AUTHOR=Naylor Peter , Lazard Tristan , Bataillon Guillaume , Laé Marick , Vincent-Salomon Anne , Hamy Anne-Sophie , Reyal Fabien , Walter Thomas TITLE=Prediction of Treatment Response in Triple Negative Breast Cancer From Whole Slide Images JOURNAL=Frontiers in Signal Processing VOLUME=2 YEAR=2022 URL=https://www.frontiersin.org/journals/signal-processing/articles/10.3389/frsip.2022.851809 DOI=10.3389/frsip.2022.851809 ISSN=2673-8198 ABSTRACT=

The automatic analysis of stained histological sections is becoming increasingly popular. Deep Learning is today the method of choice for the computational analysis of such data, and has shown spectacular results for large datasets for a large variety of cancer types and prediction tasks. On the other hand, many scientific questions relate to small, highly specific cohorts. Such cohorts pose serious challenges for Deep Learning, typically trained on large datasets. In this article, we propose a modification of the standard nested cross-validation procedure for hyperparameter tuning and model selection, dedicated to the analysis of small cohorts. We also propose a new architecture for the particularly challenging question of treatment prediction, and apply this workflow to the prediction of response to neoadjuvant chemotherapy for Triple Negative Breast Cancer.