Predicting Residual Cancer Burden in a triple negative breast cancer cohort

Peter Naylor (1) , Joseph Boyd (1) , Marick Lae (2) , Fabien Reyal (3) , Thomas Walter (1, 4, 5)

1 CBIO - Centre de Bioinformatique
2 U932 - Immunité et cancer
3 Institut Curie [Paris]
4 Cancer et génome: Bioinformatique, biostatistiques et épidémiologie d'un système complexe
5 PSL - Université Paris sciences et lettres

Analysis and interpretation of stained histopathology sections is one of the main tools in cancer diagnosis and prognosis. In addition to the information which is typically extracted by trained pathologists, there is also information that is not yet exploited, simply because we do not yet understand the impact of all cellular and tissular features that could be predictive of outcome. In this paper, we address a question that can currently not be solved by pathologists: the prediction of treatment efficiency for Triple Negative Breast Cancer (TNBC) patients from biopsy data.

Mots clés: Breast Cancer; Computer-aided detection and diagnosis; Deep Learn- ing; Digital Pathology; Histopathology; Triple Negative Breast Cancer

Domaines: Bio-informatique [q-bio.QM]; Vision par ordinateur et reconnaissance de formes [cs.CV]; Traitement des images [eess.IV]; Bio-Informatique, Biologie Systémique [q-bio.QM]

Peter Naylor, Joseph Boyd, Marick Lae, Fabien Reyal, Thomas Walter. Predicting Residual Cancer Burden in a triple negative breast cancer cohort. 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI), Apr 2019, Venice, Italy. pp.933-937, ⟨10.1109/ISBI.2019.8759205⟩. ⟨hal-02440647⟩