Endocytoscopy (EC) facilitates real-time histological diagnosis of esophageal lesions in vivo. We developed a deep-learning artificial intelligence (AI) system for analysis of EC images and compared its diagnostic ability with that of an expert pathologist and nonexpert endoscopists. Our new AI was based on a vision transformer model (DeiT) and trained using 7983 EC images of the esophagus (2368 malignant and 5615 nonmalignant). The AI evaluated 114 randomly arranged EC pictures (33 ESCC and 81 nonmalignant lesions) from 38 consecutive cases. An expert pathologist and two nonexpert endoscopists also analyzed the same image set according to the modified type classification (adding four EC features of nonmalignant lesions to our previous classification). The area under the curve calculated from the receiver-operating characteristic curve for the AI analysis was 0.92. In per-image analysis, the overall accuracy of the AI, pathologist, and two endoscopists was 91.2%, 91.2%, 85.9%, and 83.3%, respectively. The kappa value between the pathologist and the AI, and between the two endoscopists and the AI showed moderate concordance; that between the pathologist and the two endoscopists showed poor concordance. In per-patient analysis, the overall accuracy of the AI, pathologist, and two endoscopists was 94.7%, 92.1%, 86.8%, and 89.5%, respectively. The modified type classification aided high overall diagnostic accuracy by the pathologist and nonexpert endoscopists. The diagnostic ability of the AI was equal or superior to that of the experienced pathologist. AI is expected to support endoscopists in diagnosing esophageal lesions based on EC images.

Duodenal cancer is a leading cause of death after colectomy in patients with familial adenomatous polyposis (FAP). Detailed endoscopic evaluation of duodenal lesions with potential for carcinoma development is therefore mandatory. Here we investigated the features of duodenal lesions in FAP patients using an endocytoscopy system (ECS). We retrospectively reviewed duodenal lesions in 15 cases of FAP using an ECS (GIF-H290EC) with methylene blue (MB) as the vital dye. With reference to the Spigelman classification, we investigated the number of lesions using white light (WL), narrow-band imaging (NBI), and MB staining. Using the maximum magnification power of the ECS we investigated the histology (duct openings or finger-like projections) and grade of dysplasia (presence or absence of enlarged oval-shaped nuclei) of the lesions. The number of duodenal lesions increased in ascending order of WL, NBI, and MB (P < 0.05). Among 51 MB-unstained lesions, 46 (90.2%) were proven to be duodenal neoplasms histologically. Duct openings were seen in 90.2% of tubular adenomas and tubulovillous adenomas. Finger-like projections were seen in 33.3% of tubular adenomas and in 88.2% of tubulovillous adenomas. Enlarged oval-shaped nuclei were observed in 100% of duodenal cancers, 33.3% of high-grade adenomas, and 9.4% of low-grade adenomas. MB staining allows more accurate detection of duodenal neoplasms in comparison to conventional WL and NBI observation. In cases of FAP, use of the maximum magnification power of the ECS may allow selection of lesions with high malignant potential.

The development of a ganglion in the hip joint is a rare cause of lower limb swelling. We herein describe a case of a ganglion of the hip with compression of the femoral vein that produced signs and symptoms that mimicked a deep vein thrombosis. Needle aspiration of the ganglion was easily performed, and swelling of the left lower limb promptly improved. Intensive follow-up of this case was important because the recurrence rate of ganglions after needle aspiration is high.