Séminaire (Organisé par l’Equipe de recherche DI)

Chunfeng LIAN

Doctorant Heudiasyc

Outcome Prediction In Tumor Therapy Based on Dempster-Shafer Theory

Lundi 16 février 2015 à 14h00 en salle GI042

Résumé :

Outcome prediction plays a vital role in cancer treatment. It can help to update and optimize the treatment planning. We aim to find discriminant features from both PET images and clinical characteristics, so as to predict the outcome of a treatment to adapt the therapy. As both information sources are imprecise, we propose a feature selection method based on Dempster-Shafer theory to tackle this problem.

A specific objective function with sparsity constraint is developed to search for a feature subset that leads to increasing prediction performance and decreasing data imprecision simultaneously. Our approach was applied to several real datasets and two clinical datasets concerning to lung tumor et esophageal tumor, showing good performance.


FR SHIC 3272

Collegium UTC/CNRS