Premium
Effectiveness of systematic training in the application of narrow‐band imaging international colorectal endoscopic (NICE) classification for optical diagnosis of colorectal polyps: Experience from a single center in China
Author(s) -
Sikong Yinhe,
Lin Xiangchun,
Liu Kuiliang,
Wu Jing,
Lin Wu,
Wei Nan,
Jiang Guojun,
Tai Weiping,
Su Hui,
Liu Hong,
Meng Mingming
Publication year - 2016
Publication title -
digestive endoscopy
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.5
H-Index - 56
eISSN - 1443-1661
pISSN - 0915-5635
DOI - 10.1111/den.12600
Subject(s) - medicine , nice , narrow band imaging , magnification , colorectal polyp , colorectal cancer screening , receiver operating characteristic , single center , radiology , colorectal cancer , artificial intelligence , endoscopy , colonoscopy , cancer , computer science , programming language
Background and Aim To evaluate the effectiveness of systematic training in the application of narrow‐band imaging (NBI) International Colorectal Endoscopic (NICE) classification for the prediction of the histology of colorectal polyps. Methods This is a single‐center study. In total, 260 still images of colorectal polyps from 225 patients were included. Two series of 130 images obtained using NBI and white light without magnification were distributed to 10 endoscopists—five highly experienced endoscopists (HEE group) and five less experienced endoscopists (LEE group)—for assessment using the NICE classification, before and after systematic training. Results Before systematic training, accuracy in both groups was 79.54% and specificity was relatively poor (HEE: 62.46%, LEE: 69.23%, P = 0.297). After systematic training, specificity significantly improved (HEE: 96.61%, LEE: 97.54%, P = 0.000 and P = 0.013, respectively). Accuracy also significantly increased to 94.93% and 96.46% in the HEE and LEE groups, respectively. Sensitivity and negative predictive value did not significantly improve in the post‐test; however, both were high in both the pre‐ and post‐test. The κ‐values in both groups were excellent (HEE: 0.93, LEE: 0.91). Among the components of the NICE classification, surface pattern yielded the highest performance, whereas color yielded the lowest. Conclusion Systematic, feedback‐based, training programs can help achieve high accuracy and good interobserver agreement in the application of the NICE classification for the prediction of the histology of colorectal polyps by endoscopists with different levels of experience.