PADICS is a research project that aims to improve the quality and accuracy of the diagnosis of breast cancer. Indeed, at present, the test used is the Oncotype DX  (ODX), which makes it possible to determine a score of recurrence at 10 years and to predict a benefit from opaque and expensive chemotherapy. The medical community aspires to a more transparent and less expensive alternative to ODX testing. The Ki-67 proves to be a determining vector for prediction. To do this, we propose in this work an approach focused on the determination, interpretation and precision of the Ki-67 marker as well as to extract an alternative prediction approach to ODX.

The purpose of this project is to:

  • develop a tool based on deep neural networks and machine learning algorithms in order to make the measurement more precise (superimposed cells, colors, etc.)
  • include expertise in the development process
  • deal with the problem of data quality
  • automate the counting process to help pathologists at North Franche-Comté Hospital in the diagnostic process

This project is held by FEMTO-ST (Z. Al Masry and is supported by the Bourgogne Franche-Comté region and the National Research Agency (ANR):

One Post-Doc position is opened for this project. To apply please contact directly by email Z Al Masry.