Xuejian He

Title: Xuejian He: Advancing AI-Assisted Gastrointestinal Endoscopy

Introduction

Xuejian He is an inventor based in Hong Kong, CN, whose work focuses on applying artificial intelligence to gastrointestinal endoscopy. He is associated with the Hong Kong Applied Science and Technology Research Institute and is named on 4 patents.

Latest Patents

He’s recent patent work addresses AI systems designed to support endoscopic diagnosis and improve examination quality.

“Portable Edge AI-Assisted Diagnosis and Quality Control System for Gastrointestinal Endoscopy” describes a decision-support system that analyzes endoscopic images using convolutional neural networks. The invention uses lightweight kernel models, knowledge distillation, and multi-task learning to reduce computational demands, enabling deployment on edge-computing devices near examination sites. It can be configured for either esophagogastroduodenoscopy or colonoscopy, and combines lesion-detection and quality-control results into integrated information for endoscopists.

“Multi-Functional Computer-Aided Gastroscopy System Optimized with Integrated AI Solutions and Method” presents a system that uses multiple deep-learning models to perform gastroscopy-related tasks with low latency. Its optimizations span model architecture, modular design, and system functionality. The approach combines tasks such as Helicobacter pylori infection classification and lesion detection, uses temporal information to improve site recognition, and speeds up inference and image preprocessing.

Career Highlights

He’s patent portfolio reflects an emphasis on practical AI deployment in medical settings. The inventions aim to balance diagnostic support and examination quality with the computational and latency constraints of clinical use.

Collaborations

The supplied information identifies Lu Wang and Ping Shun Leung as He’s coworkers. His work is associated with the Hong Kong Applied Science and Technology Research Institute.

Conclusion

Xuejian He’s inventions apply efficient AI methods to gastrointestinal endoscopy, combining diagnostic assistance, quality control, and resource-conscious system design.