P0409 - Automatic Detection of Small Bowel Protruding Lesion in Double-balloon Enteroscopy Using a Convolutional Neural Network: A Proof-of-concept Study

Award: Presidential Poster Award

Award: Presidential Poster Award

Joao Afonso, MD¹, Miguel Mascarenhas, MD¹, Tiago Ribeiro, MD¹, João Ferreira, PhD², Hélder Cardoso, MD¹, Patrícia Andrade, MD¹, Marco Parente, PhD², Renato Natal, PhD², Guilherme Macedo, MD, PhD, FACG^1
1Centro Hospitalar de São João, Porto, Porto, Portugal; ²University of Porto, Porto, Porto, Portugal

Introduction: Double-balloon enteroscopy (DBE) allows deep exploration of the small bowel, enabling tissue sampling and the application of endoscopic therapy. DBE plays a central role in the management of patients with suspected small bowel tumors. The application of artificial intelligence (AI) algorithms to different endoscopic techniques has provided promising results. Convolutional Neural Networks (CNN) are a multi-layer artificial intelligence architecture with high performance levels for image analysis. The application of these automated algorithms for detection of lesions in DBE has not been explored. We aimed to develop and test a CNN-based algorithm for automatic detection of protruding lesions in DBE exams.

Methods: : A convolutional neural network was developed based on 72 DBE exams. A total of 7925 images were included, 2535 images containing protruding lesions (polyps, epithelial tumors, subepithelial lesions and nodules). The remaining images showed normal mucosa (n=5390). A training dataset comprising 80% of the total pool of images (n=6340) was used for development of the network. The performance of the CNN was evaluated using an independent validation dataset (20% of total image pool, n=1585). The output provided by the network was compared to a consensus classification provided by two gastroenterologists with experience in DBE (more than 200 exams each) (Figure 1a). The sensitivity, specificity, accuracy, positive and negative predictive values (PPV and NPV, respectively), and area under the curve (AUC) were calculated.

Results: Our model automatically detected small bowel protruding lesions with an accuracy of 97.3% (Figure 1b). The sensitivity, specificity, PPV and NPV were, respectively, 97.0%, 97.4%, 94.6%, and 98.6%. The AUC was 1.00 (Figure 1c). The CNN analyzed the validation dataset in 7 seconds, at a rate of approximately 239 frames per second.

Discussion: We developed a pioneer AI algorithm for automatic detection of enteric protruding lesions during DBE. The development of these tools may enhance the diagnostic yield of deep enteroscopy techniques, which may have significant impact in the management of these patients.


Figure: Figure 1: 1a - Output obtained from the application of the convolutional neural network. A blue bar represents a correct prediction. 1b - Evolution of the accuracy of the convolutional neural network during training and validation phases, as the training and validation datasets were repeatedly inputted in the neural network. 1c - ROC analyses of the network’s performance. N: normal mucosa; PR – protruding lesions.


Disclosures:
Joao Afonso indicated no relevant financial relationships.
Miguel Mascarenhas indicated no relevant financial relationships.
Tiago Ribeiro indicated no relevant financial relationships.
João Ferreira indicated no relevant financial relationships.
Hélder Cardoso indicated no relevant financial relationships.
Patrícia Andrade indicated no relevant financial relationships.
Marco Parente indicated no relevant financial relationships.
Renato Natal indicated no relevant financial relationships.
Guilherme Macedo indicated no relevant financial relationships.


Joao Afonso, MD¹, Miguel Mascarenhas, MD¹, Tiago Ribeiro, MD¹, João Ferreira, PhD², Hélder Cardoso, MD¹, Patrícia Andrade, MD¹, Marco Parente, PhD², Renato Natal, PhD², Guilherme Macedo, MD, PhD, FACG¹. P0409 - Automatic Detection of Small Bowel Protruding Lesion in Double-balloon Enteroscopy Using a Convolutional Neural Network: A Proof-of-concept Study, ACG 2021 Annual Scientific Meeting Abstracts. Las Vegas, Nevada: American College of Gastroenterology.