Back

Artificial intelligence assisted standard white light endoscopy accurately characters early colorectal cancer: a multicenter diagnostic study

Meng, S.; Zheng, Y.; Su, R.; Wang, W.; Zhang, Y.; Xiao, H.; Han, Z.; Zhang, W.; Qin, W.; Yang, C.; Yan, L.; Xu, H.; Bu, Y.; Zhong, Y.; Zhang, Y.; He, Y.; Qiu, H.; Xu, W.; Chen, H.; Wu, S.; Jiang, Z.; Zhang, Y.; Dong, C.; Hu, Y.; Xie, L.; Li, X.; Jiang, J.; Zhu, H.; Li, W.; Wen, Z.; Zheng, X.; Sun, Y.; Zhou, X.; Ding, L.; Zhang, C.; Pan, W.; Wu, S.; Hu, Y.

2020-02-23 oncology
10.1101/2020.02.21.20025650 medRxiv
Show abstract

Colorectal cancer (CRC) is the third in incidence and mortality1 of cancer. Screening with colonoscopy has been shown to reduce mortality by 40-60%2. Challenge for screening indistinguishable precancerous and noninvasive lesion using conventional colonoscopy was still existing3. We propose to establish a propagable artificial intelligence assisted high malignant potential early CRC characterization system (ECRC-CAD). 4,390 endoscopic images of early CRC were used to establish the model. The diagnostic accuracy of high malignant potential early CRC was 0.963 (95% CI, 0.941-0.978) in the internal validation set and 0.835 (95% CI, 0.805-0.862) in external datasets. It achieved better performance than the expert endoscopists. Spreading of ECRC-CAD to regions with different medical levels can assist in CRC screening and prevention.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.