A Neurodegenerative Disease (ND) is progressive damage to brain neurons, which the human body cannot repair or replace. The well-known examples of such conditions are Dementia and Alzheimer’s Disease (AD), which affect millions of lives each year. Although conducting numerous researches, there are no effective treatments for the mentioned diseases today. However, early diagnosis is crucial in disease management. Diagnosing NDs is challenging for neurologists and requires years of training and experience. So, there has been a trend to harness the power of deep learning, including state-of-the-art Convolutional Neural Network (CNN), to assist doctors in diagnosing such conditions using brain scans. The CNN models lead to promising results comparable to experienced neurologists in their diagnosis. But, the advent of transformers in the Natural Language Processing (NLP) domain and their outstanding performance persuaded Computer Vision (CV) researchers to adapt them to solve various CV tasks in multiple areas, including the medical field. This research aims to develop Vision Transformer (ViT) models using Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset to classify NDs. More specifically, the models can classify three categories (Cognitively Normal (CN), Mild Cognitive Impairment (MCI), Alzheimer’s Disease (AD)) using brain Fluorodeoxyglucose (18F-FDG) Positron Emission Tomography (PET) scans. Also, we take advantage of Automated Anatomical Labeling (AAL) brain atlas and attention maps to develop explainable models. We propose three ViTs, the best of which obtains an accuracy of 82% on the test dataset with the help of transfer learning. Also, we encode the AAL brain atlas information into the best performing ViT, so the model outputs the predicted label, the most critical region in its prediction, and overlaid attention map on the input scan with the crucial areas highlighted. Furthermore, we develop two CNN models with 2D and 3D convolutional kernels as baselines to classify NDs, which achieve accuracy of 77% and 73%, respectively, on the test dataset. We also conduct a study to find out the importance of brain regions and their combinations in classifying NDs using ViTs and the AAL brain atlas. / <p>This thesis was awarded a prize of 50,000 SEK by Getinge Sterilization for projects within Health Innovation.</p>

Ces dernières années, le domaine de l’apprentissage profond a connu des progrès énormes dans des applications allant de la génération d’images, détection d’objets, modélisation du langage à la réponse aux questions visuelles. Les approches classiques telles que l’apprentissage supervisé nécessitent de grandes quantités de données étiquetées et spécifiques à la tâches. Cependant, celles-ci sont parfois coûteuses, peu pratiques, ou trop longues à collecter. La modélisation efficace en données, qui comprend des techniques comme l’apprentissage few-shot (à partir de peu d’exemples) et l’apprentissage self-supervised (auto-supervisé), tentent de remédier au manque de données spécifiques à la tâche en exploitant de grandes quantités de données plus “générales”. Les progrès de l’apprentissage profond, et en particulier de l’apprentissage few-shot, s’appuient sur les benchmarks (suites d’évaluation), les métriques d’évaluation et les jeux de données, car ceux-ci sont utilisés pour tester et départager différentes méthodes sur des tâches précises, et identifier l’état de l’art. Cependant, du fait qu’il s’agit de versions idéalisées de la tâche à résoudre, les benchmarks sont rarement équivalents à la tâche originelle, et peuvent avoir plusieurs limitations qui entravent leur rôle de sélection des directions de recherche les plus prometteuses. De plus, la définition de métriques d’évaluation pertinentes peut être difficile, en particulier dans le cas de sorties structurées et en haute dimension, telles que des images, de l’audio, de la parole ou encore du texte. Cette thèse discute des limites et des perspectives des benchmarks existants, des fonctions de coût (training losses) et des métriques d’évaluation (evaluation metrics), en mettant l’accent sur la modélisation générative - les Réseaux Antagonistes Génératifs (GANs) en particulier - et la modélisation efficace des données, qui comprend l’apprentissage few-shot et self-supervised. La première contribution est une discussion de la tâche de modélisation générative, suivie d’une exploration des propriétés théoriques et empiriques des fonctions de coût des GANs. La deuxième contribution est une discussion sur la limitation des few-shot classification benchmarks, certains ne nécessitant pas de généralisation à de nouvelles sémantiques de classe pour être résolus, et la proposition d’une méthode de base pour les résoudre sans étiquettes en phase de testing. La troisième contribution est une revue sur les méthodes few-shot et self-supervised de détection d’objets , qui souligne les limites et directions de recherche prometteuses. Enfin, la quatrième contribution est une méthode efficace en données pour la description de vidéo qui exploite des jeux de données texte et vidéo non supervisés. / In recent years, the field of deep learning has seen tremendous progress for applications ranging from image generation, object detection, language modeling, to visual question answering. Classic approaches such as supervised learning require large amounts of task-specific and labeled data, which may be too expensive, time-consuming, or impractical to collect. Data-efficient methods, such as few-shot and self-supervised learning, attempt to deal with the limited availability of task-specific data by leveraging large amounts of general data. Progress in deep learning, and in particular, few-shot learning, is largely driven by the relevant benchmarks, evaluation metrics, and datasets. They are used to test and compare different methods on a given task, and determine the state-of-the-art. However, due to being idealized versions of the task to solve, benchmarks are rarely equivalent to the original task, and can have several limitations which hinder their role of identifying the most promising research directions. Moreover, defining meaningful evaluation metrics can be challenging, especially in the case of high-dimensional and structured outputs, such as images, audio, speech, or text. This thesis discusses the limitations and perspectives of existing benchmarks, training losses, and evaluation metrics, with a focus on generative modeling—Generative Adversarial Networks (GANs) in particular—and data-efficient modeling, which includes few-shot and self-supervised learning. The first contribution is a discussion of the generative modeling task, followed by an exploration of theoretical and empirical properties of the GAN loss. The second contribution is a discussion of a limitation of few-shot classification benchmarks, which is that they may not require class semantic generalization to be solved, and the proposal of a baseline method for solving them without test-time labels. The third contribution is a survey of few-shot and self-supervised object detection, which points out the limitations and promising future research for the field. Finally, the fourth contribution is a data-efficient method for video captioning, which leverages unsupervised text and video datasets, and explores several multimodal pretraining strategies.