219P Benchmarking state-of-the-art deep learning models for tumor-infiltrating lymphocytes (TILs) quantification

Résumé fourni par la source

TILs provide both prognostic information and predictive value for immunotherapy response. Their objective assessment has commonly relied on classical machine learning or convolutional neural networks (CNNs). More recently, vision transformer-based architectures have shown promising performance, but their efficiency and scalability remain limited due to the high computational demand of processing millions of cells.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.