Subscribe to RSS
Make the entrance wider and the depth deeperReferring to Huang L et al. p. 4–11
The use of artificial intelligence (AI) in medicine is under active investigation. In recent years, AI tools have emerged that require little or no operator training; these are increasingly being used in the clinic. Indeed, advances in imaging diagnostics, as well as automatic diagnosis using radiography and three-dimensional images that do not require human intervention or operation are being put to use . AI may perform the functions of physicians in various fields.
“A two-step AI system would be useful: the first step would involve pre-procedural imaging for scheduling, and the second step would make use of information obtained after endoscope insertion, including endoscopic retrograde cholangiography, to prepare for the next stage.”
Regarding the use of AI in endoscopy, polyps can be detected by colonoscopy, lesions in the small intestine can be detected and diagnosed by capsule endoscopy , esophageal cancer and gastric cancer can be detected by esophagogastroduodenoscopy, and the depth of invasion can be predicted . In the biliopancreatic field, AI research has focused on pancreatic annotation in endoscopic ultrasonography, detection of pancreatic tumors, and differentiation of benign and malignant lesions  . Urinary tract stones can be detected and their composition analyzed using AI , but AI is not yet used to evaluate biliary tract stones.
Bile duct stones can be analyzed, and the difficulty of treatment predicted by AI . Endoscopic treatment for bile duct stones is evolving, and almost all bile duct stones can be removed endoscopically. Endoscopic sphincterotomy followed by large-balloon dilation can treat large stones without lithotripsy. Peroral cholangioscopic lithotripsy can be used to remove intrahepatic and even cystic duct stones, which were formerly indications for surgery . However, institutions employ different strategies and devices, and thus experience varying difficulties.
In this issue of Endoscopy, Huang et al. define “difficulty” and describe the development of a computer-assisted (CAD) system, potentially heralding a new era of endoscopic treatment . Many high-volume centers are experiencing increasing numbers of difficult cases. A more effective and efficient treatment strategy is thus needed.
However, several problems need to be overcome before practical application of this CAD system. First, endoscopic retrograde cholangiography (ERC) is used for diagnosis. The bile duct was cannulated at the time of ERC, but the cannulation step was the problem in most cases. An additional AI system to assess the difficulty of biliary cannulation is needed and, for scheduling purposes, preoperative diagnosis by a diagnostic imaging modality such as magnetic resonance cholangiopancreatography and computed tomography would be helpful.
Second, the CAD system has only two outputs – “easy” and “difficult.” The difficult outcome encompasses a wide range of situations, from stones removable using a mechanical lithotripter to those in which peroral cholangioscopy is required. Therefore, it is necessary to determine not only the difficulty but also the optimal method of stone removal to determine whether treatment is feasible. Third, the grading scale needs to be re-examined. The scale is based on four factors: 1) size and number of stones and the 2) angle, 3) distance, and 4) minimum diameter of the distal bile duct. However, other factors may contribute to the difficulty of stone removal, for example, the age and condition of the patient, the type of medication, the presence or absence of parapapillary diverticula, the position and size of the duodenal papilla, and the stability of the endoscope. Most factors related to the difficulty of endoscopic treatment exert their influence after insertion of the endoscope; however, we want to assess the difficulty of stone removal pre-procedure. A two-step AI system would be useful in this regard: the first step would involve pre-procedural imaging for scheduling, and the second step would make use of information obtained after endoscope insertion, including ERC, to prepare for the next stage.
Diagnosis is considered a relatively easy application of AI in medicine; however, treatment is more important. AI systems could also contribute to treatment decision making, taking into account the efficacy, efficiency, and cost-effectiveness of the available options.
Article published online:
26 September 2022
© 2022. Thieme. All rights reserved.
Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany
- 1 Zhang Y, Liang Y, Ding J. et al. A prior knowledge guided deep learning based semi-automatic segmentation for complex anatomy on MRI. Int J Radiat Oncol Biol Phys 2022; DOI: 10.1016/j.ijrobp.2022.05.039.
- 2 Sullivan P, Gupta S, Powers PD. et al. Artificial intelligence research and development for application in video capsule endoscopy. Gastrointest Endosc Clin N Am 2021; 31: 387-397
- 3 Nagao S, Tsuji Y, Sakaguchi Y. et al. Highly accurate artificial intelligence systems to predict the invasion depth of gastric cancer: efficacy of conventional white-light imaging, nonmagnifying narrow-band imaging, and indigo-carmine dye contrast imaging. Gastrointest Endosc 2020; 92: 866-873
- 4 Kuwahara T, Hara K, Mizuno N. et al. Artificial intelligence using deep learning analysis of endoscopic ultrasonography images for the differential diagnosis of pancreatic masses. Endoscopy 2022; 54 DOI: 10.1055/a-1850-6717.
- 5 Kuwahara T, Hara K, Mizuno N. et al. Current status of artificial intelligence analysis for endoscopic ultrasonography. Dig Endosc 2021; 33: 298-305
- 6 Kim US, Kwon HS, Yang W. et al. Prediction of the composition of urinary stones using deep learning. Investig Clin Urol 2022; 63: 441-447
- 7 Huang L, Xu Y, Chen J. et al. An artificial intelligence difficulty scoring system for stone removal during ERCP: a prospective validation. Endoscopy 2022; DOI: 10.1055/a-1850-6717.
- 8 Manes G, Paspatis G, Aabakken L. et al. Endoscopic management of common bile duct stones: European Society of Gastrointestinal Endoscopy (ESGE) guideline. Endoscopy 2019; 51: 472-491
- 9 Huang L, Xu Y, Chen J. et al. An artificial intelligence difficulty scoring system for stone removal during ERCP: a prospective validation. Endoscopy 2022; 55: 4-11 DOI: 10.1055/a-1850-6717.