| International Session（Panel Discussion）1(JGES・JSGE・JSGS・JSGCS) |
| Sat. November 4th 14:00 - 17:00 Room 9: Portopia Hotel Main Building Kairaku 3 |
| Artificial Intelligence Quantifying Endoscopic Severity of Ulcerative Colitis in Gradation Scale | |||
| Kaoru Takabayashi1, Takanori Kanai2, Haruhiko Ogata1 | |||
| 1Center for Diagnostic and Therapeutic Endoscopy, School of Medicine, Keio University, 2Division of Gastroenterology and Hepatology Department of Internal Medicine, School of Medicine, Keio University | |||
|
|||
|
|||
| Background The existing endoscopic scores for Ulcerative Colitis(UC) are objectively categorize the severity of the disease based on presence or absence of endoscopic findings. The purpose of this study was to develop AI that can accurately represent the assessment of endoscopic severity of UC by IBD experts. Methods This study was conducted using a method that incorporates data on relationships identified by comparing the severity of paired images created from 59595 endoscopic images by IBD experts into a Ranking- Convolutional Neural Network, and then the severity was then expressed on a scale called UC Endoscopic Gradation Scale (UCEGS) rather than a score. Correlation coefficients were calculated to ensure that there were no inconsistencies in assessments of severity made using UCEGS diagnosed by AI and Mayo Endoscopic Subscore(MES), and also calculated correlation coefficients of the mean for test images that had been assessed using UCEGS by five IBD experts as well as the AI. Results Spearman's correlation coefficient between UCEGS diagnosed by AI and MES was approximately 0.89. Correlation coefficient between IBD experts and the AI of the evaluation results were all higher than 0.95 (P<0.01). Conclusions The AI developed here can act as a proxy of expert IBD endoscopists by presenting comprehensive assessments regarding the severity of UC. |
|||
|
Index Term 1: Artificial intelligence Index Term 2: Ulcerative Colitis |
|||
| Page Top | |||