Artificial intelligence-based diagnosis of upper gastrointestinal subepithelial lesions on endoscopic ultrasonography images.

BACKGROUND: Endoscopic ultrasonography (EUS) is useful for the differential diagnosis of subepithelial lesions (SELs); however, not all of them are easy to distinguish. Gastrointestinal stromal tumors (GISTs) are the commonest SELs, are considered potentially malignant, and differentiating them from benign SELs is important. Artificial intelligence (AI) using deep learning has developed remarkably in the medical field. This study aimed to investigate the efficacy of an AI system for classifying SELs on EUS images. METHODS: EUS images of pathologically confirmed upper gastrointestinal SELs (GIST, leiomyoma, schwannoma, neuroendocrine tumor [NET], and ectopic pancreas) were collected from 12 hospitals. These images were divided into development and test datasets in the ratio of 4:1 using random sampling; the development dataset was divided into training and validation datasets. The same test dataset was diagnosed by two experts and two non-experts. RESULTS: A total of 16,110 images were collected from 631 cases for the development and test datasets. The accuracy of the AI system for the five-category classification (GIST, leiomyoma, schwannoma, NET, and ectopic pancreas) was 86.1%, which was significantly higher than that of all endoscopists. The sensitivity, specificity, and accuracy of the AI system for differentiating GISTs from non-GISTs were 98.8%, 67.6%, and 89.3%, respectively. Its sensitivity and accuracy were significantly higher than those of all the endoscopists. CONCLUSION: The AI system, classifying SELs, showed higher diagnostic performance than that of the experts and may assist in improving the diagnosis of SELs in clinical practice.

Feasibility of endoscopic ultrasonography using a 60-MHz ultrasound miniature probe in the upper gastrointestinal tract.

PURPOSE: The use of higher frequencies in ultrasound allows for a more detailed image. This study aimed to investigate the feasibility of delineating the gastrointestinal wall using a 60-MHz miniature ultrasound probe. METHODS: A phantom study was performed using a multipurpose ultrasonic phantom model, and the depth of imaging was evaluated using 60-MHz and 20-MHz miniature probes and 7.5-MHz conventional convex-type endoscopic ultrasonography. A total of 25 visualized areas from a total of 16 specimens from 16 patients were enrolled. The structures of the layers of the esophagus, stomach, and duodenum were evaluated using a 60-MHz probe and a pathological specimen created from endoscopically or surgically resected specimens. RESULTS: The 60-MHz probe was able to render to a depth of 2 mm and visualize the esophagus, stomach, and duodenum in five layers, respectively, within the depiction range. The depiction ranges of the 20-MHz probe and 7.5-MHz conventional endoscopic ultrasonography were 5 mm and 60 mm, respectively. The 60-MHz probe visualized the muscularis mucosae as the fourth layer in the esophagus, the fourth layer in the stomach, and the second layer in the duodenum. Muscularis mucosae were delineated in almost all cases, except in two cases where the layered structure disappeared. CONCLUSION: The 60-MHz probe provided good visualization of the muscularis mucosae and structure of the layers down to the submucosa, which improves the ability to diagnose the depth of early cancer invasion of the upper gastrointestinal tract, leading to more appropriate treatments.