Deep learning–based lesion detection and severity grading of small-bowel Crohn’s disease ulcers on double-balloon endoscopy images - 20/04/24
Abstract |
Background and Aims |
Double-balloon endoscopy (DBE) is widely used in diagnosing small-bowel Crohn’s disease (CD). However, CD misdiagnosis frequently occurs if inexperienced endoscopists cannot accurately detect the lesions. The CD evaluation may also be inaccurate owing to the subjectivity of endoscopists. This study aimed to use artificial intelligence (AI) to accurately detect and objectively assess small-bowel CD for more refined disease management.
Methods |
We collected 28,155 small-bowel DBE images from 628 patients from January 2018 to December 2022. Four expert gastroenterologists labeled the images, and at least 2 endoscopists made the final decision with agreement. A state-of-the-art deep learning model, EfficientNet-b5, was trained to detect CD lesions and evaluate CD ulcers. The detection included lesions of ulcer, noninflammatory stenosis, and inflammatory stenosis. Ulcer grading included ulcerated surface, ulcer size, and ulcer depth. A comparison of AI model performance with endoscopists was performed.
Results |
The EfficientNet-b5 achieved high accuracies of 96.3% (95% confidence interval [CI], 95.7%-96.7%), 95.7% (95% CI, 95.1%-96.2%), and 96.7% (95% CI, 96.2%-97.2%) for the detection of ulcers, noninflammatory stenosis, and inflammatory stenosis, respectively. In ulcer grading, the EfficientNet-b5 exhibited average accuracies of 87.3% (95% CI, 84.6%-89.6%) for grading the ulcerated surface, 87.8% (95% CI, 85.0%-90.2%) for grading the size of ulcers, and 85.2% (95% CI, 83.2%-87.0%) for ulcer depth assessment.
Conclusions |
The EfficientNet-b5 achieved high accuracy in detecting CD lesions and grading CD ulcers. The AI model can provide expert-level accuracy and objective evaluation of small-bowel CD to optimize the clinical treatment plans.
Le texte complet de cet article est disponible en PDF.Abbreviations : AI, CD, CDEIS, CNN, DBE, Grad-CAM, SES-CD, AUC, ROC, PPV, NPV
Plan
Vol 99 - N° 5
P. 767 - mai 2024 Retour au numéro
Article précédent
- Validation of a new optical diagnosis training module to improve dysplasia characterization in inflammatory bowel disease: a multicenter international study
- Marietta Iacucci, Stefanos Bonovas, Alina Bazarova, Rosanna Cannatelli, Richard J.M. Ingram, Nunzia Labarile, Olga Maria Nardone, Tommaso Lorenzo Parigi, Daniele Piovani, Keith Siau, Samuel C.L. Smith, Irene Zammarchi, Jose G.P. Ferraz, Gionata Fiorino, Ralph Kiesslich, Remo Panaccione, Adolfo Parra-Blanco, Mariabeatrice Principi, Gian Eugenio Tontini, Toshio Uraoka, Subrata Ghosh, OPTIC-IBD Study Group, Abdullah Abbasi, Adele Wargen, Ahmed Feroz, Alessandra Dell’Era, Alessandra Piagnani, Alessandro Rimondi, Alessia Chini, Alessia D. Guarino, Alessia Todeschini, Amar Srinivasa, Andrea Sorge, Angelica Toppeta, Anna M.C. Gabrielli, Anna Testa, Anthony MacLean, Antonella Contaldo, Antonia Churchhouse, Anupama De Silva, Beatrice Marinoni, Chiara Lillo, Christopher N. Andrews, Ciro Lentano, Costantino Sgamato, Daniele Gridavilla, Daniele Noviello, Danny Cheung, Dhanai Di Paolo, Domenico Novielli, Dominic King, Edoardo Borsotti, Eleanor Liu, Elena Arsiè, Elisa Farina, Elisabetta Filippi, Emanuele Annoscia, Fabiana Castiglione, Fenella Marley, Francesca Ferretti, Francesco Conforti, Francis Egbuonu, Fulvio S. D’Abramo, Giulia Scardino, Giuseppe Indellicati, Giuseppe Losurdo, Antonietta Gerarda Gravina, Ian Beales, Ibrahim Al Bakir, Ilaria Ditonno, Imma Di Luna, Imran Tahir, Irene Bergns, Irene V. Brescia, Isabel Carbery, Ismaeel Al-Talib, Jawad Azhar, Jeffrey Butterworth, Joel James, Joëlle St-Pierre, John Jacob, Jordan Iannuzzi, Katharine Todd, Kelly Chatten, Leah Gilroy, Lekshmy S. Pillai, Luca Pastorelli, Lucienne Pellegrini, Lushen Pillay, Marco Romano, Maria C. Monico, Mariapaola Piazzolla, Marius Paraoan, Marta Patturelli, Martino Mezzapesa, Matthew Woo, Maxime Delisle, Melissa Chan, Michael Gomez, Michael Z. Ma, Milli Gupta, Misha Kabir, Mohammad F. Khattak, Mohit Inani, Muaad Abdulla, Muhammad Saad, Munaa Khaliq-Kareemi, Nauman Idrees, Nick Burr, Nicola Henderson, Nurulamin Noor, Odhran Doherty, Oliver Bendall, Oriana Olmo, Philip Harvey, Philip Oppong, Puja Kumar, Rachid Mohamed, Rahman H.M. Abdul, Rebecca Carruthers, Rebecca O’Kane, Roberto de Sire, Salvatore Rizzi, Samantha Horley, Sarah Al-Shakhshir, Sarah Townsend, Sherif Abdelbadiee, Sofia Ridolfo, Sonika Sethi, Stefania De Lisi, Stefania Marangi, Tim Ambrose, Tom Troth, Vincenzo Occhipinti, Wai L. Lam, Yasmin Nasser, Zia Rahman
- Endoscopic powered resection device for residual colonic lesions: the first multicenter, prospective, international clinical study
- Mate Knabe, Roberta Maselli, Elodie Cesbron-Metivier, Stephan Hollerbach, Lucio Petruzziello, Frédéric Prat, Harshit S. Khara, Mathieu Pioche, Dirk Hartmann, Paola Cesaro, Federico Barbaro, Arthur Berger, Cristiano Spada, David L. Diehl, Andrea May, Thierry Ponchon, Alessandro Repici, Guido Costamagna
Bienvenue sur EM-consulte, la référence des professionnels de santé.
L’accès au texte intégral de cet article nécessite un abonnement.
Déjà abonné à cette revue ?