Background: A computer-assisted (CAD) system was developed to assess and score the technical difficulty factors to classify the technical difficult degree of common bile duct (CBD) stones removal during Endoscopic Retrograde Cholangiopancreatography (ERCP). Subsequently, the efficacy of the CAD system was assessed through a multi-center, prospective, observational study.
Method: All patients who met the inclusion criteria were included in this study. Based on cholangiogram images, the CAD system analyzed the extent of difficulty in stone removal and classified it into difficult and easy groups. Further, we compared the differences in clinical endpoints, including attempts of stone extraction, stone extraction time, the whole operation time, and stone clearance rates between the two groups.
Results: In total, 173 patients with CBD stones from three hospitals were included in the study. Consequently, the difficult group had more extraction attempts (7.20 vs. 4.20, P<0.001), more frequent machine lithotripsy (30.43% vs. 7.09%, P<0.001), longer stone extraction time (16.59min vs. 7.65min, P<0.001), lower single-session stone clearance rate (73.91 % vs. 94.49 %, P<0.001) and lower total stone clearance rate (87.13% vs. 97.64%, P=0.019) compared to the easy group.
Conclusion: The CAD system can effectively assess and classify the extent of technical difficulty in endoscopic stone extraction during ERCP. Besides, it automatically provides a quantitative evaluation of CBD and stones, which in turn would help endoscopists to apply suitable procedures and interventional methods to minimize the possible risks of endoscopic stone removal.