P2025 - Artificial Intelligence and Capsule Endoscopy: Automatic Detection of Multiple Gastric Lesions in Wireless Capsule Endoscopy

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

Introduction: In recent years, capsule endoscopy (CE) has become a minimally invasive alternative to conventional esophagogastroduodenoscopy for assessment of upper gastrointestinal disease. However, the performance of CE for detection of gastric lesions has been shown to be suboptimal. 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 gastric lesions in wireless capsule endoscopy has not been explored.

Methods: Our group aimed to develop and test a CNN-based algorithm for the automatic detection of multiple gastric lesions, including vascular lesions (angiectasia, varices and red spots), protruding lesions, ulcers and erosions.
A total of 1483 CE exams from a single center were included to develop a CNN capable of automatic detection of multiple gastric lesions. Selected images of gastric lesions were inserted into a CNN model with transfer learning (n = 3735). A training dataset was used for development of the model (80% of the full image dataset, n = 3735). The performance of the network was evaluated using an independent dataset (20% of the full image dataset, n = 747). The output provided by the network was compared to a consensus classification provided by two gastroenterologists with experience in CE.
The performance of the network was evaluated by calculating its sensitivity, specificity, accuracy, positive predictive and negative predictive values (PPV and NPV, respectively), and area under the receiving operating characteristic curve (AUC).

Results: The trained CNN automatically detected gastric lesions with a sensitivity of 98.2%, a specificity of 99.4%, and a PPV and NPV of 99.8% and 94.6%. The overall accuracy was 98.5%. The AUC was 1.00. The CNN’s image processing time was 68 images per second. A subanalysis of the CNN performance in each subgroup (vascular lesions; ulcers/erosions; protruding lesions; blood/hematic residues) was performed. The neural network excelled in all the subgroups, with the best results in the subgroup of blood/hematic residues (overal accuracy of 99.3%) and the worst performance in the detection of ulcers and erosions (overal accuracy of 93.8%).

Discussion: Our group developed, for the first time, a CNN capable of automatically detecting gastric lesions in wireless capsule endoscopy with high accuracy, sensitivity and specificity.


Figure: Figure 1: 1a - Output obtained from the application of the convolutional neural network. A blue bar represents a correct prediction. A red bar represents an incorrect 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 for detection of different types of lesions. CNN – convolutional neural network; N – normal mucosa; P3 – Blood or hematic residues; PP - Pleomorphic gastric lesions


Disclosures:
Miguel Mascarenhas indicated no relevant financial relationships.
Joao Afonso 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.
Renato Natal indicated no relevant financial relationships.
Guilherme Macedo indicated no relevant financial relationships.


Miguel Mascarenhas, MD¹, Joao Afonso, MD¹, Tiago Ribeiro, MD¹, João Ferreira, PhD², Hélder Cardoso, MD¹, Patrícia Andrade, MD¹, Renato Natal, PhD², Guilherme Macedo, MD, PhD, FACG¹. P2025 - Artificial Intelligence and Capsule Endoscopy: Automatic Detection of Multiple Gastric Lesions in Wireless Capsule Endoscopy, ACG 2021 Annual Scientific Meeting Abstracts. Las Vegas, Nevada: American College of Gastroenterology.