Segmentation et Annotation d’Images Histologiques pour l’Identification Morphologique et Fonctionnelle des Cellules par IA

Abstract

Histological image analysis is essential in biomedical research and digital pathology, but manual evaluation remains time-consuming and subjective. This final-year project focuses on the annotation and segmentation of histological images of brain sections from healthy rats and rats affected by Alzheimer’s disease in order to automate the identification of cell nuclei and extract relevant morphometric features. The methodology combines QuPath for annotation and cell detection, CIAseg for AI-assisted segmentation, and Python with Visual Studio Code for data processing and analysis. The obtained results enabled the extraction of quantitative parameters such as area, perimeter, circularity, and staining intensity, highlighting the potential of artificial intelligence to improve the speed, objectivity, and reproducibility of histological analysis and to support future computer-aided diagnostic applications.