Yuan Peng, PhD | Cancer Hospital, Chinese Academy of Medical Sciences;Yue Jian,PHD；Wang Xue,PhD；Yang ZiXuan，PhD
Competition Sponsor: Chinese Academy of Medical Sciences
Awardee Year: 2023
Breast cancer has become the world’s largest cancer. While its curative effect is improving day by day, how to achieve non-invasive evaluation of the efficacy of neoadjuvant therapy (NAT) in breast cancer patients so as to avoid surgery has become a bottleneck problem in clinical diagnosis and treatment that needs to be broken through. Existing non-invasive assessment strategies are limited to single-center, single-modality modeling, and suffer from low performance and poor versatility. Therefore, in the early stage of this study, the data of breast cancer patients from multiple centers across the country were collected and the establishment of an artificial intelligence (AI) curative effect prediction model was preliminarily completed. On this basis, this project intends to build a high-quality, large-sample, multi-time series, and multi-modal standard data set powered by AI, and design a self-supervised multi-modal feature mining method and a multicenter-based federated meta-learning strategy. Finally, the establishment and verification of a high-performance and versatile judgment model for pathological complete remission (pCR) after NAT will be completed, and a 5G-enabled remote diagnosis solution will be promoted.This project proposes an intelligent analysis method for breast cancer curative effect from data to model and fully empowered by AI, which will help the theoretical research of multimodal medical data feature mining and model construction, and will help solve the problem of non-invasive judgment of NAT curative effect in breast cancer patients Scientific issues,providing a new paradigm for the research of high-performance AI diagnosis and treatment assistance systems applicable to multi-centers.