MD, Chief Physician, Young Backbone Supported by the Beijing Excellent Talent Training Program. Currently serves as Deputy Director (in charge of daily operations) of Ward I, Gastrointestinal Cancer Center, Peking University Cancer Hospital. His main research directions include minimally invasive surgery and peri‑operative comprehensive therapy for gastric cancer, as well as the development and application of methodological frameworks for surgical clinical research. He has extensive experience in data‑driven clinical practices, such as surgical video data analysis, construction and quality control of specialized gastric cancer disease databases. He has published more than 120 academic papers in key journals including Nature Medicine, JAMA Surgery and Gastric Cancer. His publications have received over 1,970 citations on Google Scholar (h‑index: 24). He holds 4 authorized national invention patents.
Representative Achievements
1. Randomized controlled trial of laparoscopic surgery after chemotherapy for gastric cancer. As the world’s first randomized controlled study in this field, it demonstrated that laparoscopic surgery can effectively alleviate postoperative pain, reduce postoperative complications by 26% (20% vs. 46%), and improve tolerance to adjuvant chemotherapy in patients with advanced gastric cancer following neoadjuvant therapy. (JAMA Surgery, 2019)2. Development of computer‑vision‑based surgical quality assessment tools. Taking laparoscopic gastric cancer surgery as the research subject, this work firstly developed a subjective quality assessment tool and an annotation tool for surgical phases and events for laparoscopic gastric cancer surgery, and preliminarily built a subjective evaluation dataset for laparoscopic gastric cancer surgery quality. Based on the above work, preliminary exploration was carried out on computer‑aided quality assessment methods for laparoscopic gastric cancer surgery. Multiple video feature extraction and time‑series model design approaches have been mastered. A partitioning method based on hard‑sample mining and a reinforcement learning approach were proposed. An unsupervised surgical instrument segmentation method and an objective surgical quality evaluation model based on surgical field of view were designed. Early experimental results have shown promising prospects of these methods. He has been invited to give oral presentations at top computer vision conferences for multiple times and obtained 1 authorized national invention patent. (MICCAI 2019, MICCAI 2020, CVPR 2021)3. First‑line study of PD‑1/CTLA‑4 dual‑target monoclonal antibody combined with chemotherapy for advanced gastric cancer. This international‑first clinical study adopted Cadonilimab, a domestically‑developed innovative drug, to investigate PD‑1/CTLA‑4 dual‑target monoclonal antibody plus chemotherapy as first‑line treatment for advanced gastric cancer. It proved that this regimen yields favorable safety and efficacy and can significantly prolong patient survival. The publication of study findings provided solid evidence supporting the approval of Cadonilimab for the first‑line treatment indication of advanced gastric cancer. (Nature Medicine, 2024)
English