The MULTISTANCE project is aimed at tackling one of the most challenging issues in cancer treatment today: chemotherapy resistance in ovarian high-grade serous carcinoma (HGSC), which is the most prevalent and deadly type of ovarian cancer. At the heart of MULTISTANCE is the development of computational models that combine genomic data from cancer cells with the characteristics of these cells in histopathological images obtained during routine diagnosis. These models leverage a visual language model based on a "Foundation Model," which has been trained on over a million pairs of histopathological images and their clinical descriptions. The model is designed to interpret complex medical images and provide detailed, understandable insights into the cancer's characteristics. The outcome of MULTISTANCE will be publicly available models that are able to integrate multiple types of data and predict chemotherapy outcomes.

2025

2027