Title : Application of artificial intelligence in the gastrointestinal endoscopy
Abstract:
This lecture will show the most advanced artificial intelligence (AI) researches related to the gastrointestinal endoscopy, and introduced the most proven AI device in the field of digestive endoscopy till now.
Early detection is the key to improve the prognosis of cancer. The quality of endoscopy examination plays a vital role for the detection of early cancers, which is the prerequisite for patients’ welfare. To ensure endoscopy quality, guidelines and expert consensus prescribed quality indicators for endoscopy. However, the quality of examination varied greatly due to the recommendations are not well implemented due to insufficient practical quality control tools and strong supervision. Here the first AI system I will introduce to you is ENDOANGEL, it aims at monitoring blind spots in the upper endoscopy, monitoring the withdraw speed, calculate the BBPS score, prompt blind spots, and report the percent of overspeed time in the lower endoscopy. Furthermore, the second AI system is the BP-MASTER, the system aims at providing real-time pancreaticobiliary and mediastinum EUS navigation.
Besides quality control, it is also vital to enhance the ability of lesion recognition. Here I will introduce a series of functions aimed at assisting in detecting lesions and diagnosing early cancer in real-time, for both upper and lower endoscopy. The systems were proved to have great performance in detecting lesions, which could improve the performance of endoscopists. Furthermore, explainable systems were constructed to further fulfill the clinical needs.
Electronic endoscopic reports facilitate effective communication and understanding among endoscopists, patients, and second-opinion advisors. Photo-documentation is a prerequisite for generating comprehensive and qualified reports. The manual procedure was usually problematic. Then, I will introduce an endoscopic automatic image reporting system, aiming to generate high quality image report. Automatic report is an important subject in future AI studies, especially in the field of digestive endoscopy.
Audience Take Away:
- Information about one of the most proven AI device in the field of digestive endoscopy.
- Experience of designing and conducting studies related to the artificial intelligence.
- Experience of introducing researches into clinical practice.
- Learning about how to use AI in endoscopy and how to make your own practice.