Expert Endoscopists Research Articles

Development and validation of a simulation-based assessment tool in colonoscopy

BackgroundColonoscopy is difficult to learn. Virtual reality simulation training is helpful, but how and when novices should progress to patient-based training has yet to be established. To date, there is no assessment tool for credentialing novice endoscopists prior to clinical practice. The aim of this study was to develop such an assessment tool based on metrics provided by the simulator. The metrics used for the assessment tool should be able to discriminate between novices, intermediates, and experts and include essential checklist items for patient safety.MethodsThe validation process was conducted based on the Standards for Educational and Psychological Testing. An expert panel decided upon three essential checklist items for patient safety based on Lawshe’s method: perforation, hazardous tension to the bowel wall, and cecal intubation. A power calculation was performed. In this study, the Simbionix GI Mentor II simulator was used. Metrics with discriminatory ability were identified with variance analysis and combined to form an aggregate score. Based on this score and the essential items, pass/fail standards were set and reliability was tested.ResultsTwenty-four participants (eight novices, eight intermediates, and eight expert endoscopists) performed two simulated colonoscopies. Four metrics with discriminatory ability were identified. The aggregate score ranged from 4.2 to 51.2 points. Novices had a mean score of 10.00 (SD 5.13), intermediates 24.63 (SD 7.91), and experts 30.72 (SD 11.98). The difference in score between novices and the other two groups was statistically significant (p<0.01). Although expert endoscopists had a higher score, the difference was not statistically significant (p=0.40). Reliability was good (Cronbach’s alpha=0.86). A pass/fail score was defined at 17.1 points with correct completion of three essential checklist items, resulting in three experts and three intermediates failing and one novice passing the assessment.ConclusionWe established a valid and reliable assessment tool with a pass/fail standard on the simulator. We suggest using the assessment after simulation-based training before commencing work-based learning.

Accuracy of Narrow-Band Imaging International Colorectal Endoscopic Classification for Predicting the Histology of Colon Polyps by Experienced Endoscopists and Trainees.

Digital chromoendoscopy has proven to be useful in the histological prediction of premalignant lesions in the colon. The aim of the study was to describe the diagnostic performance of Narrow-Band Imaging International Colorectal Endoscopic Classification in the histological differentiation of colonic lesions, applied by expert endoscopists and trainees. Cross-sectional study that includes high-definition endoscopic images and histopathological reports of 94 patients over 50 years. Images were evaluated and classified as Narrow-Band Imaging International Colorectal Endoscopic 1, 2, or 3 by 2 experts and 2 trainee endoscopists, all of them blinded to histological results. Diagnostic accuracy for each Narrow-Band Imaging International Colorectal Endoscopic category was calculated for trainees and expert endoscopists. Intra-observer agreement was evaluated by means of Cohen's kappa coefficient; meanwhile, inter-observer agreement was calculated by means of Fleiss' kappa. Evaluations performed by expert and trainee endoscopists showed a performance for Narrow-Band Imaging International Colorectal Endoscopic category 1: sensitivity 62%, specificity 85%, area under receiver operator characteristic 0.73; Narrow-Band Imaging International Colorectal Endoscopic category 2: sensitivity 61%, specificity 73%, area under receiver operator characteristic 0.66; and Narrow-Band Imaging International Colorectal Endoscopic category 3: sensitivity 88%, specificity 91%, area under receiver operator characteristic 0.86. The total agreement of the evaluations was 72.5%, with an inter-observer variability of K 0.60 (95% CI 0.52-0.74). When the diagnostic performance for non-dysplastic lesions and dysplastic lesions (Narrow-Band Imaging International Colorectal Endoscopic 1 vs 2 and 3) was compared, we observed an increase in sensitivity for differentiated adenomas (Narrow-Band Imaging International Colorectal Endoscopic 2). Narrow-Band Imaging International Colorectal Endoscopic Classification applied in the histological prediction of static images of colonic lesions has a good diagnostic performance for Narrow-Band Imaging International Colorectal Endoscopic category 3, as well as an acceptable performance for Narrow-Band Imaging International Colorectal Endoscopic category 1, with a moderate agreement among observers.

A real-time interpretable artificial intelligence model for thecholangioscopic diagnosis of malignant biliary stricture (with videos).

It is crucial to accurately determine malignant biliary strictures (MBSs) for early curative treatment. This study aimed to develop a real-time interpretable artificial intelligence (AI) system to predict MBSs under digital single-operator cholangioscopy (DSOC). A novel interpretable AI system called MBSDeiT was developed consisting of 2 models to identify qualified images and then predict MBSs in real time. The overall efficiency of MBSDeiT was validated at the image level on internal, external, and prospective testing data sets and subgroup analyses, and at the video level on the prospective data sets; these findings were compared with those of the endoscopists. The association between AI predictions and endoscopic features was evaluated to increase the interpretability. MBSDeiT can first automatically select qualified DSOC images with an area under the curve (AUC) of .963 and .968 to .973 on the internal testing data set and the external testing data sets, and then identify MBSs with an AUC of .971 on the internal testing data set, an AUC of .978 to .999 on the external testing data sets, and an AUC of .976 on the prospective testing data set, respectively. MBSDeiT accurately identified 92.3% of MBSs in prospective testing videos. Subgroup analyses confirmed the stability and robustness of MBSDeiT. The AI system achieved superior performance to that of expert and novice endoscopists. The AI predictions were significantly associated with 4 endoscopic features (nodular mass, friability, raised intraductal lesion, and abnormal vessels; P< .05) under DSOC, which is consistent with the endoscopists' predictions. The study findings suggest that MBSDeiT could be a promising approach for the accurate diagnosis of MBSs under DSOC.

Essential Surgical Techniques During Fully Endoscopic Spine Surgery: A Historical Review and Future Perspectives

Full-endoscopic spine surgery (FESS) offers minimally invasive techniques for treating spinal lesions with smaller incisions, leading to reduced tissue damage and consequently more rapid patient recovery. The essential techniques in FESS —namely, the transforaminal and interlaminar approaches—have evolved with technological advances in endoscopic systems and surgeon expertise. The transforaminal approach consists of “inside-out” and “outside-in” techniques that have distinct applications depending on the patient’s condition. The interlaminar approach, similar to traditional spinal surgery techniques, is gaining wider acceptance due to its efficacy in addressing unique anatomical challenges. Advanced procedures, such as endoscopic laminectomy and interbody fusion surgery, coupled with navigation systems, have broadened the scope of endoscopic spine surgery, making it an increasingly vital tool in modern spinal surgery.

Computer-aided diagnosis in predicting the invasion depth of early colorectal cancer: a systematic review and meta-analysis of diagnostic test accuracy.

Endoscopic resection (ER) is widely applied to treat early colorectal cancer (CRC). Predicting the invasion depth of early CRC is critical in determining treatment strategies. The use of computer-aided diagnosis (CAD) algorithms could theoretically make accurate and objective predictions regarding the suitability of lesions for ER indication based on invasion depth. This study aimed to assess diagnostic test accuracy of CAD algorithms in predicting the invasion depth of early CRC and to compare the performance between the CAD algorithms and endoscopists. Multiple databases were searched until June 30, 2022 for studies that evaluated the diagnostic performance of CAD algorithms for invasion depth of CRC. Meta-analysis of diagnostic test accuracy using a bivariate mixed-effects model was performed. Ten studies consisting of 13 arms (13,918 images from 1472 lesions) were included. Due to significant heterogeneity, studies were stratified into Japan/Korea-based or China-based studies. For the former, the area under the curve (AUC), sensitivity, and specificity of the CAD algorithms were 0.89 (95% CI 0.86-0.91), 62% (95% CI 50-72%), and 96% (95% CI 93-98%), respectively. For the latter, AUC, sensitivity, and specificity were 0.94 (95% CI 0.92-0.96), 88% (95% CI 78-94%), and 88% (95% CI 80-93%), respectively. The performance of the CAD algorithms in Japan/Korea-based studies was not significantly different from that of all endoscopists (0.88 vs. 0.91, P = 0.10) but was inferior to that of expert endoscopists (0.88 vs. 0.92, P = 0.03). The performance of the CAD algorithms in China-based studies was better than that of all endoscopists (0.94 vs. 0.90, P = 0.01). The CAD algorithms showed comparable accuracy for prediction of invasion depth of early CRC compared to all endoscopists, which was still lower than expert endoscopists in diagnostic accuracy; more improvements should be achieved before it can be extensively applied to clinical practice.

New Perspectives in Endoscopic Treatment of Gastroesophageal Reflux Disease

Gastroesophageal reflux disease has a high incidence and prevalence in the general population. Clinical manifestations are heterogenous, and so is the response to medical treatment. Proton pump inhibitors are still the most common agents used to control reflux symptoms and for healing esophagitis, but they are not a one-size-fits-all solution for the disease. Patients with persistent troublesome symptoms despite medical therapy, those experiencing some adverse drug reaction, or those unwilling to take lifelong medications deserve valid alternatives. Anti-reflux Nissen fundoplication is an effective option, but the risk of adverse events has limited its spread. In recent years, advancements in therapeutic endoscopy have been made, and three major endoluminal alternatives are now available, including (1) the delivery of radiofrequency energy to the esophago–gastric junction, (2) transoral incisionless fundoplication (TIF), and (3) anti-reflux mucosal interventions (ARMI) based on mucosal resection (ARMS) and mucosal ablation (ARMA) techniques to remodel the cardia. Endoscopic techniques have shown interesting results, but their diffusion is still limited to expert endoscopists in tertiary centers. This review discusses the state of the art in the endoscopic approach to gastroesophageal reflux disease.

Postcolonoscopy colorectal cancer: Prevalence, categorization and root-cause analysis based on the World Endoscopic Organization system

AimsThe World Endoscopy Organization (WEO) recommends that endoscopy units implement a process to identify postcolonoscopy colorectal cancer (PCCRC). The aims of this study were to assess the 3-year PCCRC rate and to perform root-cause analyses and categorization in accordance with the WEO recommendations. Patients and methodsCases of colorectal cancers (CRCs) in a tertiary care center were retrospectively included from January 2018 to December 2019. The 3-year and 4-year PCCRC rates were calculated. A root-cause analysis and categorization of PCCRCs (interval and type A, B, C noninterval PCCRCs) were performed. The level of agreement between two expert endoscopists was assessed. ResultsA total of 530 cases of CRC were included. A total of 33 were deemed PCCRCs (age 75.8±9.5 years; 51.5% women). The 3-year and 4-year PCCRC rates were 3.4% and 4.7%, respectively. The level of agreement between the two endoscopists was acceptable either for the root-cause analysis (k=0.958) or for the categorization (k=0.76).The most plausible explanations of the PCCRCs were 8 “likely new PCCRCs”, 1 (4%) “detected, not resected”, 3 (12%) “detected, incomplete resection”, 8 (32%) “missed lesion, inadequate examination”, and 13 (52%) “missed lesion, adequate examination”. Most PCCRCs were deemed noninterval Type C PCCRCs (N=17, 51.5%). ConclusionWEO recommendations for root-cause analysis and categorization are useful to detect areas for improvement. Most PCCRCs were avoidable and were likely due to missed lesions during an otherwise adequate examination.

Comparison of Texture and Color Enhancement Imaging with White Light Imaging in 52 Patients with Short-Segment Barrett's Esophagus.

BACKGROUND Texture and color enhancement imaging (TXI), a new type of image-enhanced endoscopy, may improve the detection of gastrointestinal lesions. Barrett's esophagus (BE) requires an accurate diagnosis since it may undergo neoplastic transformation. We aimed to evaluate the usefulness of TXI compared with white light imaging (WLI) in BE. MATERIAL AND METHODS In this prospective study at a single hospital from February 2021 to February 2022, we enrolled 52 consecutive patients with BE. Endoscopic images of BE using WLI, TXI mode 1 (TXI-1), TXI mode 2 (TXI-2), and narrow-band imaging (NBI) were compared by 10 endoscopists (5 experts and 5 trainees). Endoscopists scored visibility for the images as follows: 5 (improved), 4 (somewhat improved), 3 (equivalent), 2 (somewhat decreased), and 1 (decreased). Total visibility scores for all 10 endoscopists, and subgroups composed of the 5 expert endoscopists and the 5 trainee endoscopists, were evaluated. Main-group (10 endoscopists) scores of ≥40, 21-39, and ≤20, and subgroup (5 endoscopists) scores of ≥20, 11-19, and ≤10, were considered "improved", "equivalent", and "decreased", respectively. Inter-rater reliability (intra-class correlation coefficient [ICC]) was calculated and images were objectively assessed based on L*a*b* color values and color differences (ΔE*). RESULTS All 52 cases were diagnosed as short-segment BE (SSBE). TXI-1/TXI-2 improved visibility compared with WLI was: 78.8%/32.7% for all endoscopists; 82.7%/40.4% for trainees; and 76.9%/34.6% for experts. NBI did not improve visibility. The ICC for TXI-1 and TXI-2 compared with WLI was "excellent" for all endoscopists. The ΔE* between esophageal and Barrett's mucosae, and between Barrett's and gastric mucosae, was higher for TXI-1 than for WLI (P<0.01, P<0.05, respectively). CONCLUSIONS TXI, especially TXI-1, improves the endoscopic diagnosis of SSBE compared with WLI, regardless of the endoscopist's skill.

Endoscopic clearance of non-complex biliary stones using fluoroscopy-free direct solitary cholangioscopy: Initial multicenter experience.

Fluoroscopy-free endoscopic retrograde cholangiopancreatography for common bile duct stone (CBDS) clearance is usually offered only to pregnant patients. We initiated a multicenter, randomized controlled trial comparing clearance of non-complex CBDSs using fluoroscopy-free direct solitary cholangioscopy (DSC) to standard endoscopic retrograde cholangiography (ERC) to evaluate the wider applicability of the DSC-based approach. Here we report the initial results of stone clearance and safety in roll-in cases for the randomized controlled trial. Twelve expert endoscopists at tertiary care centers in four countries prospectively enrolled 47 patients with non-complex CBDSs for DSC-assisted CBDS removal in an index procedure including fluoroscopy-free cannulation. Successful CBDS clearance was first determined by DSC and subsequently validated by final occlusion cholangiogram as the ERC gold standard. Fully fluoroscopy-free cannulation was successful in 42/47 (89.4%) patients. Brief fluoroscopy with minimal contrast injection was used in 4/47 (8.5%) patients during cannulation. Cannulation failed in 1/47 (2.1%) patients. Fluoroscopy-free complete stone clearance was reached in 38/46 (82.6%) cases. Residual stones were detected in the validation ERC occlusion cholangiogram in three cases. Overall serious adverse event rate was 2.1% (95% confidence interval 0.1-11.3): postprocedural pancreatitis in one patient. In patients with non-complex CBDS, the fluoroscopy-free technique is easily transferred to endoscopic retrograde cholangiopancreatography experts with acceptable rates of cannulation and stone clearance and few serious adverse events. (ClinicalTrials.gov number, NCT03421340).

Diagnosis of nasopharyngeal carcinoma with convolutional neural network on narrowband imaging

Objective:To evaluate the diagnostic accuracy of the convolutional neural network（CNN） in diagnosing nasopharyngeal carcinoma using endoscopic narrowband imaging. Methods:A total of 834 cases with nasopharyngeal lesions were collected from the People's Hospital of Guangxi Zhuang Autonomous Region between 2014 and 2016. We trained the DenseNet201 model to classify the endoscopic images, evaluated its performance using the test dataset, and compared the results with those of two independent endoscopic experts. Results:The area under the ROC curve of the CNN in diagnosing nasopharyngeal carcinoma was 0.98. The sensitivity and specificity of the CNN were 91.90% and 94.69%, respectively. The sensitivity of the two expert-based assessment was 92.08% and 91.06%, respectively, and the specificity was 95.58% and 92.79%, respectively. There was no significant difference between the diagnostic accuracy of CNN and the expert-based assessment （P=0.282, P=0.085）. Moreover, there was no significant difference in the accuracy in discriminating early-stage and late-stage nasopharyngeal carcinoma（P=0.382）. The CNN model could rapidly distinguish nasopharyngeal carcinoma from benign lesions, with an image recognition time of 0.1 s/piece. Conclusion:The CNN model can quickly distinguish nasopharyngeal carcinoma from benign nasopharyngeal lesions, which can aid endoscopists in diagnosing nasopharyngeal lesions and reduce the rate of nasopharyngeal biopsy.

Diagnostic validity and learning curve of non-NBI expert endoscopists in gastric intestinal metaplasia diagnosis.

Endoscopists' experience influences narrow-band imaging (NBI)-guided gastric intestinal metaplasia (GIM) diagnostic performance. We aimed to evaluate the general gastroenterologists (GE) performance in NBI-guided GIM diagnosis compared to NBI experts (XP) and assess GEs' learning curve. A cross-sectional study was conducted between 10/2019 and 2/2022. Histology-proven GIM who underwent esophagogastroduodenoscopy (EGD) were randomly assessed by 2XPs or 3GEs. Endoscopists' performance on NBI-guided diagnoses were compared to the pathological diagnosis (gold standard) in five areas of the stomach according to the Sydney protocol. The primary outcome were GIM diagnosis validity scores of GEs compared to XPs. The secondary outcome was the minimum number of lesions required for GEs to achieve an accuracy of GIM diagnosis ≥ 80%. One thousand one hundred and fifty-five lesions from 189 patients (51.3% male, mean age 66 ± 10years) were examined. GEs performed EGD in 128 patients with 690 lesions. the GIM diagnosis sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of GEs compared to the XPs, were 91% vs.93%, 73% vs.83%, 79% vs.83%, 89% vs.93%, and 83% vs.88%, respectively. GEs demonstrated lower specificity (mean difference - 9.4%; 95%CI - 16.3, 1.4; p = 0.008) and accuracy (mean difference - 5.1%; 95%CI - 3.3, 6.3; p = 0.006) compared to XPs. After 100 lesions (50% GIM), GEs achieved an accuracy of ≥ 80% and all diagnostic validity scores were comparable to the XPs (p < 0.05 all). Compared to XPs, GEs had lower specificity and accuracy for GIM diagnosis. The learning curve for a GE to achieve comparable performance to XPs would necessitate at least 50 GIM lesions. Created with BioRender.com.

High quality colonoscopy: using textbook process as a composite quality measure.

BACKGROUND : High quality colonoscopy is fundamental to good patient outcomes. "Textbook outcome" has proven to be a feasible multidimensional measure for quality assurance between surgical centers. In this study, we sought to establish the "textbook process" (TP) as a new composite measure for the optimal colonoscopy process and assessed how frequently TP was attained in clinical practice and the variation in TP between endoscopists. METHODS : To reach consensus on the definition of TP, international expert endoscopists completed a modified Delphi consensus process. The achievement of TP was then applied to clinical practice. Prospectively collected data in two endoscopy services were retrospectively evaluated. Data on colonoscopies performed for symptoms or surveillance between 1 January 2018 and 1 August 2021 were analyzed. RESULTS : The Delphi consensus process was completed by 20 of 27 invited experts (74.1 %). TP was defined as a colonoscopy fulfilling the following items: explicit colonoscopy indication; successful cecal intubation; adequate bowel preparation; adequate withdrawal time; acceptable patient comfort score; provision of post-polypectomy surveillance recommendations in line with guidelines; and the absence of the use of reversal agents, early adverse events, readmission, and mortality. In the two endoscopy services studied, TP was achieved in 5962/8227 colonoscopies (72.5 %). Of 48 endoscopists performing colonoscopy, attainment of TP varied significantly, ranging per endoscopist from 41.0 % to 89.1 %. CONCLUSION : This study proposes a new composite measure for colonoscopy, namely "textbook process." TP gives a comprehensive summary of performance and demonstrates significant variation between endoscopists, illustrating the potential benefit of TP as a measure in future quality assessment programs.

Usefulness of a new polyvinyl alcohol hydrogel (PVA-H)-based simulator for endoscopic submucosal dissection training: a pilot study.

We developed a new endoscopic submucosal dissection (ESD) simulator and evaluated its efficacy and realism for use training endoscopists. An ESD simulator was constructed using polyvinyl alcohol hydrogel sheets and compared to a previous ESD simulator. Between March 1, 2020, and December 30, 2021, eight expert endoscopists from three different centers analyzed the procedure-related factors of the simulator. Five trainees performed gastric ESD exercises under the guidance of these experts. Although the two ESD simulators provided overall favorable outcomes in terms of ESD-related factors, the new simulator had several benefits, including better marking of the target lesion's limits (p<0.001) and overall handling (p<0.001). Trainees tested the usefulness of the new ESD simulator. The complete resection rate improved after 3 ESD training sessions (9 procedures), and the perforation rate decreased after 4 sessions (12 procedures). We have developed a new ESD simulator that can help beginners achieve a high level of technical experience before performing real-time ESD procedures in patients.

Anatomical endoscopic retrograde cholangiopancreatography simulator using moulded meshed silicone: A novel simulator pilot study.

Endoscopic retrograde cholangio-pancreatography (ERCP) has higher rates of morbidity and mortality compared to upper or lower gastrointestinal tract endoscopy. The availability of magnetic resonance cholangiopancreatography means ERCP is usually performed for therapeutic purposes. Simulation could provide an adjunct to patient-based training in ERCP however models to date have been unconvincing. This ERCP simulation model was constructed from moulded meshed silicone by co-designers: Jean Wong and Kai Cheng. The anatomical orientation was based on a combination of anatomical specimens, sectional atlases, and the clinical experience of expert endoscopists. From March to October 2022, we recruited 5 surgeons/gastroenterologists to the expert group and 14 medical students, junior doctors, or surgical/gastroenterological trainees to the novice group. Most experts either agreed or strongly agreed that the simulation anatomy appearance (100%), anatomical orientation (83%), tactile feedback (66%), traversal actions (67%), cannula positioning (66%) and papilla cannulation (67%) resembled the procedure in humans. Experts statistically significantly outperformed novices in obtaining a cannulating position (80% vs. 14%, P = 0.006) and successful papilla cannulation (80% vs. 7%, P = 0.0015) on their first attempt. The novice group had statistically significant improvements in time to obtaining a cannulating position (3.53 vs. 11.5 min, P = 0.006) and passing the duodenoscope to the papilla (2.55 vs. 4 passes, P = 0.009). The simulator showed statistically significant results in face, content, and construct validity. A follow-up validation study should recruit participants across multiple institutions. External validity could be assessed by comparing expert proceduralist simulator performance against clinical ERCP performance.

Hemorrhoids as a risk factor for colorectal adenomas on colonoscopy

Background and study aims Colorectal premalignant polyps and hemorrhoids are important findings in colonoscopy; however, the association between them is unclear. Therefore, we investigated the association between the presence and severity of hemorrhoids and the detection of precancerous colorectal polyps on colonoscopy. Patients and methods This retrospective, single-center, cross-sectional study enrolled patients who underwent colonoscopy at Toyoshima Endoscopy Clinic between May 2017 and October 2020. The association between hemorrhoids and other outcomes (patient age, sex, withdrawal time for colonoscopy, expert endoscopist, number of adenomas per colonoscopy, detection rates of adenoma, advanced neoplasia, clinically significant serrated polyp, and sessile serrated lesion) was assessed using a binomial logistic regression model. Results A total of 12,408 patients were enrolled in this study. Hemorrhoids were identified in 1,863 patients. Univariable analysis showed that patients with hemorrhoids were older (61.0 vs. 52.5 years, P < 0.001), had a higher number of adenomas per colonoscopy (1.16 vs. 0.756, P < 0.001) than those without hemorrhoids. Multivariable analyses also demonstrated that hemorrhoids were associated with a higher number of adenomas per colonoscopy (odds ratio [OR]: 1.061; P = 0.002), regardless of patient age, sex, and expert endoscopist. Among patients with hemorrhoids, severe hemorrhoids with a mucosal elevation ≥ 10 mm were associated with a higher number of adenomas per colonoscopy than mild hemorrhoids (OR: 1.112, P = 0.044), regardless of patient age, sex, and expert endoscopist. Conclusions Hemorrhoids, especially severe ones, are associated with a high number of adenomas. Complete colonoscopy should be performed in patients with hemorrhoids.

Automatic textual description of colorectal polyp features: explainable artificial intelligence

Computer-aided diagnosis systems (CADx) can improve colorectal polyp (CRP) optical diagnosis. For integration into clinical practice, better understanding of artificial intelligence (AI) by endoscopists is needed. We aimed to develop an explainable AI CADx capable of automatically generating textual descriptions of CRPs. For training and testing of this CADx, textual descriptions of CRP size and features according to the Blue Light Imaging (BLI) Adenoma Serrated International Classification (BASIC) were used, describing CRP surface, pit pattern, and vessels. CADx was tested using BLI images of 55 CRPs. Reference descriptions with agreement by at least five out of six expert endoscopists were used as gold standard. CADx performance was analyzed by calculating agreement between the CADx generated descriptions and reference descriptions. CADx development for automatic textual description of CRP features succeeded. Gwet's AC1 values comparing the reference and generated descriptions per CRP feature were: size 0.496, surface-mucus 0.930, surface-regularity 0.926, surface-depression 0.940, pits-features 0.921, pits-type 0.957, pits-distribution 0.167, and vessels 0.778. CADx performance differed per CRP feature and was particularly high for surface descriptors while size and pits-distribution description need improvement. Explainable AI can help comprehend reasoning behind CADx diagnoses and therefore facilitate integration into clinical practice and increase trust in AI.

Evaluation of a New Animal Tissue-Free Bleeding Model for Training of Endoscopic Hemostasis.

For endoscopists, knowledge of the available hemotherapeutic devices and materials as well as competence in using them is a life-saving expertise in the treatment of patients with acute gastrointestinal bleeding. These competences can be acquired in training on live animals, animal organs, or simulators. We present an animal tissue-free training model of the upper gastrointestinal tract for bleeding therapy. An artificial, animal tissue-free mucosa and submucosa with the opportunity of injection and clipping therapy were created first. Patches with this artificial mucosa and submucosa were placed into silicone and latex organs with human-like anatomy. Esophageal bleeding situations were imitated as variceal bleeding and bleeding of a reflux esophagitis in latex organs. Finally, a modular training model with human anatomy and replaceable bleeding sources was created. Evaluation of the novel model for gastroscopic training was performed in a multicentric setting with endoscopic beginners and experts. Evaluation was carried out by 38 physicians with different levels of education in endoscopy. Evaluation of the model was made with grades from one (excellent) to six (bad): suitability for endoscopic training was 1.4, relevance of the endoscopic training was 1.6, and grading for haptic and optic impression of the model was 1.7. The creation of a gastroscopic model for the training of hemostatic techniques without animal tissues was possible and multiple endoscopic bleeding skills could be trained in it. Evaluation showed good results for this new training option, which could be used in every endoscopic unit or other places without hygienic doubts.

Usefulness of texture and color enhancement imaging (TXI) in early gastric cancer found after Helicobacter pylori eradication

Early-stage gastric cancer (EGC) found after Helicobacter pylori (Hp) eradication is often difficult to diagnose using conventional white light (WL) endoscopy. We aimed to evaluate whether Texture and Color Enhancement Imaging (TXI), a new image-enhanced endoscopy enhances the EGC lesions after Hp eradication. We also compared diagnostic accuracy and lesion detection time between WL and TXI in trainee endoscopists. 58 EGC lesions after successful Hp eradication were enrolled. Using endoscopic images in WLI, TXI mode 1 (TXI1), and TXI mode 2 (TXI2), visibility of EGC was assessed by six expert endoscopists using a subjective score. Mean color differences (ΔE) of four matched adjacent and intra-tumoral points were examined. Using randomly allocated images, diagnostic accuracy and lesion detection time were evaluated in three trainee endoscopists. Visibility score was unchanged (Score 0) in 20.7% (12/58) and 45.6% (26/57), slightly improved (Score 1) in 60.3% (35/58) and 52.6% (30/57), obviously improved (Score 2) in 45.6% (26/58) and 1.8% (1/57), in TXI1 and TXI2 compared to WL, respectively. Mean ΔE ± SEM in TXI1 (22.90 ± 0.96), and TXI2 (15.32 ± 0.71) were higher than that in WL (1.88 ± 0.26, both P < 0.0001). TXI1 presented higher diagnostic accuracy compared to WL, in two of three trainees (94.8% vs. 74.1%, 100% vs. 89.7%, P = 0.003; < 0.005, respectively). Lesion detection time was shorter in TXI1 in two of three trainees (P = 0.006, 0.004, respectively) compared to WL. TXI improves visibility of EGC after Hp eradication that may contribute to correct diagnosis.

Bariatric endoscopy-Possibilities and value in obesity medicine and surgery-A separate special field

Endoscopy has become established as an important component in all preoperative, intraoperative and postoperative areas of bariatric surgery. As guidelines are not yet available, relevant studies and international position papers have been collected in this article. Preoperatively, endoscopy should be performed before sleeve gastrectomy (SG, detection of reflux status) and before bypass procedures (status survey of the postoperatively no longer accessible distal stomach). Routine endoscopic monitoring should be performed 2-3years after SG and aOAGB (one anastomosis gastric bypass) for recording of the gastroesophageal reflex disease (GERD) status. In the first line complication management of hemorrhage and leakages, endoscopic procedures have prevailed over surgical ones due to more causal therapeutic approaches (except cases of general peritonitis where lavage and drainages are needed). In the treatment of late complications such as dumping, anastomosis dilatation and stenosis, bariatric endoscopy provides versatile, organ-preserving solutions and continues to demonstrate ahigh degree of innovation. In these postoperative fields of application, especially in the treatment of late complications, specific knowledge of bariatric anatomy and physiology as well as surgical treatment options are required in addition to endoscopic expertise. The special field of bariatric endoscopy has developed and will have an own importance closely linked to the bariatricsurgery.

Endoscopic techniques for gastric neuroendocrine tumors: An update.

Gastric neuroendocrine neoplasms (gNENs) are a rare type of gastric neoplasm, even if their frequency is increasing according to the latest epidemiologic revisions of the main registries worldwide. They are divided into three main subtypes, with different pathogeneses, biological behaviors, and clinical characteristics. GNEN heterogeneity poses challenges, therefore these neoplasms require different management strategies. Update the knowledge on the endoscopic treatment options to manage g-NENs. This manuscript is a narrative review of the literature. In recent years, many advances have been made not only in the knowledge of both the pathogenesis and the molecular profiling of gNENs but also in the endoscopic expertise towards innovative treatment options, which proved to be less aggressive without losing the capability of being radical. The endoscopic approach is increasingly applied in the field of gastrointestinal (GI) luminal neoplasms, and this is true not only for adenocarcinomas but also for gNENs. In particular, different techniques have been described for the endoscopic removal of suspected lesions, ranging from classical polypectomy (cold or hot snare) to endoscopic mucosal resection (both with "en bloc" or piecemeal technique), endoscopic submucosal dissection, and endoscopic full-thickness resection. GNENs comprise different subtypes of neoplasms with distinct management and prognosis. New endoscopic techniques offer a wide variety of approaches for GI localized neoplasms, which demonstrated to be appropriate and effective also in the case of gNENs. Correct evaluation of size, site, morphology, and clinical context allows the choice of tailored therapy in order to guarantee a definitive treatment.