Meta-Analysis
Copyright ©The Author(s) 2023.
World J Gastrointest Endosc. Aug 16, 2023; 15(8): 528-539
Published online Aug 16, 2023. doi: 10.4253/wjge.v15.i8.528
Table 1 Summary of all studies investigating the development of machine learning algorithms for the endoscopic ultrasound diagnosis of subepithelial lesions
Ref.
Geographical setting
Study type
Patients
Sex (male/female)
GIST
GIL
Other SELs
Images
Tumor location
AI model
External validation
Endoscopists comparison
Histopathology
Minoda et al[29], 20221EasternRetrospective5233/19361422718Esophagus, duodenum, and colonCNNYesYesYes
Hirai et al[33], 2022EasternRetrospective664231/1884359710016110Esophagus, stomach, and duodenumCNNYesYesYes
Tanaka et al[34], 2022EasternRetrospective5328/254211-10600StomachCNNNoYesYes
Yang et al[32], 2022EasternRetrospective752337/415348404-10439Esophagus, stomach, duodenum, colon, and rectumCNNYesYesYes
Oh et al[30], 2021EasternRetrospective168NI12543-546StomachCNNYesYesYes
Seven et al[31], 2022EasternRetrospective14572/7310936-1362Esophagus, stomach, and duodenumCNNYesYesYes
Kim et al[27], 2020EasternRetrospective248111/13715755351117StomachCNNYesYesYes
Minoda et al[28], 2020EasternRetrospective273138/13518465243980StomachCNNYesYesYes
Table 2 Artificial intelligence and experienced endoscopists’ diagnostic performance for gastrointestinal stromal tumor and differentiation of leiomyoma

Prevalence (mean)
Sensitivity
Specificity
LR+
LR-
AUC
PTP+
PTP-
AI
GIST 68%92%80%4.260.090.94990%16%
GIST vs GIL70%193%90%6.480.060.96694%12%
Endoscopists
GIST67%72%70%2.510.420.77784%46%
GIST vs GIL67%173%75%2.610.370.81984%43%

  • Citation: Gomes RSA, de Oliveira GHP, de Moura DTH, Kotinda APST, Matsubayashi CO, Hirsch BS, Veras MO, Ribeiro Jordão Sasso JG, Trasolini RP, Bernardo WM, de Moura EGH. Endoscopic ultrasound artificial intelligence-assisted for prediction of gastrointestinal stromal tumors diagnosis: A systematic review and meta-analysis. World J Gastrointest Endosc 2023; 15(8): 528-539
  • URL: https://www.wjgnet.com/1948-5190/full/v15/i8/528.htm
  • DOI: https://dx.doi.org/10.4253/wjge.v15.i8.528