Introduction

Device-assisted enteroscopy (DAE) plays an important role in management of enteric lesions. Endoscopic observation of erosions or ulcers is frequent and can be associated in many nosologic entities, including Crohn’s disease and non-steroid anti-inflammatory drug-induced enteropathy. Despite the rapid expansion of artificial intelligence (AI) application in different image-base procedures of gastroenterology, there is still lack of evidence of AI relevance during DAE. This study aimed to develop and test a Convolutional Neural Network (CNN)-based model for automatic detection of ulcers and erosions during DAE.

Aims & Methods

A unicentric retrospective study was conducted for the development of a CNN, based on a total of 258 DAE exams. DAE was performed by two experienced endoscopists using the double-balloon enteroscopy system Fujifilm EN-580T (n = 152), the single-balloon enteroscopy system Olympus EVIS EXERA II SIF-Q180 (n = 98) and the Olympus PowerSpiral Motorized Enteroscope PSF-1 (n = 8). After double validation, a total of 29513 images were used, of which 633 were considered ulcers or erosions. Data was separated into two sets: training and validation, with the latter being used for the evaluate the model’s performance. Our primary outcome measures were sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV), and area under the curve precision-recall curve (AUC-PR).

Results

Sensitivity and specificity were 86.2% and 99.8%, respectively. PPV and NPV were 91.8% and 99.7%, respectively. Overall accuracy of the CNN was 99.5%. The AUC-PR was 1.00.

Conclusion

To the best of our knowledge, this is the first CNN constructed and validated for automatic detection of ulcers and erosions during DAE, including spiral enteroscopy. The high diagnostic performance of this CNN in multidevice solves a significant issue of technological interoperability, allowing it to be replicated in many technological contexts.

Introduction

Device-assisted enteroscopy (DAE) allows deep exploration of the gastrointestinal (GI) tract, combining its diagnostic abilities with tissue sampling and the application of endoscopic therapy. Although the application of artificial intelligence (AI) is growing exponentially in various imaged-based gastroenterology procedures, there is still a lack of evidence of the AI's technical feasibility and clinical applicability of DAE. Convolutional Neural Networks (CNN) are a multi-layer AI architecture with high-performance levels for image analysis.

Aims & Methods

This study aimed to develop and test a multi-brand CNN-based algorithm for automatically detecting multiple lesions in DAE.
A unicentric retrospective study was conducted to develop a CNN based on 258 DAE exams. DAE was performed by two experienced endoscopists using the double-balloon enteroscopy system Fujifilm EN-580T (n = 152), the single-balloon enteroscopy system Olympus EVIS EXERA II SIF-Q180 (n = 98), and the Olympus PowerSpiral Motorized Enteroscope PSF-1 (n = 8). A total of 11205 images from different segments of the GI tract (stomach, small bowel, and colon) were included, 2535 images containing protruding lesions (polyps, epithelial tumors, subepithelial lesions, and nodules), 1435 containing blood or hematic residues, 1395 images containing angioectasias and 450 images containing ulcers/erosions. The remaining images showed normal mucosa (n=5390). A training dataset comprising 80% of the total pool of images (n=8964) was used to develop the network. The output provided by the network was compared to a consensus classification provided. Our primary outcome measures were sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV), and an area under the curve precision–recall curve (AUC-PR).

Results

The CNN model for automatic detection of GI protruding lesions identified these lesions with a sensitivity of 97.0%, a specificity of 97.4%, and positive and negative predictive values of 94.6% and 98.6%, respectively. The component for automatic detection of blood/hematic residues had a sensitivity of 95.8%, a specificity of 97.6%, and positive and negative predictive values of 91.4% and 98.9%, respectively. The CNN’s angioectasia model automatically detected angioectasia with a sensitivity of 88.5%, a specificity of 97.1%, and positive and negative predictive values of 88.1% and 97.0%, respectively. Furthermore, the CNN ulcers/erosions model automatically detected ulcers/erosions with a sensitivity of 100%, a specificity of 96.4%, and positive and negative predictive values of 96.8% and 100%, respectively. The algorithm’s accuracy was 98.7%. The AUC-PR was 1.00. The CNN analyzed images at approximately 239 frames per second, enabling AI live application in a real-life clinical setting in DAE.

Conclusion

The authors have developed a pioneer combined deep learning system for automatically detecting a wide range of lesions in different gastrointestinal tract segments. To our knowledge, this is the first study regarding the automatic multi-brand panendoscopic detection using Olympus PowerSpiral Motorized Enteroscope, overcoming a relevant interoperability challenge. Our results highlight that using a CNN to detect pleomorphic lesions is associated with high overall accuracy. The development of these tools may enhance the diagnostic yield of device-assisted enteroscopy in the setting of AI applications in digestive endoscopy, particularly for panendoscopic evaluation.

References

1. Goenka, M.K.; Majumder, S.; Kumar, S.; Sethy, P.K.; Goenka, U. Single center experience of capsule endoscopy in patients with obscure gastrointestinal bleeding. World J. Gastroenterol. 2011, 17, 774–778. [CrossRef] [PubMed]
2. Shahidi, N.C.; Ou, G.; Svarta, S.; Law, J.K.; Kwok, R.; Tong, J.; Lam, E.C.; Enns, R. Factors associated with positive findings from capsule endoscopy in patients with obscure gastrointestinal bleeding. Clin. Gastroenterol. Hepatol. 2012, 10, 1381–1385. [CrossRef] [PubMed]
3. Iddan, G.; Meron, G.; Glukhovsky, A.; Swain, P. Wireless capsule endoscopy. Nature 2000, 405, 417. [CrossRef]
4. Aoki, T.; Yamada, A.; Aoyama, K.; Saito, H.; Tsuboi, A.; Nakada, A.; Niikura, R.; Fujishiro, M.; Oka, S.; Ishihara, S.; et al. Automatic detection of erosions and ulcerations in wireless capsule endoscopy images based on a deep convolutional neural network. Gastrointest. Endosc. 2019, 89, 357–363.e2. [CrossRef] [PubMed]
5. Pennazio, M.; Rondonotti, E.; Despott, E.J.; Dray, X.; Keuchel, M.; Moreels, T.; Sanders, D.S.; Spada, C.; Carretero, C.; Cortegoso Valdivia, P.; et al. Small-bowel capsule endoscopy and device-assisted enteroscopy for diagnosis and treatment of small-bowel disorders: European Society of Gastrointestinal Endoscopy (ESGE) Guideline—Update 2022. Endoscopy 2023, 55, 58–95. [CrossRef]

Introduction

Device-assisted enteroscopy (DAE) is the only exam capable of a panendoscopic diagnostic and therapeutic intervention in gastrointestinal tract. Despite the limited data, DAE appears to be a safe procedure in pediatric population. In fact, a few studies demonstrate ​the feasibility and safetiness of DAE in several clinical contexts, namely small ​bowel ​bleeding, polyposis syndromes or inflammatory bowel disease.

Aims & Methods

With this study, the authors aim to describe DAE experience, specifically balloon assisted enteroscopy, in a pediatric population of a tertiary center.
A unicentric retrospective study was performed in patients up to 18 years old undergoing DAE (either by anterograde or retrograde pathway), between January 2011 and February 2024.

Results

A total of 23 DAE exams from 16 patients were included in the study, 56% from the male gender. The study population had a medium age of 13 years (ranging from 4 to 18 years old). DAE was performed due to polyposis syndromes in 8 cases (6 cases of Peutz-Jeghers syndrome and 2 cases of juvenile polyposis) and in suspicious cases of middle gastrointestinal bleeding in 8 patients (5 with melena and 3 with ferropenic anemia). There was a total of 17 anterograde DAE, 5 retrograde DAE and 1 intraoperatory enteroscopy. 35% of the procedures were performed in an inpatient setting.
DAE had a diagnostic yield of 95%, identifying polyps in the small bowel or other locations, ulcers, small bowel stenosis and foreign body. A therapeutic procedure was performed in 56% of the cases, namely polypectomy, argon plasma coagulation and foreign body removal. A self-limited bleeding was observed in the first 24 hours after an anterograde DAE, solved without the need of an additional endoscopic or surgical procedure. DAE led to a modification in the clinical outcome in over 90% of the patients.

Conclusion

The present study confirms the safety profile of DAE in the pediatric population. In a cohort from a tertiary referral center, DAE led to therapeutic procedure in the majority of patients, without significant procedure related complications.

References

1. Pennazio, Marco et al. “Small-bowel capsule endoscopy and device-assisted enteroscopy for diagnosis and treatment of small-bowel disorders: European Society of Gastrointestinal Endoscopy (ESGE) Guideline - Update 2022.” Endoscopy vol. 55,1 (2023): 58-95.
2. Di Nardo, Giovanni et al. “Enteroscopy in children.” United European gastroenterology journal vol. 6,7 (2018): 961-969.

Disclosure

1. Pennazio, Marco et al. “Small-bowel capsule endoscopy and device-assisted enteroscopy for diagnosis and treatment of small-bowel disorders: European Society of Gastrointestinal Endoscopy (ESGE) Guideline - Update 2022.” Endoscopy vol. 55,1 (2023): 58-95.
2. Di Nardo, Giovanni et al. “Enteroscopy in children.” United European gastroenterology journal vol. 6,7 (2018): 961-969.

Introduction

Capsule endoscopy (CE) is a minimally invasive exam suitable of panendoscopic evaluation of the gastrointestinal tract. Nevertheless, CE is time-consuming, with suboptimal diagnostic yield in the upper GI tract. Convolutional neural networks (CNN) are human brain architecture-based models suitable for image analysis. However, there is no study about their role in capsule panendoscopy.

Aims & Methods

Our aim was to develop an artificial intelligence (AI) model for panendoscopic automatic detection of pleomorphic lesions (namely vascular lesions, protuberant lesions, hematic residues, ulcers, and erosions). 355110 images (6977 esophageal, 12918 gastric, 258443 small bowel, 76772 colonic) from eight different CE and colon CE (CCE) devices were divided into a training and validation dataset in a patient split design. The model classification was compared to three CE experts’ classifications. The model’s performance was evaluated by its sensitivity, specificity, accuracy, positive predictive value, negative predictive value, and area under the precision-recall curve.

Results

The binary esophagus CNN had a diagnostic accuracy for pleomorphic lesions of 83.6%. The binary gastric CNN identified pleomorphic lesions with a 96.6% accuracy. The undenary small bowel CNN distinguished pleomorphic lesions with different hemorrhagic potential with 97.6% accuracy. The trinary colonic CNN (detection and differentiation of normal mucosa, pleomorphic lesions, and hematic residues) had 94.9% global accuracy.

Conclusion

Our group developed the first AI model for panendoscopic automatic detection of pleomorphic lesions in both CE and colon CE from multiple brands, solving a critical interoperability technological challenge. Deep learning-based tools may change the landscape of minimally invasive capsule panendoscopy.

References

1. Pennazio M, Rondonotti E, Despott EJ, Dray X, Keuchel M, Moreels T, et al. Small-bowel capsule endoscopy and device-assisted enteroscopy for diagnosis and treatment of small-bowel disorders: European Society of Gastrointestinal Endoscopy (ESGE) Guideline - Update 2022. Endoscopy. 2023 Jan;55(1):58-95. PubMed PMID: 36423618. Epub 20221124.
2. Le Berre C, Trang-Poisson C, Bourreille A. Small bowel capsule endoscopy and treat-to-target in Crohn's disease: A systematic review. World J Gastroenterol. 2019 Aug 21;25(31):4534-54. PubMed PMID: 31496630. PMCID: PMC6710184.
3. Iddan G, Meron G, Glukhovsky A, Swain P. Wireless capsule endoscopy. Nature. 2000 May 25;405(6785):417. PubMed PMID: 10839527.
4. Rondonotti E, Pennazio M. Colon capsule for panendoscopy: a narrow window of opportunity. Endosc Int Open. 2021 Dec;9(12):E1860-E2. PubMed PMID: 34917450. PMCID: PMC8670997. Epub 20211214.
5. Vuik FER, Moen S, Spaander MCW. Colon capsule endoscopy as panendoscopy: Using current knowledge to enhance possibilities. Endosc Int Open. 2022 May;10(5):E584. PubMed PMID: 35571484. PMCID: PMC9106432. Epub 20220513.

Introduction

Subepithelial lesions (SELs) are frequently detected during esophagogastroduodenoscopy, yet their diagnostic workup remains challenging. Accurate histological characterization is essential for optimal management. Endoscopic ultrasound (EUS) enables both lesion assessment and sampling and is the first-line modality for evaluating SELs >10 mm. However, EUS interpretation is operator-dependent and suffers from suboptimal diagnostic accuracy. This multicenter study aimed to develop and validate a deep learning model for automated detection and classification of upper gastrointestinal SELs, specifically distinguishing leiomyomas from gastrointestinal stromal tumors (GISTs).

Aims & Methods

A convolutional neural network (YOLO11Net) was trained using 57,190 EUS images from 65 patients across five countries (Portugal, USA, Spain, Brazil, and the UK) and five different EUS platforms. The dataset was split into training (70%), validation (20%), and testing (10%) subsets. The model was designed to perform binary classification between GISTs and leiomyomas. Performance was evaluated using accuracy and area under the precision-recall curve (AUC-PR).

Results

The model achieved an overall diagnostic accuracy of 95%. For GIST detection, the AUC-PR was 0.992, and for leiomyomas, 0.982, demonstrating excellent discriminative performance.

Conclusion

This is the first global AI model capable of detecting and differentiating the most prevalent SELs using EUS. The model’s high accuracy and robust AUC-PR values underscore its clinical potential to enhance diagnostic precision and guide targeted tissue sampling. The inclusion of data from multiple devices and international centers strengthens the model's generalizability and reduces demographic and technical bias. Larger, prospective multicenter studies are warranted to validate these findings and support clinical integration.

Introduction

Endoscopic ultrasound (EUS) is an essential tool in the evaluation of lymph nodes (LNs), especially for oncologic staging and guiding therapeutic decisions. Characteristics such as hypoechoic appearance, well-defined borders, round shape, and a size exceeding 10 mm are typically associated with malignancy, offering 80–100% accuracy when observed collectively—though this ideal combination is rare. While EUS-guided tissue acquisition (EUS-TA) contributes to diagnosis, its reliability can be limited.

Aims & Methods

This multicenter study aimed to assess the performance of a novel artificial intelligence (AI) model in predicting LN malignancy based on EUS images.
The study involved EUS images from nine centers across Spain, Portugal, Brazil, the United Kingdom, and the United States. Lesions were annotated as “malignant” or “benign” and outlined using bounding boxes. Final diagnoses were established through positive fine-needle aspiration or biopsy (FNA/FNB), or, when negative, confirmed with a minimum of six months of clinical follow-up. A convolutional neural network (CNN) was created using the YOLO (“You Only Look Once”) architecture, incorporating both detection and classification components. Key performance indicators included detection rate (DR), sensitivity, specificity, overall accuracy, and the area under the curve (AUC).

Results

The model was trained on 59,992 images derived from 82 EUS procedures. The CNN achieved a sensitivity of 98.8% (95% CI: 98.5–99.2%), a specificity of 99.0% (95% CI: 98.3–99.7%), and a precision of 99.0% (95% CI: 98.4–99.7%) in differentiating malignant from benign LNs. The negative predictive value (NPV) was 98.8% (95% CI: 98.4–99.2%) for malignant and 97.0% (95% CI: 95.0–98.9%) for benign nodes. The overall diagnostic accuracy for both malignant and benign findings reached 96.99% (95% CI: 87.8–96.2%).

Conclusion

This is the first study to explore the application of deep learning for LN analysis through EUS imaging. Reliable LN characterization is vital in determining disease stage and informing treatment pathways, with direct implications for patient management. The AI-based model demonstrated strong performance in both detection and classification tasks, highlighting its potential to enhance LN evaluation with EUS and contribute to more individualized, effective care.

Disclosure

No co

Introduction

Capsule endoscopy (CE) is commonly used as the initial exam in situations of suspected mid-gastrointestinal bleeding, after normal upper and lower endoscopy. Although the assessment of the small bowel is the primary focus of CE, detection of upstream or downstream vascular lesions may also be clinically significant. This study aimed to develop and test a Convolutional Neural Network (CNN)-based model for panendoscopic automatic detection of vascular lesions during CE.

Aims & Methods

A multicentric retrospective study was conducted, based on 1022 CE exams. We used a total of 35655 frames, from seven types of CE devices (PillCam Colon 1, PillCam Crohn's, PillCam SB1, PillCam SB3, OMOM HD, Olympus, Mirocam), of which 11091 were considered to have vascular lesions (angiectasia or varices) after triple validation. We divided data into a training and a validation set, the latter being used to evaluate the model’s performance. At time of division, all frames from a given patient were assigned to the same dataset. Our primary outcome measures were sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV) and an area under the precision-recall curve (AUC-PR).

Results

A total of 35655 frames were used, with 11091 including vascular lesions. The validation data set contained 2433 frames with normal mucosa and 933 vascular lesions.
Sensitivity and specificity were 86.4% and 98.3%, respectively. PPV was 95.2%, while the NPV was 95.0%. Overall accuracy was 95.0%. The AUC-PR value was 0.96. The CNN processed 115 frames per second.

Conclusion

The study presents promising results for the application of AI deep learning models in panendoscopic automatic detection of vascular lesions in various gastrointestinal topographies. The methodological details, including the patient-split design and the use of frames from different types of CE devices, suggest that the model could be effectively used in routine clinical practice independently of the device brand. This study highlights the potential of panendoscopic evaluation of all gastrointestinal tract and the need for further research in this area. Overall, the high diagnostic performance of this CNN in multibrand devices addresses an important issue of technological interoperability, allowing it to be replicated in multiple technological settings. However, the study has some limitations, including a small number of frames and a retrospective design, which may introduce selection bias, therefore prospective and multicentric studies are needed before introducing this deep learning model into clinical practice. This is the first proof-of concept AI deep learning model developed for panendoscopic automatic detection of vascular lesions during CE.

References

Triester, S.L., et al., A meta-analysis of the yield of capsule endoscopy compared to other diagnostic modalities in patients with obscure gastrointestinal bleeding. Am J Gastroenterol, 2005. 100(11): p. 2407-18.
Liao, Z., et al., Indications and detection, completion, and retention rates of small-bowel capsule endoscopy: a systematic review. Gastrointest Endosc, 2010. 71(2): p. 280-6.
Rondonotti, E., et al., Small-bowel capsule endoscopy and device-assisted enteroscopy for diagnosis and treatment of small-bowel disorders: European Society of Gastrointestinal Endoscopy (ESGE) Technical Review. Endoscopy, 2018. 50(4): p. 423-446.
Ribeiro, T.F.-U.I.C.H., Colon capsule endoscopy and artificial intelligence: a perfect match por panendoscopy, in Artificial Intelligence in Capsule Endoscopy, C.H. Mascarenhas M, Macedo G, Editor. 2023, Academic Press.
Tsuboi, A., et al., Artificial intelligence using a convolutional neural network for automatic detection of small-bowel angioectasia in capsule endoscopy images. Dig Endosc, 2020. 32(3): p. 382-390.
Saurin, J.C., et al., Diagnostic value of endoscopic capsule in patients with obscure digestive bleeding: blinded comparison with video push-enteroscopy. Endoscopy, 2003. 35(7): p. 576-84.
Mascarenhas, M., et al., Deep learning and colon capsule endoscopy: automatic detection of blood and colonic mucosal lesions using a convolutional neural network. Endosc Int Open, 2022. 10(2): p. E171-e177