EXPERIMENTALLY-GENERATED GROUND TRUTH FOR DETECTING CELL TYPES IN AN IMAGE-BASED IMMUNOTHERAPY SCREEN

Joseph Boyd , Zelia Gouveia , Franck Perez , Thomas Walter (1, 2, 3)

1 CBIO - Centre de Bioinformatique
2 Cancer et génome: Bioinformatique, biostatistiques et épidémiologie d'un système complexe
3 PSL - Université Paris sciences et lettres

Chimeric antigen receptor is an immunotherapy whereby T lympho-cytes are engineered to selectively attack cancer cells. Image-based screens of CART cells, combining phase contrast and fluorescence microscopy, suffer from the gradual quenching of the fluorescent signal, making the reliable monitoring of cell populations across time-lapse imagery difficult. We propose to leverage the available fluorescent markers as an experimentally-generated ground truth, without recourse to manual annotation. With some simple image processing, we are able to segment and assign cell type classes automatically. This ground truth is sufficient to train a neural object detection system from the phase contrast signal alone, potentially eliminating the need for the cumbersome fluorescent markers. This approach will underpin the development of cheap and robust microscope-based protocols to quantify CART activity against tumor cells in vitro.

Mots clés: High Content Screening; deep learning; object detection; phase contrast microscopy; Computer Vision

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]

Joseph Boyd, Zelia Gouveia, Franck Perez, Thomas Walter. EXPERIMENTALLY-GENERATED GROUND TRUTH FOR DETECTING CELL TYPES IN AN IMAGE-BASED IMMUNOTHERAPY SCREEN. 2020 IEEE 17th International Symposium on Biomedical Imaging (ISBI), Apr 2020, Iowa City, United States. pp.886-890, ⟨10.1109/ISBI45749.2020.9098696⟩. ⟨hal-02976141⟩