Improving Early Breast Cancer Detection in Diverse Ethnic Groups in Israel

Breast cancer (BC) is the most common cancer diagnosed among women in Western countries. As early detection is vital for improved survival, nationwide mammography screens are conducted to detect BC malignancies at an early stage.

The current recommendation of health providers in Israel is that most women should take mammography screens every other year, starting at the age of 50. However, this 'one-size fits all' approach is clearly sub-optimal and our project aims at better and more personalized screening strategies where decisions around the starting age and frequency of screening are based on the individual risk of each woman.

To this goal, using AI-based transfer learning techniques we are constructing genetic risk score models that take into account the cumulative effect of hundreds of genetic BC risk variants (called polygenic risk score (PRS), and optimize these models to women of different Israeli ethnicities.

Our results will form a significant milestone towards the implementation of risk-based BC screening programs in Israel.