LXY-Medical Image and Health Informatics Lab

Ph.D. Student

Xinyue Li

2024-Present

Medical image-based Early Diagnosis of Tumor
Generalization

School of Biomedical Engineering
Shanghai Jiao Tong University
Room 419, Med-X Research Institute
1954 Hua Shan Road, Shanghai, China

Research Interests

My primary focus is to develop generally viable intelligent models for early diagnosis of tumor via medical images, where the key challenge is the stability of small tumor detection. To address this challenge, my research efforts span multi-instance learning, graph neural networks, and generalizable classification.

Brief Biography

Ph.D. student, 2024-Present, School of Biomedical Engineering, Shanghai Jiao Tong University, China.

M.S., 2021-2024, School of Biomedical Engineering, Shanghai Jiao Tong University, China.

B.S., 2017-2021, School of Biomedical Engineering, Shanghai Jiao Tong University, China.

Current Research Topics

Non-contrast CT-based early diagnosis of pancreatic cancer

Pancreatic cancer is one of the most fatal digestive system tumors, mainly because there are no characteristic symptoms in the early stages, and patients are typically at late stage when diagnosed. Currently, there is no effective and convenient early diagnosis method for pancreatic cancer. To achieve universal early diagnosis, we are committed to exploring the potential of widely accessible non-contrast CT for pancreatic cancer diagnosis through designing low-contrast image generalizable classification algorithms for tumors (TMI 2023). Finally, we expect to build a pancreatic cancer early diagnosis system that can be used for physical examinations, providing timely and accurate assessment for high-risk groups.