networks and investigates plasticity mechanisms at a systems level. This thesis comprises of two independent studies that aim at understanding the importance of considering broad cortical networks when predicting the outcome of timing-dependent associative plasticity induction in the human brain. In the first study we developed a new protocol (ipsilateral PAS (ipsiPAS)) that required timing- and regional-specific information transfer across hemispheres for the induction of timing-dependent plasticity within M1 (see chapter 3). In the second study, we tested the influence of individual brain structure, as measured with voxel-based cortical thickness, on a standard PAS protocol (see chapter 4). In summary, we observed that the near-synchronous associativity taking place within M1 is not the only determinant influencing the outcome of PAS protocols. Rather, the online interaction of the cortical networks integrating information during a PAS intervention determines the outcome of the pairing of inputs in M1.