Hen graduated with honors from his undergraduate degree and is currently working on developing AI-based methods to help doctors diagnose breast cancer, a topic he plans to pursue for his PhD at the University of Oxford.
Hen’s research focuses on developing AI-based tools to improve medical decision-making, particularly for predicting patient survival. One of the key challenges in this area is dealing with the inherent uncertainty in predictions – it is sometimes difficult to assess their accuracy because information about patient survival is only available after long-term follow-up.
For example, if an AI system predicts that a patient with breast cancer will survive five years but there is only two years of follow-up, it is not possible to know whether the prediction is accurate – we have no information about the future. Such uncertainty makes it difficult to assess the model’s performance and make informed decisions based on it.
In collaboration with Dr. Yaniv Romano, Shai Feldman, Dr. Gil Shamai and Prof. Ron Kimmel, the research team developed an innovative statistical method that provides an envelope of reliability for survival predictions using confidence intervals. The envelope allows for the integration of artificial intelligence capabilities with a reliable assessment of the level of risk, thus supporting more informed decision-making.
The article he wrote, “Conformalized Survival Analysis for General Right-Censored Data,” was recently accepted for the prestigious ICLR conference to be held this year in Singapore.
Yaniv Romano, who is supervising Hen in his thesis, says that “Hen is a brilliant and exceptionally talented student – his selection for the prestigious Rhodes Scholars list is not surprising at all. He embodies the excellence of the Technion in general and the Faculty of Computer Science in particular. I am curious to see how his skills will be expressed in the future of his academic and professional career".
To the article in the Jerusalem Post, link
[Back to the news index]