Yoni Schirris

PhD Candidate


Yoni has an interdisciplinary background with a formal education in neuroscience, biomedical science, economics, and artificial intelligence. With years of experience in operational and technical roles, he now wishes to bridge the gap between artificial intelligence, medical science, and clinical implementation of AI models. He obtained his master's degree in AI at the University of Amsterdam in 2020. His thesis titled "Predicting DNA Damage Repair Deficiencies directly from H&E Whole-Slide Images using Deep Learning" was supervised by dr Jonas Teuwen, dr Hugo Horlings and dr Efstratios Gavves.

He is now a PhD candidate under the HISTO-AI project (2020-2024), with the aim to develop deep learning methods to predict genomic and transcriptomic information directly from H&E WSIs. The larger goal of this project is to predict which patient can benefit from immune therapy or targeted therapy, directly from commonly available H&E WSIs.

Yoni pitched the topic of his PhD for the 3MT (three minute thesis) competition, and won the UvA-wide finals. You can see the winning pitch here.

  1. Y. Schirris, E. Gavves, I. Nederlof, H. M. Horlings, J. Teuwen, "DeepSMILE: Contrastive self-supervised pre-training benefits MSI and HRD classification directly from H&E whole-slide images in colorectal and breast cancer", Medical Image Analysis, 2022-07-01, 79
  2. Y. Schirris, M. Engelaer, A. Panteli, H. M. Horlings, E. Gavves, J. Teuwen, J. E. Tomaszewski, A. D. Ward, "WeakSTIL: weak whole-slide image level stromal tumor infiltrating lymphocyte scores are all you need", Medical Imaging 2022: Digital and Computational Pathology, 2022-4-4
  3. Y. Schirris, E. Gavves, I. Nederlof, H. M. Horlings, J. Teuwen, "DeepSMILE: Self-supervised heterogeneity-aware multiple instance learning for DNA damage response defect classification directly from H&E whole-slide images", arXiv:2107.09405 [cs, eess], 2021-07-20