合作项目
Virtual Doctor for GIST
This project targets end-to-end care for gastrointestinal stromal tumors (GIST), building a "virtual doctor" that serves both clinicians and patients. Powered by a multimodal foundation model, the system integrates CT, endoscopy, ultrasound, pathology whole-slide imaging (WSI), clinical report text, and follow-up data collected via the mini-program. On this basis it executes a closed loop of risk stratification → individualized treatment planning → therapy efficacy and recurrence-risk prediction → follow-up and medication management. For clinicians, it provides an MDT (multidisciplinary team) decision dashboard, standardized documentation, and a traceable evidence chain; for patients, it offers medication reminders, symptom reporting, and risk alerts. Technically, the system combines our self-developed multimodal encoder and diagnostic-therapeutic agent, aiming to significantly reduce information curation and decision-making costs in real clinical workflows while improving consistency and interpretability of care.
Hebei Medical University Fourth Hospital
Rheumatology and Immunology Lupus Erythematosus Agent-Assisted Diagnosis and Treatment Project
The laboratory focuses on the research and development of a Systemic Lupus Erythematosus (SLE)-assisted diagnosis large model and its clinical application, and has launched cooperation with The Seventh Affiliated Hospital of Sun Yat-sen University.
The Seventh Affiliated Hospital, Sun Yat-sen University
Intelligent multi-modal medical data platform.
Our laboratory is collaborating with The Sixth Affiliated Hospital of Sun Yat-sen University (SYSU Sixth Hospital) to develop an intelligent multi-modal medical data platform.
Sixth Affiliated Hospital of Sun Yat-sen University
Cross-modal Data Processing and Machine Learning Model Training
Pathological diagnosis requires the integration of cross-modal information such as slice images, clinical texts, and genetic data. However, traditional AI struggles to efficiently integrate this information, which limits the accuracy of diagnosis. The laboratory and the Bingli Team of Anbiping Group have jointly developed and applied for the patent of "Cross-Modal Data Processing and Model Training Technology".
LBP
The project with Hebei Medical University Fourth Hospital
The project is predicting the mutation status of the PIK3CA gene based on HE slices and the efficacy of neoadjuvant therapy.
Hebei Medical University Fourth Hospital
Cervical histology
We cooperated with Peking Union Medical College and Shenzhen Maternal and Child Health Hospital to develop an efficient and accurate artificial intelligence diagnosis system for cervical cancer omics pathological images.
Peking Union Medical College
Colposcopy Report Interpretation
In cooperation with Peking Union Medical College and Shenzhen Maternal and Child Health Hospital, the intelligent interpretation system is designed to improve the understanding of colposcopy reports.
Peking Union Medical CollegeThe project with Hebei Medical University Fourth Hospital
a multi-modal deep learning model based on HE images and HER2 immunohistochemistry images for predicting the prognosis of metastatic breast cancer.
Hebei Medical University Fourth HospitalThe project with Hebei Medical University Fourth Hospital
We jointly dedicated to researching the application of artificial intelligence in the field of breast cancer.
Hebei Medical University Fourth Hospital
Treating Gastroesophageal Reflux (GER)
This project aims to develop a multimodal AI model for predicting the efficacy of various drug therapies in treating GER symptoms.
The First Affiliated Hospital of Sun Yat-sen University (Nansha)
Lung cancer case-level large model
Lung cancer is the most common and diverse type of cancer in modern society. This project focuses on case-level lung cancer tasks that require comprehensive judgment based on multiple slices, particularly the task of evaluating the immunohistochemical heterogeneity of lung cancer. It develops report generation and diagnostic assistance models to support pathologists in making case-level diagnoses.
Zhejiang Cancer Hospital
Recurrence patterns of liver cancer
The collaboration focuses on exploring the recurrence patterns of liver cancer based on liver WSI slices.
Wuhan Tongji Hospital
LBP Labeling
We have established a cooperative relationship with LBP Group to develop an intelligent labeling tool for pathological images for problems.
LBP