Search

Background/Objectives: Artificial intelligence (AI)-assisted endoscopic ultrasonography (EUS) diagnostic tools have shown excellent performance in diagnosing gastric mesenchymal tumors. This study aimed to assess whether incorporating clinical and endoscopic factors into AI-assisted EUS classification models based on digital image analysis could improve the diagnostic performance of AI-assisted EUS diagnostic tools. Methods: We retrospectively analyzed the data of 464 patients who underwent both EUS and surgical resection of gastric mesenchymal tumors, including 294 gastrointestinal stromal tumors (GISTs), 52 leiomyomas, and 41 schwannomas. AI-assisted classification models for GISTs and non-GIST tumors were developed utilizing clinical and endoscopic factors and digital EUS image analysis. Results: Regarding the baseline EUS classification models, the area under the receiver operating characteristic (AUC) values of the logistic regression, decision tree, random forest, K-nearest neighbor (KNN), and support vector machine (SVM) models were 0.805, 0.673, 0.781, 0.740, and 0.791, respectively. Using the new classification models incorporating clinical and endoscopic factors into the baseline classification models, the AUC values of the logistic regression, decision tree, random forest, KNN, and SVM models increased to 0.853, 0.715, 0.896, 0.825, and 0.794, respectively. In particular, the random forest and KNN models exhibited significant improvement in performance in Delong’s test (both p < 0.001). Conclusion: The diagnostic performance of the AI-assisted EUS classification models improved when clinical and endoscopic factors were incorporated. Our results provided direction for developing new AI-assisted EUS models for gastric mesenchymal tumors.

Background: Owing to the rising number of screening endoscopies and instrumental advances in endoscopic ultrasound (EUS), colorectal subepithelial tumors (SETs) are being increasingly detected. We aimed to determine the feasibility of endoscopic resection (ER) and the impact of EUS-based surveillance on colorectal SETs. Methods: The medical records of 984 patients with incidentally detected colorectal SETs between 2010 and 2019 were retrospectively reviewed. Overall, 577 colorectal SETs underwent ER, and 71 colorectal SETs underwent serial colonoscopy for > 12 months. Results: The mean tumor size (± standard deviation) of 577 colorectal SETs for which ER was performed was 7.0 ± 5.7 (median, 55; range, 1–50) mm; 475 tumors were located in the rectum and 102, in the colon. En bloc resection was achieved in 560/577 treated lesions (97.1%), and complete resection was achieved in 516/577 (89.4%). ER-related adverse events occurred in 15/577 (2.6%) patients. SETs originating from the muscularis propria showed a higher risk of ER-related adverse events and perforation than SETs arising from the mucosal or submucosal layer (odds ratio [OR] 19.786, 95% confidence interval [CI] 4.556–85.919; P = 0.002 and OR 141.250, 95% CI 11.596–1720.492; P = 0.046, respectively). Seventy-one patients were followed up after EUS without any treatment for > 12 months, during which three showed progression; eight, regression; and sixty, no changes. Conclusions: ER for colorectal SETs showed excellent efficacy and safety. Additionally, colorectal SETs without high-risk features in surveillance with colonoscopy showed an excellent prognosis. [ABSTRACT FROM AUTHOR]

Background and Aims: Endoscopic ultrasonography (EUS) is a useful diagnostic modality for evaluating gastric mesenchymal tumors; however, differentiating gastrointestinal stromal tumors (GISTs) from benign mesenchymal tumors such as leiomyomas and schwannomas remains challenging. For this reason, we developed a convolutional neural network computer-aided diagnosis (CNN-CAD) system that can analyze gastric mesenchymal tumors on EUS images. Methods: A total of 905 EUS images of gastric mesenchymal tumors (pathologically confirmed GIST, leiomyoma, and schwannoma) were used as a training dataset. Validation was performed using 212 EUS images of gastric mesenchymal tumors. This test dataset was interpreted by three experienced and three junior endoscopists. Results: The sensitivity, specificity, and accuracy of the CNN-CAD system for differentiating GISTs from non-GIST tumors were 83.0%, 75.5%, and 79.2%, respectively. Its diagnostic specificity and accuracy were significantly higher than those of two experienced and one junior endoscopists. In the further sequential analysis to differentiate leiomyoma from schwannoma in non-GIST tumors, the final diagnostic accuracy of the CNN-CAD system was 75.5%, which was significantly higher than that of two experienced and one junior endoscopists. Conclusions: Our CNN-CAD system showed high accuracy in diagnosing gastric mesenchymal tumors on EUS images. It may complement the current clinical practices in the EUS diagnosis of gastric mesenchymal tumors.

Duodenal neuroendocrine tumors (NETs) are rare subepithelial tumors that arise from the neuroendocrine cells beneath the epithelial layer. However, an accurate histopathological diagnosis is difficult when tissue samples are obtained using conventional endoscopic forceps biopsy alone. This study aimed to evaluate the magnifying endoscopy with narrow-band imaging (ME-NBI) findings of duodenal NETs. We retrospectively analyzed a database of 22 duodenal NETs from 21 patients who underwent ME-NBI between January 2011 and June 2022. The ME-NBI, endosonographic, and histopathologic findings of duodenal NETs were analyzed. Nineteen lesions were located in the bulb, two were located in the superior duodenal angle, and one was located in the second portion of the duodenum. Eighteen lesions (82%) had IIa morphology, and nine (41%) had central depression on the surface. On endoscopic ultrasonography, almost all lesions (20/22, 91%) were located in the second and/or third layers, and the median tumor size was 6 mm. During ME-NBI, the microsurface pattern was regular in 18 lesions (82%) and absent in 4 (18%). The microvascular pattern was regular in 17 lesions (77%), irregular in 4 (18%), and absent in 1 (5%). Thickened subepithelial vessels were observed in 15 (68%) lesions. There was no difference in tumor size according to the presence or absence of thickened subepithelial vessels (6.1 ± 1.8 mm vs. 5.9 ± 3.8 mm, p = 0.860). In conclusion, the characteristic ME-NBI findings of duodenal NETs were regular microsurface and microvascular patterns and the presence of thickened subepithelial vessels. These ME-NBI features may be useful for differentiating duodenal NETs from other duodenal subepithelial lesions. [ABSTRACT FROM AUTHOR]

Gastric mesenchymal tumors (GMTs) are incidentally discovered in national gastric screening programs in Korea. Endoscopic ultrasonography (EUS) is the most useful diagnostic modality for evaluating GMTs. The differentiation of gastrointestinal stromal tumors from benign mesenchymal tumors, such as schwannomas or leiomyomas, is important to ensure appropriate clinical management. However, this is difficult and operator dependent because of the subjective interpretation of EUS images. Digital image analysis computes the distribution and spatial variation of pixels using texture analysis to extract useful data, enabling the objective analysis of EUS images and decreasing interobserver and intraobserver agreement in EUS image interpretation. This review aimed to summarize the usefulness and future of digital EUS image analysis for GMTs based on published reports and our experience. [ABSTRACT FROM AUTHOR]

Objective. Predicting the invasion depth of superficial esophageal squamous cell carcinoma (SESCC) is important when selecting among therapeutic strategies. The aim of this study was to compare magnifying endoscopy with narrow-band imaging (ME-NBI) with endoscopic ultrasonography (EUS) for predicting the depth of tumor invasion in patients with SESCC. Methods. This study enrolled 51 patients with SESCC (52 SESCC lesions) who underwent both ME-NBI and EUS at Pusan National University Hospital during 2010-2013. We reviewed the patients' medical records and compared ME-NBI and EUS findings with histopathological results according to clinicopathological factors. Results. A total of 46 lesions in 45 patients were included in the final analysis. ME-NBI and EUS had overall accuracies of 76.1% and 84.8%, respectively, in distinguishing mucosal from non-mucosal cancers. There were no differences between ME-NBI and EUS in terms of sensitivities and specificities in distinguishing mucosal from non-mucosal cancers ( p = 0.500 and p = 0.688, respectively). When both ME-NBI and EUS suggested a mucosal depth of lesion invasion, the frequency of mucosal cancer in the final histopathology was 94%. However, if either ME-NBI or EUS suggested a non-mucosal depth of invasion, the frequency of mucosal cancer was only 21%. Conclusion. ME-NBI and EUS are accurate predictors of SESCC invasion depth. If both methods suggest a mucosal depth of lesion invasion, the accuracy of the prediction is increased. Therefore, when possible, it would be better to evaluate the invasion depth of SESCC using both ME-NBI and EUS before deciding to perform endoscopic resection. [ABSTRACT FROM AUTHOR]

During dissection, the tumor was tightly attached to the proper muscle layer; therefore, a clip attached with dental floss was deployed to secure the dissection plane between the tumor and muscle layer (Fig. Histopathologically, the cystic lesion was lined with columnar epithelial mucosa and had its own muscle layer, and ectopic pancreatic tissue of acinar structure was present in the cystic wall (Fig. 3 A clip attached with dental floss was deployed to secure the dissection plane between the tumor and the muscle layer. [Extracted from the article]

Background and Aims: Endoscopic ultrasonography (EUS) is a useful diagnostic modality for evaluating gastric mesenchymal tumors; however, differentiating gastrointestinal stromal tumors (GISTs) from benign mesenchymal tumors such as leiomyomas and schwannomas remains challenging. For this reason, we developed a convolutional neural network computer-aided diagnosis (CNN-CAD) system that can analyze gastric mesenchymal tumors on EUS images. Methods: A total of 905 EUS images of gastric mesenchymal tumors (pathologically confirmed GIST, leiomyoma, and schwannoma) were used as a training dataset. Validation was performed using 212 EUS images of gastric mesenchymal tumors. This test dataset was interpreted by three experienced and three junior endoscopists. Results: The sensitivity, specificity, and accuracy of the CNN-CAD system for differentiating GISTs from non-GIST tumors were 83.0%, 75.5%, and 79.2%, respectively. Its diagnostic specificity and accuracy were significantly higher than those of two experienced and one junior endoscopists. In the further sequential analysis to differentiate leiomyoma from schwannoma in non-GIST tumors, the final diagnostic accuracy of the CNN-CAD system was 75.5%, which was significantly higher than that of two experienced and one junior endoscopists. Conclusions: Our CNN-CAD system showed high accuracy in diagnosing gastric mesenchymal tumors on EUS images. It may complement the current clinical practices in the EUS diagnosis of gastric mesenchymal tumors. [ABSTRACT FROM AUTHOR]

Background/Aims The mural nodule is a key point of differential diagnosis of malignant intraductal papillary mucinous neoplasm (IPMN). As the consensus guideline of IPMN was revised in 2017, the criteria for the worrisome feature and high risk stigmata of the mural nodule were changed. The purpose of this study was to investigate the predictive rate of malignancy between 2012 and 2017 guideline. Methods We retrospectively analyzed patients who underwent surgery caused of cystic neoplasm from March 2011 to August 2018. A total of 61 patients were included. Among these patients, MCN and main duct type IPMN were excluded. Finally, total of 25 patients were included. Results Fourteen patients were male and the median age was 68 years old. Four patients were diagnosed as malignant IPMN. The median serum CA 19-9 was 8.5 IU/mL; carcinoembryonic antigen was 2.5 IU/mL; and median largest cyst size was 35 mm. Computed tomography (CT) and magnetic resonance imaging (MRI) were revealed CE mural nodules in three cases and non-CE mural nodule in four cases. Endoscopic ultrasound (EUS) was revealed mural nodule in seven cases. Total 11 patients have mural nodule that was proven by CT/MRI/EUS. Nine patients have a mural nodule 5 mm or more at EUS or CE mural nodule at CT/MRI. The sensitivity of malignant IPMN diagnosis was 100%; the specificity was 66.6%; and the positive predictive value was 36.4% in 2012 guideline. When we investigate the mural nodule based on the 2017 guideline; there were 100% of sensitivity; 76.2% of specificity and 44.4% of positive predictive value. Conclusions The 2017 revised guideline have slightly higher specificity and positive predictive value than those of the 2012 guideline for malignant IPMN diagnosis. The 2017 revised guideline was more accurate method to detect malignancy in terms of mural nodule. [ABSTRACT FROM AUTHOR]