Prof. Dr. Jakob N. Kather
Predicting cancer biomarkers from pathology images with artificial intelligence
Precision oncology requires molecular and genetic testing of tumor tissue. For many of these tests, universal implementation in clinical practice is limited. However, for virtually every cancer patient, tissue slides stained with hematoxylin and eosin (H&E) is available. Recently, we and others have demonstrated that we can use deep learning to predict biomarkers directly from routine H&E histology images. This talk will summarize the state of the art of deep learning in oncology, demonstrate emerging use cases and discuss the clinical implications of these new biomarkers.