29 March 2023
Abstract:This thesis introduces a novel neural network architecture for predicting relapse-free survival curves using prostate cancer histopathology images as input. The proposed model employs state-of-the-art deep learning techniques, and experimental results demonstrate that it reaches a predictive performance similar to a pathologist. Furthermore, a combination of out-of-distribution detection and color transfer is proposed to make the model more robust to differences in the data acquisition protocol.
Date and Time: 29 March 2023 13:00 CET.
Online presentation, contact the Informatics study office for the Zoom meeting ID and password.