@article{EbigboMendelScheppachetal., author = {Ebigbo, Alanna and Mendel, Robert and Scheppach, Markus W. and Probst, Andreas and Shahidi, Neal and Prinz, Friederike and Fleischmann, Carola and R{\"o}mmele, Christoph and G{\"o}lder, Stefan Karl and Braun, Georg and Rauber, David and R{\"u}ckert, Tobias and Souza Jr., Luis Antonio de and Papa, Jo{\~a}o Paulo and Byrne, Michael F. and Palm, Christoph and Messmann, Helmut}, title = {Vessel and tissue recognition during third-space endoscopy using a deep learning algorithm}, series = {Gut}, volume = {71}, journal = {Gut}, number = {12}, publisher = {BMJ}, address = {London}, doi = {10.1136/gutjnl-2021-326470}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:898-opus4-54293}, pages = {2388 -- 2390}, abstract = {In this study, we aimed to develop an artificial intelligence clinical decision support solution to mitigate operator-dependent limitations during complex endoscopic procedures such as endoscopic submucosal dissection and peroral endoscopic myotomy, for example, bleeding and perforation. A DeepLabv3-based model was trained to delineate vessels, tissue structures and instruments on endoscopic still images from such procedures. The mean cross-validated Intersection over Union and Dice Score were 63\% and 76\%, respectively. Applied to standardised video clips from third-space endoscopic procedures, the algorithm showed a mean vessel detection rate of 85\% with a false-positive rate of 0.75/min. These performance statistics suggest a potential clinical benefit for procedure safety, time and also training.}, language = {en} } @article{MeinikheimMendelPalmetal., author = {Meinikheim, Michael and Mendel, Robert and Palm, Christoph and Probst, Andreas and Muzalyova, Anna and Scheppach, Markus W. and Nagl, Sandra and Schnoy, Elisabeth and R{\"o}mmele, Christoph and Schulz, Dominik Andreas Helmut Otto and Schlottmann, Jakob and Prinz, Friederike and Rauber, David and R{\"u}ckert, Tobias and Matsumura, Tomoaki and Fern{\´a}ndez-Esparrach, Gl{\`o}ria and Parsa, Nasim and Byrne, Michael F. and Messmann, Helmut and Ebigbo, Alanna}, title = {Influence of artificial intelligence on the diagnostic performance of endoscopists in the assessment of Barrett's esophagus: a tandem randomized and video trial}, series = {Endoscopy}, volume = {56}, journal = {Endoscopy}, publisher = {Georg Thieme Verlag}, address = {Stuttgart}, doi = {10.1055/a-2296-5696}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:898-opus4-72818}, pages = {641 -- 649}, abstract = {Background This study evaluated the effect of an artificial intelligence (AI)-based clinical decision support system on the performance and diagnostic confidence of endoscopists in their assessment of Barrett's esophagus (BE). Methods 96 standardized endoscopy videos were assessed by 22 endoscopists with varying degrees of BE experience from 12 centers. Assessment was randomized into two video sets: group A (review first without AI and second with AI) and group B (review first with AI and second without AI). Endoscopists were required to evaluate each video for the presence of Barrett's esophagus-related neoplasia (BERN) and then decide on a spot for a targeted biopsy. After the second assessment, they were allowed to change their clinical decision and confidence level. Results AI had a stand-alone sensitivity, specificity, and accuracy of 92.2\%, 68.9\%, and 81.3\%, respectively. Without AI, BE experts had an overall sensitivity, specificity, and accuracy of 83.3\%, 58.1\%, and 71.5\%, respectively. With AI, BE nonexperts showed a significant improvement in sensitivity and specificity when videos were assessed a second time with AI (sensitivity 69.8\% [95\%CI 65.2\%-74.2\%] to 78.0\% [95\%CI 74.0\%-82.0\%]; specificity 67.3\% [95\%CI 62.5\%-72.2\%] to 72.7\% [95\%CI 68.2\%-77.3\%]). In addition, the diagnostic confidence of BE nonexperts improved significantly with AI. Conclusion BE nonexperts benefitted significantly from additional AI. BE experts and nonexperts remained significantly below the stand-alone performance of AI, suggesting that there may be other factors influencing endoscopists' decisions to follow or discard AI advice.}, language = {en} } @misc{EbigboRauberAyoubetal., author = {Ebigbo, Alanna and Rauber, David and Ayoub, Mousa and Birzle, Lisa and Matsumura, Tomoaki and Probst, Andreas and Steinbr{\"u}ck, Ingo and Nagl, Sandra and R{\"o}mmele, Christoph and Meinikheim, Michael and Scheppach, Markus W. and Palm, Christoph and Messmann, Helmut}, title = {Early Esophageal Cancer and the Generalizability of Artificial Intelligence}, series = {Endoscopy}, volume = {56}, journal = {Endoscopy}, number = {S 02}, publisher = {Thieme}, address = {Stuttgart}, doi = {10.1055/s-0044-1783775}, pages = {S428}, abstract = {Aims Artificial Intelligence (AI) systems in gastrointestinal endoscopy are narrow because they are trained to solve only one specific task. Unlike Narrow-AI, general AI systems may be able to solve multiple and unrelated tasks. We aimed to understand whether an AI system trained to detect, characterize, and segment early Barrett's neoplasia (Barrett's AI) is only capable of detecting this pathology or can also detect and segment other diseases like early squamous cell cancer (SCC). Methods 120 white light (WL) and narrow-band endoscopic images (NBI) from 60 patients (1 WL and 1 NBI image per patient) were extracted from the endoscopic database of the University Hospital Augsburg. Images were annotated by three expert endoscopists with extensive experience in the diagnosis and endoscopic resection of early esophageal neoplasias. An AI system based on DeepLabV3+architecture dedicated to early Barrett's neoplasia was tested on these images. The AI system was neither trained with SCC images nor had it seen the test images prior to evaluation. The overlap between the three expert annotations („expert-agreement") was the ground truth for evaluating AI performance. Results Barrett's AI detected early SCC with a mean intersection over reference (IoR) of 92\% when at least 1 pixel of the AI prediction overlapped with the expert-agreement. When the threshold was increased to 5\%, 10\%, and 20\% overlap with the expert-agreement, the IoR was 88\%, 85\% and 82\%, respectively. The mean Intersection Over Union (IoU) - a metric according to segmentation quality between the AI prediction and the expert-agreement - was 0.45. The mean expert IoU as a measure of agreement between the three experts was 0.60. Conclusions In the context of this pilot study, the predictions of SCC by a Barrett's dedicated AI showed some overlap to the expert-agreement. Therefore, features learned from Barrett's cancer-related training might be helpful also for SCC prediction. Our results allow different possible explanations. On the one hand, some Barrett's cancer features generalize toward the related task of assessing early SCC. On the other hand, the Barrett's AI is less specific to Barrett's cancer than a general predictor of pathological tissue. However, we expect to enhance the detection quality significantly by extending the training to SCC-specific data. The insight of this study opens the way towards a transfer learning approach for more efficient training of AI to solve tasks in other domains.}, language = {en} } @misc{ScheppachMendelRauberetal., author = {Scheppach, Markus W. and Mendel, Robert and Rauber, David and Probst, Andreas and Nagl, Sandra and R{\"o}mmele, Christoph and Meinikheim, Michael and Palm, Christoph and Messmann, Helmut and Ebigbo, Alanna}, title = {Artificial Intelligence (AI) improves endoscopists' vessel detection during endoscopic submucosal dissection (ESD)}, series = {Endoscopy}, volume = {56}, journal = {Endoscopy}, number = {S 02}, publisher = {Thieme}, address = {Stuttgart}, doi = {10.1055/s-0044-1782891}, pages = {S93}, abstract = {Aims While AI has been successfully implemented in detecting and characterizing colonic polyps, its role in therapeutic endoscopy remains to be elucidated. Especially third space endoscopy procedures like ESD and peroral endoscopic myotomy (POEM) pose a technical challenge and the risk of operator-dependent complications like intraprocedural bleeding and perforation. Therefore, we aimed at developing an AI-algorithm for intraprocedural real time vessel detection during ESD and POEM. Methods A training dataset consisting of 5470 annotated still images from 59 full-length videos (47 ESD, 12 POEM) and 179681 unlabeled images was used to train a DeepLabV3+neural network with the ECMT semi-supervised learning method. Evaluation for vessel detection rate (VDR) and time (VDT) of 19 endoscopists with and without AI-support was performed using a testing dataset of 101 standardized video clips with 200 predefined blood vessels. Endoscopists were stratified into trainees and experts in third space endoscopy. Results The AI algorithm had a mean VDR of 93.5\% and a median VDT of 0.32 seconds. AI support was associated with a statistically significant increase in VDR from 54.9\% to 73.0\% and from 59.0\% to 74.1\% for trainees and experts, respectively. VDT significantly decreased from 7.21 sec to 5.09 sec for trainees and from 6.10 sec to 5.38 sec for experts in the AI-support group. False positive (FP) readings occurred in 4.5\% of frames. FP structures were detected significantly shorter than true positives (0.71 sec vs. 5.99 sec). Conclusions AI improved VDR and VDT of trainees and experts in third space endoscopy and may reduce performance variability during training. Further research is needed to evaluate the clinical impact of this new technology.}, language = {en} } @misc{ScheppachNunesArizietal., author = {Scheppach, Markus W. and Nunes, Danilo Weber and Arizi, X. and Rauber, David and Probst, Andreas and Nagl, Sandra and R{\"o}mmele, Christoph and Meinikheim, Michael and Palm, Christoph and Messmann, Helmut and Ebigbo, Alanna}, title = {Procedural phase recognition in endoscopic submucosal dissection (ESD) using artificial intelligence (AI)}, series = {Endoscopy}, volume = {56}, journal = {Endoscopy}, number = {S 02}, publisher = {Thieme}, address = {Stuttgart}, doi = {10.1055/s-0044-1783804}, pages = {S439}, abstract = {Aims Recent evidence suggests the possibility of intraprocedural phase recognition in surgical operations as well as endoscopic interventions such as peroral endoscopic myotomy and endoscopic submucosal dissection (ESD) by AI-algorithms. The intricate measurement of intraprocedural phase distribution may deepen the understanding of the procedure. Furthermore, real-time quality assessment as well as automation of reporting may become possible. Therefore, we aimed to develop an AI-algorithm for intraprocedural phase recognition during ESD. Methods A training dataset of 364385 single images from 9 full-length ESD videos was compiled. Each frame was classified into one procedural phase. Phases included scope manipulation, marking, injection, application of electrical current and bleeding. Allocation of each frame was only possible to one category. This training dataset was used to train a Video Swin transformer to recognize the phases. Temporal information was included via logarithmic frame sampling. Validation was performed using two separate ESD videos with 29801 single frames. Results The validation yielded sensitivities of 97.81\%, 97.83\%, 95.53\%, 85.01\% and 87.55\% for scope manipulation, marking, injection, electric application and bleeding, respectively. Specificities of 77.78\%, 90.91\%, 95.91\%, 93.65\% and 84.76\% were measured for the same parameters. Conclusions The developed algorithm was able to classify full-length ESD videos on a frame-by-frame basis into the predefined classes with high sensitivities and specificities. Future research will aim at the development of quality metrics based on single-operator phase distribution.}, language = {en} } @misc{ScheppachRauberStallhoferetal., author = {Scheppach, Markus W. and Rauber, David and Stallhofer, Johannes and Muzalyova, Anna and Otten, Vera and Manzeneder, Carolin and Schwamberger, Tanja and Wanzl, Julia and Schlottmann, Jakob and Tadic, Vidan and Probst, Andreas and Schnoy, Elisabeth and R{\"o}mmele, Christoph and Fleischmann, Carola and Meinikheim, Michael and Miller, Silvia and M{\"a}rkl, Bruno and Palm, Christoph and Messmann, Helmut and Ebigbo, Alanna}, title = {Performance comparison of a deep learning algorithm with endoscopists in the detection of duodenal villous atrophy (VA)}, series = {Endoscopy}, volume = {55}, journal = {Endoscopy}, number = {S02}, publisher = {Thieme}, doi = {10.1055/s-0043-1765421}, pages = {S165}, abstract = {Aims VA is an endoscopic finding of celiac disease (CD), which can easily be missed if pretest probability is low. In this study, we aimed to develop an artificial intelligence (AI) algorithm for the detection of villous atrophy on endoscopic images. Methods 858 images from 182 patients with VA and 846 images from 323 patients with normal duodenal mucosa were used for training and internal validation of an AI algorithm (ResNet18). A separate dataset was used for external validation, as well as determination of detection performance of experts, trainees and trainees with AI support. According to the AI consultation distribution, images were stratified into "easy" and "difficult". Results Internal validation showed 82\%, 85\% and 84\% for sensitivity, specificity and accuracy. External validation showed 90\%, 76\% and 84\%. The algorithm was significantly more sensitive and accurate than trainees, trainees with AI support and experts in endoscopy. AI support in trainees was associated with significantly improved performance. While all endoscopists showed significantly lower detection for "difficult" images, AI performance remained stable. Conclusions The algorithm outperformed trainees and experts in sensitivity and accuracy for VA detection. The significant improvement with AI support suggests a potential clinical benefit. Stable performance of the algorithm in "easy" and "difficult" test images may indicate an advantage in macroscopically challenging cases.}, language = {en} } @misc{ScheppachWeberNunesArizietal., author = {Scheppach, Markus W. and Weber Nunes, Danilo and Arizi, X. and Rauber, David and Probst, Andreas and Nagl, Sandra and R{\"o}mmele, Christoph and Palm, Christoph and Messmann, Helmut and Ebigbo, Alanna}, title = {Single frame workflow recognition during endoscopic submucosal dissection (ESD) using artificial intelligence (AI)}, series = {Endoscopy}, volume = {57}, journal = {Endoscopy}, number = {S 02}, publisher = {Thieme}, address = {Stuttgart}, doi = {10.1055/s-0045-1806324}, pages = {S511}, abstract = {Aims Precise surgical phase recognition and evaluation may improve our understanding of complex endoscopic procedures. Furthermore, quality control measurements and endoscopy training could benefit from objective descriptions of surgical phase distributions. Therefore, we aimed to develop an artificial intelligence algorithm for frame-by-frame operational phase recognition during endoscopic submucosal dissection (ESD). Methods Full length ESD-videos from 31 patients comprising 6.297.782 single images were collected retrospectively. Videos were annotated on a frame-by-frame basis for the operational macro-phases diagnostics, marking, injection, dissection and bleeding. Further subphases were the application of electrical current, visible injection of fluid into the submucosal space and scope manipulation, leading to 11 phases in total. 4.975.699 frames (21 patients) were used for training of a video swin transformer using uniform frame sampling for temporal information. Hyperparameter tuning was performed with 897.325 further frames (6 patients), while 424.758 frames (4 patients) were used for validation. Results The overall F1 scores on the test dataset for the macro-phases and all 11 phases were 0.96 and 0.90, respectively. The recall values for diagnostics, marking, injection, dissection and bleeding were 1.00, 1.00, 0.95, 0.96 and 0.93, respectively. Conclusions The algorithm classified operational phases during ESD with high accuracy. A precise evaluation of phase distribution may allow for the development of objective quality metrics for quality control and training.}, language = {en} } @misc{RoserMeinikheimMendeletal., author = {Roser, David and Meinikheim, Michael and Mendel, Robert and Palm, Christoph and Probst, Andreas and Muzalyova, Anna and Scheppach, Markus W. and Nagl, Sandra and Schnoy, Elisabeth and R{\"o}mmele, Christoph and Schulz, Dominik Andreas Helmut Otto and Schlottmann, Jakob and Prinz, Friederike and Rauber, David and R{\"u}ckert, Tobias and Matsumura, Tomoaki and Fernandez-Esparrach, G. and Parsa, Nasim and Byrne, Michael F. and Messmann, Helmut and Ebigbo, Alanna}, title = {Human-Computer Interaction: Impact of Artificial Intelligence on the diagnostic confidence of endoscopists assessing videos of Barrett's esophagus}, series = {Endoscopy}, volume = {56}, journal = {Endoscopy}, number = {S 02}, publisher = {Georg Thieme Verlag}, issn = {1438-8812}, doi = {10.1055/s-0044-1782859}, pages = {79}, abstract = {Aims Human-computer interactions (HCI) may have a relevant impact on the performance of Artificial Intelligence (AI). Studies show that although endoscopists assessing Barrett's esophagus (BE) with AI improve their performance significantly, they do not achieve the level of the stand-alone performance of AI. One aspect of HCI is the impact of AI on the degree of certainty and confidence displayed by the endoscopist. Indirectly, diagnostic confidence when using AI may be linked to trust and acceptance of AI. In a BE video study, we aimed to understand the impact of AI on the diagnostic confidence of endoscopists and the possible correlation with diagnostic performance. Methods 22 endoscopists from 12 centers with varying levels of BE experience reviewed ninety-six standardized endoscopy videos. Endoscopists were categorized into experts and non-experts and randomly assigned to assess the videos with and without AI. Participants were randomized in two arms: Arm A assessed videos first without AI and then with AI, while Arm B assessed videos in the opposite order. Evaluators were tasked with identifying BE-related neoplasia and rating their confidence with and without AI on a scale from 0 to 9. Results The utilization of AI in Arm A (without AI first, with AI second) significantly elevated confidence levels for experts and non-experts (7.1 to 8.0 and 6.1 to 6.6, respectively). Only non-experts benefitted from AI with a significant increase in accuracy (68.6\% to 75.5\%). Interestingly, while the confidence levels of experts without AI were higher than those of non-experts with AI, there was no significant difference in accuracy between these two groups (71.3\% vs. 75.5\%). In Arm B (with AI first, without AI second), experts and non-experts experienced a significant reduction in confidence (7.6 to 7.1 and 6.4 to 6.2, respectively), while maintaining consistent accuracy levels (71.8\% to 71.8\% and 67.5\% to 67.1\%, respectively). Conclusions AI significantly enhanced confidence levels for both expert and non-expert endoscopists. Endoscopists felt significantly more uncertain in their assessments without AI. Furthermore, experts with or without AI consistently displayed higher confidence levels than non-experts with AI, irrespective of comparable outcomes. These findings underscore the possible role of AI in improving diagnostic confidence during endoscopic assessment.}, language = {en} } @article{RoserMeinikheimMuzalyovaetal., author = {Roser, David and Meinikheim, Michael and Muzalyova, Anna and Mendel, Robert and Palm, Christoph and Probst, Andreas and Nagl, Sandra and Scheppach, Markus W. and R{\"o}mmele, Christoph and Schnoy, Elisabeth and Parsa, Nasim and Byrne, Michael F. and Messmann, Helmut and Ebigbo, Alanna}, title = {Artificial intelligence-assisted endoscopy and examiner confidence : a study on human-artificial intelligence interaction in Barrett's Esophagus (With Video)}, series = {DEN Open}, volume = {6}, journal = {DEN Open}, number = {1}, publisher = {Wiley}, doi = {10.1002/deo2.70150}, pages = {8}, abstract = {Objective Despite high stand-alone performance, studies demonstrate that artificial intelligence (AI)-supported endoscopic diagnostics often fall short in clinical applications due to human-AI interaction factors. This video-based trial on Barrett's esophagus aimed to investigate how examiner behavior, their levels of confidence, and system usability influence the diagnostic outcomes of AI-assisted endoscopy. Methods The present analysis employed data from a multicenter randomized controlled tandem video trial involving 22 endoscopists with varying degrees of expertise. Participants were tasked with evaluating a set of 96 endoscopic videos of Barrett's esophagus in two distinct rounds, with and without AI assistance. Diagnostic confidence levels were recorded, and decision changes were categorized according to the AI prediction. Additional surveys assessed user experience and system usability ratings. Results AI assistance significantly increased examiner confidence levels (p < 0.001) and accuracy. Withdrawing AI assistance decreased confidence (p < 0.001), but not accuracy. Experts consistently reported higher confidence than non-experts (p < 0.001), regardless of performance. Despite improved confidence, correct AI guidance was disregarded in 16\% of all cases, and 9\% of initially correct diagnoses were changed to incorrect ones. Overreliance on AI, algorithm aversion, and uncertainty in AI predictions were identified as key factors influencing outcomes. The System Usability Scale questionnaire scores indicated good to excellent usability, with non-experts scoring 73.5 and experts 85.6. Conclusions Our findings highlight the pivotal function of examiner behavior in AI-assisted endoscopy. To fully realize the benefits of AI, implementing explainable AI, improving user interfaces, and providing targeted training are essential. Addressing these factors could enhance diagnostic accuracy and confidence in clinical practice.}, language = {en} } @article{ScheppachRauberStallhoferetal., author = {Scheppach, Markus W. and Rauber, David and Stallhofer, Johannes and Muzalyova, Anna and Otten, Vera and Manzeneder, Carolin and Schwamberger, Tanja and Wanzl, Julia and Schlottmann, Jakob and Tadic, Vidan and Probst, Andreas and Schnoy, Elisabeth and R{\"o}mmele, Christoph and Fleischmann, Carola and Meinikheim, Michael and Miller, Silvia and M{\"a}rkl, Bruno and Stallmach, Andreas and Palm, Christoph and Messmann, Helmut and Ebigbo, Alanna}, title = {Detection of duodenal villous atrophy on endoscopic images using a deep learning algorithm}, series = {Gastrointestinal Endoscopy}, journal = {Gastrointestinal Endoscopy}, publisher = {Elsevier}, doi = {10.1016/j.gie.2023.01.006}, abstract = {Background and aims Celiac disease with its endoscopic manifestation of villous atrophy is underdiagnosed worldwide. The application of artificial intelligence (AI) for the macroscopic detection of villous atrophy at routine esophagogastroduodenoscopy may improve diagnostic performance. Methods A dataset of 858 endoscopic images of 182 patients with villous atrophy and 846 images from 323 patients with normal duodenal mucosa was collected and used to train a ResNet 18 deep learning model to detect villous atrophy. An external data set was used to test the algorithm, in addition to six fellows and four board certified gastroenterologists. Fellows could consult the AI algorithm's result during the test. From their consultation distribution, a stratification of test images into "easy" and "difficult" was performed and used for classified performance measurement. Results External validation of the AI algorithm yielded values of 90 \%, 76 \%, and 84 \% for sensitivity, specificity, and accuracy, respectively. Fellows scored values of 63 \%, 72 \% and 67 \%, while the corresponding values in experts were 72 \%, 69 \% and 71 \%, respectively. AI consultation significantly improved all trainee performance statistics. While fellows and experts showed significantly lower performance for "difficult" images, the performance of the AI algorithm was stable. Conclusion In this study, an AI algorithm outperformed endoscopy fellows and experts in the detection of villous atrophy on endoscopic still images. AI decision support significantly improved the performance of non-expert endoscopists. The stable performance on "difficult" images suggests a further positive add-on effect in challenging cases.}, language = {en} } @misc{MeinikheimMendelProbstetal., author = {Meinikheim, Michael and Mendel, Robert and Probst, Andreas and Scheppach, Markus W. and Schnoy, Elisabeth and Nagl, Sandra and R{\"o}mmele, Christoph and Prinz, Friederike and Schlottmann, Jakob and Golger, Daniela and Palm, Christoph and Messmann, Helmut and Ebigbo, Alanna}, title = {AI-assisted detection and characterization of early Barrett's neoplasia: Results of an Interim analysis}, series = {Endoscopy}, volume = {55}, journal = {Endoscopy}, number = {S02}, publisher = {Thieme}, doi = {10.1055/s-0043-1765437}, pages = {S169}, abstract = {Aims Evaluation of the add-on effect an artificial intelligence (AI) based clinical decision support system has on the performance of endoscopists with different degrees of expertise in the field of Barrett's esophagus (BE) and Barrett's esophagus-related neoplasia (BERN). Methods The support system is based on a multi-task deep learning model trained to solve a segmentation and several classification tasks. The training approach represents an extension of the ECMT semi-supervised learning algorithm. The complete system evaluates a decision tree between estimated motion, classification, segmentation, and temporal constraints, to decide when and how the prediction is highlighted to the observer. In our current study, ninety-six video cases of patients with BE and BERN were prospectively collected and assessed by Barrett's specialists and non-specialists. All video cases were evaluated twice - with and without AI assistance. The order of appearance, either with or without AI support, was assigned randomly. Participants were asked to detect and characterize regions of dysplasia or early neoplasia within the video sequences. Results Standalone sensitivity, specificity, and accuracy of the AI system were 92.16\%, 68.89\%, and 81.25\%, respectively. Mean sensitivity, specificity, and accuracy of expert endoscopists without AI support were 83,33\%, 58,20\%, and 71,48 \%, respectively. Gastroenterologists without Barrett's expertise but with AI support had a comparable performance with a mean sensitivity, specificity, and accuracy of 76,63\%, 65,35\%, and 71,36\%, respectively. Conclusions Non-Barrett's experts with AI support had a similar performance as experts in a video-based study.}, language = {en} } @article{ScheppachMendelMuzalyovaetal., author = {Scheppach, Markus W. and Mendel, Robert and Muzalyova, Anna and Rauber, David and Probst, Andreas and Nagl, Sandra and R{\"o}mmele, Christoph and Yip, Hon Chi and Lau, Louis Ho Shing and G{\"o}lder, Stefan Karl and Schmidt, Arthur and Kouladouros, Konstantinos and Abdelhafez, Mohamed and Walter, Benjamin M. and Meinikheim, Michael and Chiu, Philip Wai Yan and Palm, Christoph and Messmann, Helmut and Ebigbo, Alanna}, title = {Artificial intelligence improves submucosal vessel detection during third space endoscopy}, series = {Endoscopy}, journal = {Endoscopy}, publisher = {Thieme}, address = {Stuttgart}, doi = {10.1055/a-2534-1164}, abstract = {Background and study aims: While artificial intelligence (AI) shows high potential in decision support for diagnostic gastrointestinal endoscopy, its role in therapeutic endoscopy remains unclear. Third space endoscopic procedures pose the risk of intraprocedural bleeding. Therefore, we aimed to develop an AI algorithm for intraprocedural blood vessel detection. Patients and Methods: Using a test dataset with 101 standardized video clips containing 200 predefined submucosal blood vessels, 19 endoscopists were evaluated for the vessel detection rate (VDR) and time (VDT) with and without support of an AI algorithm. Test subjects were grouped according to experience in ESD. Results: With AI support, endoscopists VDR increased from 56.4\% [CI 54.1-58.6] to 72.4\% [CI 70.3-74.4]. Endoscopists' VDT dropped from 6.7sec [CI 6.2-7.1] to 5.2sec [CI 4.8-5.7]. False positive (FP) readings appeared in 4.5\% of frames and were marked significantly shorter than true positives (6.0sec [CI 5.28-6.70] vs. 0.7sec [CI 0.55-0.87]). Conclusions: AI improved the vessel detection rate and time of endoscopists during third space endoscopy. While these data need to be corroborated by clinical trials, AI may prove to be an invaluable tool for the improvement of endoscopic interventions.}, language = {en} } @misc{ScheppachWeberNunesRauberetal., author = {Scheppach, Markus W. and Weber Nunes, Danilo and Rauber, David and Arizi, X. and Probst, Andreas and Nagl, Sandra and R{\"o}mmele, Christoph and Ebigbo, Alanna and Palm, Christoph and Messmann, Helmut}, title = {K{\"u}nstliche Intelligenz-basierte Erkennung von interventionellen Phasen bei der endoskopischen Submukosadissektion}, series = {Zeitschrift f{\"u}r Gastroenterologie}, volume = {63}, journal = {Zeitschrift f{\"u}r Gastroenterologie}, number = {08}, publisher = {Thieme}, address = {Stuttgart}, doi = {10.1055/s-0045-1811093}, pages = {e612 -- e613}, abstract = {Einleitung: Die endoskopische Submukosadissektion (ESD) ist ein komplexes endoskopisches Verfahren, das technische Expertise erfordert. Objektive Methoden zur Analyse von interventionellen Abl{\"a}ufen bei ESD k{\"o}nnten f{\"u}r Qualit{\"a}tssicherung und Ausbildung, wie auch eine automatische Befunderstellung von Nutzen sein. Ziele: In dieser Studie wurde ein KI-Algorithmus f{\"u}r die Erkennung und Klassifizierung der interventionellen Phasen der ESD entwickelt, um die technische Basis f{\"u}r eine standardisierte Leistungsbewertung und automatische Befunderstellung zu schaffen. Methodik: Vollst{\"a}ndige ESD-Videoaufnahmen von 49 Patienten wurden retrospektiv zusammengestellt. Der Datensatz umfasste 6.390.151 Einzelbilder, die alle f{\"u}r die folgenden interventionellen Phasen annotiert wurden: Diagnostik, Markierung, Injektion, Dissektion und H{\"a}mostase. 3.973.712 Bilder (28 Patienten) wurden f{\"u}r das Training eines Video-Swin-Transformers genutzt. Dabei wurde temporale Information durch standardisierte BIldextraktion in festgelegten zeitlichen Abst{\"a}nden zum analysierten Bild inkorporiert. 2.416.439 separate Bilder (21 Patienten) wurden f{\"u}r eine interne Validierung genutzt. Ergebnis: Bei der internen Evaluation erreichte das System insgesamt einen F1-Wert von 0,88. Es wurden F1-Werte von 0,99, 0,89, 0,89, 0,91 und 0,52 f{\"u}r Diagnostik, Markierung, Injektion, Dissektion bzw. Blutungsmanagement gemessen. Die Sensitivit{\"a}ten f{\"u}r dieselben Parameter betrugen 1,00, 0,80, 0,94, 0,89 und 0,67, die Spezifit{\"a}ten lagen bei 1,00, 1,00, 0,98, 0,88 und 0,93. Positive pr{\"a}diktive Werte wurden mit 0,98, 1,00, 0,85, 0,94 und 0,43 gemessen. Schlussfolgerung: In dieser vorl{\"a}ufigen Studie zeigte ein KI-Algorithmus eine hohe Leistungsf{\"a}higkeit f{\"u}r die Einzelbild-Erkennung von Verfahrensphasen w{\"a}hrend der ESD. Die vergleichsweise niedrige Leistung f{\"u}r die Blutungsphase wurde auf das seltene Auftreten von Blutungsepisoden im Trainingsdatensatz zur{\"u}ckgef{\"u}hrt, der zu diesem Zeitpunkt nur Videos in voller L{\"a}nge umfasste. Die zuk{\"u}nftige Entwicklung des Algorithmus wird sich auf die Reduzierung von Klassenungleichgewichten durch selektive Annotationsprotokolle konzentrieren.}, language = {de} } @misc{ScheppachRauberZingleretal., author = {Scheppach, Markus W. and Rauber, David and Zingler, C. and Weber Nunes, Danilo and Probst, Andreas and R{\"o}mmele, Christoph and Nagl, Sandra and Ebigbo, Alanna and Palm, Christoph and Messmann, Helmut}, title = {Instrumentenerkennung w{\"a}hrend der endoskopischen Submukosadissektion mittels k{\"u}nstlicher Intelligenz}, series = {Zeitschrift f{\"u}r Gastroenterologie}, volume = {63}, journal = {Zeitschrift f{\"u}r Gastroenterologie}, number = {8}, publisher = {Thieme}, doi = {10.1055/s-0045-1811092}, abstract = {Einleitung: Die endoskopische Submukosadissektion (ESD) ist eine komplexe Technik zur Resektion gastrointestinaler Fr{\"u}hneoplasien. Dabei werden f{\"u}r die verschiedenen Schritte der Intervention spezifische endoskopische Instrumente verwendet. Die pr{\"a}zise und automatische Erkennung und Abgrenzung der verwendeten Instrumente (Injektionsnadeln, elektrochirurgische Messer mit unterschiedlichen Konfigurationen, h{\"a}mostatische Zangen) k{\"o}nnte wertvolle Informationen {\"u}ber den Fortschritt und die Verfahrensmerkmale der ESD liefern und eine automatische standardisierte Berichterstattung erm{\"o}glichen. Ziele: Ziel dieser Studie war die Entwicklung eines KI-Algorithmus zur Erkennung und Delineation von endoskopischen Instrumenten bei der ESD. Methodik: 17 ESD-Videos (9×rektal, 5×{\"o}sophageal, 3×gastrisch) wurden retrospektiv zusammengestellt. Auf 8530 Einzelbilder dieser Videos wurden durch 2 Studienmitarbeiter die folgenden Klassen eingezeichnet: Hakenmesser - Spitze, Hakenmesser - Katheter, Nadelmesser - Spitze und - Katheter, Injektionsnadel -Spitze und - Katheter sowie h{\"a}mostatische Zange - Spitze und - Katheter. Der annotierte Datensatz wurde zum Training eines DeepLabV3+-Deep-Learning-Algorithmus mit ConvNeXt-Backbone zur Erkennung und Abgrenzung der genannten Klassen verwendet. Die Evaluation erfolgte durch 5-fache interne Kreuzvalidierung. Ergebnis: Die Validierung auf Einzelpixelbasis ergab insgesamt einen F1-Score von 0,80, eine Sensitivit{\"a}t von 0,81 und eine Spezifit{\"a}t von 1,00. Es wurden F1-Scores von 1,00, 0,97, 0,80, 0,98, 0,85, 0,97, 0,80, 0,51 bzw. 0,85 f{\"u}r die Klassen Hakenmesser - Katheter und - Spitze, Nadelmesser - Katheter und - Spitze, Injektionsnadel - Katheter und - Spitze, h{\"a}mostatische Zange - Katheter und - Spitze gemessen. Schlussfolgerung: In dieser Studie wurden die wichtigsten endoskopischen Instrumente, die w{\"a}hrend der ESD verwendet werden, mit hoher Genauigkeit erkannt. Die geringere Leistung bei der h{\"a}mostatische Zange - Katheter kann auf die Unterrepr{\"a}sentation dieser Klassen in den Trainingsdaten zur{\"u}ckgef{\"u}hrt werden. Zuk{\"u}nftige Studien werden sich auf die Erweiterung der Instrumentenklassen sowie auf die Ausbalancierung der Trainingsdaten konzentrieren.}, language = {de} } @misc{MeinikheimMendelProbstetal., author = {Meinikheim, Michael and Mendel, Robert and Probst, Andreas and Scheppach, Markus W. and Nagl, Sandra and Schnoy, Elisabeth and R{\"o}mmele, Christoph and Prinz, Friederike and Schlottmann, Jakob and Messmann, Helmut and Palm, Christoph and Ebigbo, Alanna}, title = {Einfluss von K{\"u}nstlicher Intelligenz auf die Performance von niedergelassenen Gastroenterolog:innen bei der Beurteilung von Barrett-{\"O}sophagus}, series = {Zeitschrift f{\"u}r Gastroenterologie}, volume = {61}, journal = {Zeitschrift f{\"u}r Gastroenterologie}, number = {8}, publisher = {Thieme}, address = {Stuttgart}, doi = {10.1055/s-0043-1771711}, abstract = {Einleitung Die Differenzierung zwischen nicht dysplastischem Barrett-{\"O}sophagus (NDBE) und mit Barrett-{\"O}sophagus assoziierten Neoplasien (BERN) w{\"a}hrend der endoskopischen Inspektion erfordert viel Expertise. Die fr{\"u}he Diagnosestellung ist wichtig f{\"u}r die weitere Prognose des Barrett-Karzinoms. In Deutschland werden Patient:innen mit einem Barrett-{\"O}sophagus (BE) in der Regel im niedergelassenen Sektor {\"u}berwacht. Ziele Ziel ist es, den Einfluss von einem auf K{\"u}nstlicher Intelligenz (KI) basierenden klinischen Entscheidungsunterst{\"u}tzungssystems (CDSS) auf die Performance von niedergelassenen Gastroenterolog:innen (NG) bei der Evaluation von Barrett-{\"O}sophagus (BE) zu untersuchen. Methodik Es erfolgte die prospektive Sammlung von 96 unver{\"a}nderten hochaufl{\"o}senden Videos mit F{\"a}llen von Patient:innen mit histologisch best{\"a}tigtem NDBE und BERN. Alle eingeschlossenen F{\"a}lle enthielten mindestens zwei der folgenden Darstellungsmethoden: HD-Weißlichtendoskopie, Narrow Band Imaging oder Texture and Color Enhancement Imaging. Sechs NG von sechs unterschiedlichen Praxen wurden als Proband:innen eingeschlossen. Es erfolgte eine permutierte Block-Randomisierung der Videof{\"a}lle in entweder Gruppe A oder Gruppe B. Gruppe A implizierte eine Evaluation des Falls durch Proband:innen zun{\"a}chst ohne KI und anschließend mit KI als CDSS. In Gruppe B erfolgte die Evaluation in umgekehrter Reihenfolge. Anschließend erfolgte eine zuf{\"a}llige Wiedergabe der so entstandenen Subgruppen im Rahmen des Tests. Ergebnis In diesem Test konnte ein von uns entwickeltes KI-System (Barrett-Ampel) eine Sensitivit{\"a}t von 92,2\%, eine Spezifit{\"a}t von 68,9\% und eine Accuracy von 81,3\% erreichen. Mit der Hilfe von KI verbesserte sich die Sensitivit{\"a}t der NG von 64,1\% auf 71,2\% (p<0,001) und die Accuracy von 66,3\% auf 70,8\% (p=0,006) signifikant. Eine signifikante Verbesserung dieser Parameter zeigte sich ebenfalls, wenn die Proband:innen die F{\"a}lle zun{\"a}chst ohne KI evaluierten (Gruppe A). Wurde der Fall jedoch als Erstes mit der Hilfe von KI evaluiert (Gruppe B), blieb die Performance nahezu konstant. Schlussfolgerung Es konnte ein performantes KI-System zur Evaluation von BE entwickelt werden. NG verbessern sich bei der Evaluation von BE durch den Einsatz von KI.}, language = {de} } @misc{ScheppachMendelMuzalyovaetal., author = {Scheppach, Markus W. and Mendel, Robert and Muzalyova, Anna and Rauber, David and Probst, Andreas and Nagl, Sandra and R{\"o}mmele, Christoph and Yip, Hon Chi and Lau, Louis Ho Shing and G{\"o}lder, Stefan Karl and Schmidt, Arthur and Kouladouros, Konstantinos and Abdelhafez, Mohamed and Walter, B. and Meinikheim, Michael and Chiu, Philip Wai Yan and Palm, Christoph and Messmann, Helmut and Ebigbo, Alanna}, title = {K{\"u}nstliche Intelligenz erh{\"o}ht die Gef{\"a}ßerkennung von Endoskopikern bei third space Endoskopie}, series = {Zeitschrift f{\"u}r Gastroenterologie}, volume = {62}, journal = {Zeitschrift f{\"u}r Gastroenterologie}, number = {09}, publisher = {Georg Thieme Verlag KG}, doi = {10.1055/s-0044-1790087}, pages = {e830}, abstract = {Einleitung: K{\"u}nstliche Intelligenz (KI)-Algorithmen unterst{\"u}tzen Endoskopiker bei der Erkennung und Charakterisierung von Kolonpolypen in der klinischen Praxis und f{\"u}hren zu einer Erh{\"o}hung der Adenomdetektionsrate. Auch bei therapeutischen Maßnahmen wie der endoskopischen Submukosadissektion (ESD) k{\"o}nne relevante anatomische Strukturen durch KI mit hoher Genauigkeit erkannt und im endoskopischen Bild in Echtzeit markiert werden. Der Effekt einer solchen Applikation auf die Gef{\"a}ßdetektion von Endoskopikern ist bislang nicht erforscht. Ziele: In dieser Studie wurde der Effekt eines KI-Algorithmus zur Echtzeit-Gef{\"a}ßmarkierung bei ESD auf die Gef{\"a}ßdetektionsrate von Endoskopikern untersucht. Methodik: 59 third space Endoskopievideos wurde aus der Datenbank des Universit{\"a}tsklinikums Augsburg extrahiert. Auf 5470 Einzelbildern dieser Untersuchungen wurde submukosale Blutgef{\"a}ße annotiert. Zusammen mit weiteren 179681 unmarkierten Bildern wurde ein DeepLabV3+ neuronales Netzwerk mit einer semi-supervised learning Methode darin trainiert, submukosale Blutgef{\"a}ße auf dem endoskopischen Bild zu erkennen und in Echtzeit einzuzeichnen. Anhand eines Videotests mit 101 Videoclips und 200 vordefinierten Blutgef{\"a}ßen wurden 19 Endoskopiker mit und ohne KI Unterst{\"u}tzung getestet. Ergebnis: Der Algorithmus erkannte in dem Videotest 93.5\% der Gef{\"a}ße in einer Detektionszeit von im Median 0,3 Sekunden. Die Gef{\"a}ßdetektionsrate von Endoskopikern erh{\"o}hte sich durch KI Unterst{\"u}tzung von 56,4\% auf 72,4\% (p<0.001). Die Gef{\"a}ßdetektionszeit reduzierte sich durch KI-Unterst{\"u}tzung von 6,7 auf 5.2 Sekunden (p<0.001). Der Algorithmus zeigte eine Rate an falsch positiven Detektionen in 4.5\% der Einzelbilder. Falsch positiv erkannte Strukturen wurde k{\"u}rzer detektiert, als richtig positive (0.7 und 6.0 Sekunden, p<0.001). Schlussfolgerung: KI Unterst{\"u}tzung f{\"u}hrte zu einer erh{\"o}hten Gef{\"a}ßdetektionsrate und schnelleren Gef{\"a}ßdetektionszeit von Endoskopikern. Ein m{\"o}glicher klinischer Effekt auf die intraprozedurale Komplikationsrate oder Operationszeit k{\"o}nnte in prospektiven Studien ermittelt werden.}, language = {de} } @misc{ScheppachNunesArizietal., author = {Scheppach, Markus W. and Nunes, Danilo Weber and Arizi, X. and Rauber, David and Probst, Andreas and Nagl, Sandra and R{\"o}mmele, Christoph and Palm, Christoph and Messmann, Helmut and Ebigbo, Alanna}, title = {Intraoperative Phasenerkennung bei endoskopischer Submukosadissektion mit Hilfe von k{\"u}nstlicher Intelligenz}, series = {Zeitschrift f{\"u}r Gastroenterologie}, volume = {62}, journal = {Zeitschrift f{\"u}r Gastroenterologie}, number = {09}, publisher = {Georg Thieme Verlag KG}, doi = {10.1055/s-0044-1790084}, pages = {e828}, abstract = {Einleitung: K{\"u}nstliche Intelligenz (KI) wird in der Endoskopie des Gastrointestinaltraktes zur Erkennung und Charakterisierung von Kolonpolypen eingesetzt. Die Rolle von KI bei therapeutischen Maßnahmen wurde noch nicht eingehend untersucht. Eine intraprozedurale Phasenerkennung bei endoskopischer Submukoasdissektion (ESD) k{\"o}nnte die Erhebung von Qualit{\"a}tsindikatoren erm{\"o}glichen. Weiterhin k{\"o}nnte diese Technologie zu einem tieferen Verst{\"a}ndnis {\"u}ber die Eigenschaften der Prozedur f{\"u}hren und weiterf{\"u}hrende Applikationen zur automatischen Dokumentation oder standardisiertem Training vorbereiten. Ziele: Ziel dieser Studie war die Entwicklung eines KI Algorithmus zur intraprozeduralen Phasenerkennung bei endoskopischer Submukosadissektion. Methodik: 2071546 Einzelbilder aus 27 ESD Videos in voller L{\"a}nge wurden f{\"u}r die {\"u}bergeordneten Klassen Diagnostik, Markierung, Nadelinjektion, Dissektion und Blutung, sowie die untergeordneten Klassen Endoskop-Manipulation, Injektion und Applikation von elektrischem Strom annotiert. Mit einem Trainingsdatensatz (898440 Einzelbilder, 17 ESDs) wurde ein Video Swin Transformer mit uniformer Stichprobenentnahme trainiert und intern validiert (769523 Einzelbilder, 6 ESDs). Neben der internen Validierung wurde der Algorithmus anhand von einem separaten Testdatensatz (403583 Einzelbilder, 4 ESDs) evaluiert. Ergebnis: Der F1 Score des Algorithmus f{\"u}r alle Klassen lag in der internen Validierung bei 83\%, in dem separaten Test bei 90\%. Anhand des separaten Tests wurden true positive (TP)-Raten f{\"u}r Diagnostik, Markierung, Nadelinjektion, Dissektion und Blutung von 100\%, 100\%, 96\%, 97\% und 93\% ermittelt. F{\"u}r Endoskopmanipulation, Injektion und Applikation von Elektrizit{\"a}t lagen die TP-Raten bei 92\%, 98\% und 91\%. Schlussfolgerung: Der entwickelte Algorithmus klassifizierte ESD Videos in voller L{\"a}nge und anhand jedes einzelnen Bildes mit hoher Genauigkeit. Zuk{\"u}nftige Forschungsvorhaben k{\"o}nnten intraoperative Qualit{\"a}tsindikatioren auf Basis dieser Informationen entwickeln und eine automatisierte Dokumentation erm{\"o}glichen.}, language = {de} } @misc{ScheppachRauberMendeletal., author = {Scheppach, Markus W. and Rauber, David and Mendel, Robert and Palm, Christoph and Byrne, Michael F. and Messmann, Helmut and Ebigbo, Alanna}, title = {Detection Of Celiac Disease Using A Deep Learning Algorithm}, series = {Endoscopy}, volume = {53}, journal = {Endoscopy}, number = {S 01}, publisher = {Georg Thieme Verlag}, address = {Stuttgart}, doi = {10.1055/s-0041-1724970}, abstract = {Aims Celiac disease (CD) is a complex condition caused by an autoimmune reaction to ingested gluten. Due to its polymorphic manifestation and subtle endoscopic presentation, the diagnosis is difficult and thus the disorder is underreported. We aimed to use deep learning to identify celiac disease on endoscopic images of the small bowel. Methods Patients with small intestinal histology compatible with CD (MARSH classification I-III) were extracted retrospectively from the database of Augsburg University hospital. They were compared to patients with no clinical signs of CD and histologically normal small intestinal mucosa. In a first step MARSH III and normal small intestinal mucosa were differentiated with the help of a deep learning algorithm. For this, the endoscopic white light images were divided into five equal-sized subsets. We avoided splitting the images of one patient into several subsets. A ResNet-50 model was trained with the images from four subsets and then validated with the remaining subset. This process was repeated for each subset, such that each subset was validated once. Sensitivity, specificity, and harmonic mean (F1) of the algorithm were determined. Results The algorithm showed values of 0.83, 0.88, and 0.84 for sensitivity, specificity, and F1, respectively. Further data showing a comparison between the detection rate of the AI model and that of experienced endoscopists will be available at the time of the upcoming conference. Conclusions We present the first clinical report on the use of a deep learning algorithm for the detection of celiac disease using endoscopic images. Further evaluation on an external data set, as well as in the detection of CD in real-time, will follow. However, this work at least suggests that AI can assist endoscopists in the endoscopic diagnosis of CD, and ultimately may be able to do a true optical biopsy in live-time.}, language = {en} } @inproceedings{RauberMendelScheppachetal., author = {Rauber, David and Mendel, Robert and Scheppach, Markus W. and Ebigbo, Alanna and Messmann, Helmut and Palm, Christoph}, title = {Analysis of Celiac Disease with Multimodal Deep Learning}, series = {Bildverarbeitung f{\"u}r die Medizin 2022: Proceedings, German Workshop on Medical Image Computing, Heidelberg, June 26-28, 2022}, booktitle = {Bildverarbeitung f{\"u}r die Medizin 2022: Proceedings, German Workshop on Medical Image Computing, Heidelberg, June 26-28, 2022}, publisher = {Springer Vieweg}, address = {Wiesbaden}, doi = {10.1007/978-3-658-36932-3_25}, pages = {115 -- 120}, abstract = {Celiac disease is an autoimmune disorder caused by gluten that results in an inflammatory response of the small intestine.We investigated whether celiac disease can be detected using endoscopic images through a deep learning approach. The results show that additional clinical parameters can improve the classification accuracy. In this work, we distinguished between healthy tissue and Marsh III, according to the Marsh score system. We first trained a baseline network to classify endoscopic images of the small bowel into these two classes and then augmented the approach with a multimodality component that took the antibody status into account.}, language = {en} } @misc{ScheppachMendelProbstetal., author = {Scheppach, Markus W. and Mendel, Robert and Probst, Andreas and Rauber, David and R{\"u}ckert, Tobias and Meinikheim, Michael and Palm, Christoph and Messmann, Helmut and Ebigbo, Alanna}, title = {Real-time detection and delineation of tissue during third-space endoscopy using artificial intelligence (AI)}, series = {Endoscopy}, volume = {55}, journal = {Endoscopy}, number = {S02}, publisher = {Thieme}, doi = {10.1055/s-0043-1765128}, pages = {S53 -- S54}, abstract = {Aims AI has proven great potential in assisting endoscopists in diagnostics, however its role in therapeutic endoscopy remains unclear. Endoscopic submucosal dissection (ESD) is a technically demanding intervention with a slow learning curve and relevant risks like bleeding and perforation. Therefore, we aimed to develop an algorithm for the real-time detection and delineation of relevant structures during third-space endoscopy. Methods 5470 still images from 59 full length videos (47 ESD, 12 POEM) were annotated. 179681 additional unlabeled images were added to the training dataset. Consequently, a DeepLabv3+ neural network architecture was trained with the ECMT semi-supervised algorithm (under review elsewhere). Evaluation of vessel detection was performed on a dataset of 101 standardized video clips from 15 separate third-space endoscopy videos with 200 predefined blood vessels. Results Internal validation yielded an overall mean Dice score of 85\% (68\% for blood vessels, 86\% for submucosal layer, 88\% for muscle layer). On the video test data, the overall vessel detection rate (VDR) was 94\% (96\% for ESD, 74\% for POEM). The median overall vessel detection time (VDT) was 0.32 sec (0.3 sec for ESD, 0.62 sec for POEM). Conclusions Evaluation of the developed algorithm on a video test dataset showed high VDR and quick VDT, especially for ESD. Further research will focus on a possible clinical benefit of the AI application for VDR and VDT during third-space endoscopy.}, subject = {Speiser{\"o}hrenkrankheit}, language = {en} }