Early gastric cancer (EGC) is often difficult to detect accurately during endoscopic examination due to subtle morphological features and variability among endoscopists. Artificial intelligence (AI) has shown promise in improving diagnostic performance; however, most existing models lack interpretability and rely on single-modality imaging.
This study aims to develop and evaluate an explainable multimodal artificial intelligence model for the diagnosis of early gastric cancer using endoscopic imaging. The model integrates features derived from white-light imaging and image-enhanced endoscopy, along with quantitative image features and clinical data, to improve diagnostic accuracy and provide interpretable decision support.
The primary outcome is the diagnostic performance of the AI model for detecting early gastric cancer, evaluated by area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity.
The results of this study are expected to provide evidence for the clinical utility of explainable AI in endoscopic diagnosis and support the development of reliable human-AI collaborative diagnostic systems.
Age ≥18 years
Suspicious gastric lesions identified on white-light imaging (WLI)
Preoperative biopsy indicating precancerous lesions (dysplasia or intraepithelial neoplasia) or adenocarcinoma, with preoperative magnifying endoscopy with narrow-band imaging (ME-NBI) performed
Patients meeting the absolute indications for endoscopic submucosal dissection (ESD) and who underwent ESD
Non-adenocarcinoma histological types (e.g., lymphoma)
Patients who did not undergo ME-NBI examination or did not receive ESD
Lesions invading the muscularis propria or deeper layers
Missing or indeterminate postoperative histopathological results
Patients with suspected gastric lesions who underwent endoscopic examination and endoscopic submucosal dissection (ESD), with available white-light imaging (WLI), magnifying endoscopy with narrow-band imaging (ME-NBI), and histopathological diagnosis at participating centers.
Cohort
Prospective
No intervention (observational study)
Other interventionThis is an observational study with no intervention assigned to participants.
Diagnostic performance of the artificial intelligence model for detecting early gastric cancer
The primary outcome is the diagnostic performance of the artificial intelligence model for identifying early gastric cancer, evaluated by the area under the receiver operating characteristic curve (AUROC), sensitivity, specificity, accuracy, and F1 score, using histopathological diagnosis as the reference standard.
Click on the lead sponsor to view all of their trials.
