JMIR Publications

ABSTRACT

Background:

The authors previously established a deep-learning-based clinical decision support system (CDSS) for real-time endoscopy-based detection and classification of gastric neoplasms. However, preneoplastic conditions such as atrophy and intestinal metaplasia (IM) were not taken into account, and there is no established model that classifies all stages of gastric carcinogenesis.

Objective:

This study aimed to establish CDSS in real-time endoscopy for all stages of gastric carcinogenesis, including atrophy and IM.

Methods:

A total of 11,868 endoscopic images were used for the training and internal-testing. The primary outcomes were: 1. The classification model's lesion-classification accuracy (6 classes; advanced-, early gastric cancer, dysplasia, atrophy, IM, normal). 2. Atrophy and IM lesion-segmentation rates for the segmentation model. The following tests were carried out to validate the performance of lesion-classification accuracy; 1. External-test using novel 1,282 images from another institution. 2. Evaluation of the classification accuracy of atrophy and IM in real-world procedures in a prospective manner. One CDSS was constructed by combining the established 6 class lesion-classification model and the preneoplastic lesion-segmentation model with the previously established lesion-detection model.

Results:

The overall lesion-classification accuracy was 90.3% (95% confidence interval: 89.0-91.6%) in the internal-test. For the performance validation, the CDSS achieved 85.3% (83.4-97.2%) of overall accuracy. Per-class external-test accuracy of atrophy and IM was 95.3% (92.6-98.0%) and 89.3% (85.4-93.2%), respectively. CDSS-assisted endoscopy showed accuracy of 92.1% (88.8-95.4%) for atrophy and 95.5% (92.0-99.0%) for IM in the real-world application of 522 consecutive screening endoscopy. The CDSS demonstrated a segmentation rate of 93.4% (92.4-94.4%) for atrophy or IM lesion segmentation in the internal-test.

Conclusions:

The CDSS demonstrated high performance in terms of computer-aided diagnosis of all stages of gastric carcinogenesis and demonstrated real-world application potential.