Taiwan
January 31, 2024AI Cure
January 31, 20242019-2020 “Revolutionizing Gastroenterology: The Emergence of Artificial Intelligence in Endoscopic Procedures”
AI-assisted endoscopy is rapidly evolving, with systems like ENDOANGEL and GI Genius improving colonoscopy accuracy and adenoma detection rates...
The field of AI-assisted endoscopy is developing and has a bright future. Initially, colon polyps with and without cancer were identified, differentiated, and characterized using computer-aided diagnosis (CAD).
The application of artificial intelligence (AI) in endoscopic practice is expanding quickly, and it has profited from the recent technological revolution in the endoscopy and imaging fields. Two of the mentioned technologies are approved for usage, even if most have been demonstrated in concept. A CNN-based system called ENDOANGEL (Wuhan EndoAngel Medical Technology Company, Wuhan, China) was created in 2019 and has an accuracy of 91.89% when it comes to objectively assessing bowel preparation every 30 seconds during the withdrawal phase of a colonoscopy. A recent randomized controlled study showed a significant improvement in adenoma detection rates using ENDOANGEL-assisted colonoscopy versus unassisted colonoscopy (17% vs. 8%; odds ratio, 2.18; 95% confidence interval, 1.31-3.62; P =.0026). The device was used to monitor real-time withdrawal speed and colonoscopy withdrawal time. (1)
The application of artificial intelligence (AI) in endoscopic practice is expanding quickly, and it has profited from the recent technological revolution in the endoscopy and imaging fields. Two of the mentioned technologies are approved for usage, even if most have been demonstrated in concept. A CNN-based system called ENDOANGEL (Wuhan EndoAngel Medical Technology Company, Wuhan, China) was created in 2019 and has an accuracy of 91.89% when it comes to objectively assessing bowel preparation every 30 seconds during the withdrawal phase of a colonoscopy.
A recent randomized controlled study showed a significant improvement in adenoma detection rates using ENDOANGEL-assisted colonoscopy versus unassisted colonoscopy (17% vs. 8%; odds ratio, 2.18; 95% confidence interval, 1.31-3.62; P =.0026). The device was used to monitor real-time withdrawal speed and colonoscopy withdrawal time. (1)
GI Genius (Medtronic, Minneapolis, Minn., USA) is an AI-enhanced endoscopy aid device designed to show a visual marker on a live video feed during endoscopic inspection, thereby aiding in identifying colorectal polyps. It is undergoing clinical study in the United States and has been licensed for usage in Europe. In a validation trial, GI Genius found polyps 82% of the time faster than endoscopists, with a total sensitivity per lesion of 99.7%. (2)
AI has also been used to enhance Barrett’s esophagus endoscopic images. A CAD method that was recently created by de Groof et al. was 90% sensitive, 88% specific, and 89% accurate in identifying pictures as either non-carcinogenic Barrett’s esophagus or neoplastic. Compared to 53 novice endoscopists, the CAD system’s accuracy was superior (88% vs. 73%).(3) Similarly, a real-time CAD system was created and trained using 5191 precancerous and 1480 malignant narrow-band pictures. It was able to distinguish between precancerous lesions and early esophageal squamous cell carcinoma with 98% specificity and 95% sensitivity. (4)
AI has also been used to enhance Barrett’s esophagus endoscopic images. A CAD method that was recently created by de Groof et al. was 90% sensitive, 88% specific, and 89% accurate in identifying pictures as either non-carcinogenic Barrett’s esophagus or neoplastic.
Compared to 53 novice endoscopists, the CAD system’s accuracy was superior (88% vs. 73%).(3) Similarly, a real-time CAD system was created and trained using 5191 precancerous and 1480 malignant narrow-band pictures. It was able to distinguish between precancerous lesions and early esophageal squamous cell carcinoma with 98% specificity and 95% sensitivity. (4)