La segmentation d'image est une opération de s consistant à détecter et rassembler les pixels suivant des critères, notamment d'intensité ou spatiaux, l'image apparaissant ainsi formée de régions uniformes. La segmentation peut par exemple montrer les objets en les distinguant du fond avec netteté. Dans les cas où les critères divisent les pixels en deux ensembles, le traitement est une binarisation. Des algorithmes sont écrits comme substitut aux connaissances de haut niveau que l'homme mobilise dans son identification des objets et structures. Des algorithmes de segmentation de haut niveau (chaque région est un objet sémantique) sont en développement en divers domaines concernés par le traitement d'images. La segmentation est une étape primordiale en traitement d'image. Les nombreuses méthodes de segmentation se répartissent en quatre principales classes :

La segmentation fondée sur les régions (en anglais : region-based segmentation). On y trouve par exemple : la croissance de

Graph Representation Learning in Computational Pathology

Guillaume Jaume

Advances in scanning systems have enabled the digitization of pathology slides into Whole-Slide Images (WSIs), opening up opportunities to develop Computational Pathology (CompPath) methods for computer-aided cancer diagnosis and prognosis. CompPath has been primarily developed using models based on Convolutional Neural Networks (CNNs), building on the recent successes in Computer Vision. A series of promising approaches have been proposed for nuclei segmentation and classification, tumor detection, tumor grading, among others. However, CNN-based methods suffer from several limitations. First, it is challenging to model both fine-grained nuclei-level information and long-range inter-glandular dependencies. Second, there is a discrepancy between the pixel-based analysis of CNNs and the histological entity-centered analysis employed for pathological diagnosis, which in turn can hinder model transparency. Third, the inherent complexity of training networks on large histology images with limited annotations constrains its learning capabilities.Instead, we propose an analytical paradigm shift, where we view and analyze histology images as a set of biological entities interacting with each other. Specifically, an image is represented as an entity-graph where nodes depict biological entities and edges encode interactions between these entities. Entity-graphs are further processed by a Graph Neural Network (GNN) model to jointly encode the entity morphological attributes and topological distribution, towards tissue phenotyping. In this thesis, we study three research directions in CompPath, namely, scalability, interpretability and explainability, and weakly-supervised learning.First, histology images are orders of magnitude larger than natural images, where diagnostically relevant regions can represent only a fraction of the image. We propose a scalable hierarchical cell-to-tissue representation (HACT) and GNN model, HACT-Net, for learning on arbitrary large inputs. We show the capabilities of HACT-Net on our proposed BR ACS dataset, the largest cohort to date for breast tumor Regions-of-Interest subtyping.Second, computer-aided diagnostic tools must be transparent and their decision-making process justified. By shifting the analysis from pixel- to entity-based, we make the input space interpretable for pathologists that can better relate to the model input. We further propose entity-centric graph explainers, exemplified with cell-graph model explainability, along with novel metrics to evaluate explanations based on entity-level pathological concepts.Third, acquiring ground-truth data to train deep learning systems requires pathologists to provide specific annotations, which is time-consuming, expensive, and subject to inter- and intra-observer variability. We therefore propose WholeSIGHT, a method that reduces annotation requirements to WSI-level labels only, for joint classification and segmentation of WSIs. We show the capabilities of WholeSIGHT for Gleason pattern segmentation and grading on multi-source WSI prostate datasets. The generalization properties of WholeSIGHT are further evaluated on unseen cohorts, and compared to Bayesian variants to strengthen the estimation of the model uncertainty. Finally, we introduce HistoCartography, a novel python library designed to accelerate the development of graph analytics in CompPath.

EPFL2022

Video segmentation based on multiple features for interactive and automatic multimedia applications

In this thesis, a novel method for the segmentation of video sequences based on the analysis of multiple image features is presented. A key feature of the system is the distinction between two levels of segmentation, namely region and object segmentation. Regions are homogeneous areas of the images, which are extracted automatically by the computer. Semantically meaningful objects are obtained by grouping regions, automatically or through user interaction, according to the specific application. This splitting relieves the computer of ill-posed semantic problems, and allows a higher level of flexibility in the use of the results. The extraction of the regions is based the multidimensional analysis of several image features by a spatially constrained Fuzzy C-Means algorithm. The relative weighting of the different features is achieved by means of an adaptive system that takes into account the local level of reliability of each feature. The temporal tracking of the obtained regions is performed by means of a dual strategy in which the motion-compensated projection of the segmentation mask from previous frames is used to influence the segmentation of the current frame so as to achieve higher temporal coherence and stability.

EPFL1998

Graph Representation Learning in Computational Pathology

Guillaume Jaume

EPFL2022

Segmentation d'image

La segmentation fondée sur les régions (en anglais : region-based segmentation). On y trouve par exemple : la croissance de

Graph Representation Learning in Computational Pathology

Video segmentation based on multiple features for interactive and automatic multimedia applications

Video segmentation based on multiple features for interactive and automatic multimedia applications

Graph Representation Learning in Computational Pathology