Artificial intelligence across the gastric cancer care continuum: Clinical evidence, multimodal integration, and translation
Abstract
Artificial intelligence (AI) is increasingly being applied across the gastric cancer care continuum, but its clinical maturity differs substantially among endoscopy, cross-sectional imaging, digital pathology, prognostic modeling, treatment selection, and surgical support. Existing reviews have often focused on individual modalities or summarized algorithmic performance without systematically distinguishing technical feasibility from translational readiness. We conducted a structured narrative review of studies indexed in PubMed/MEDLINE and the Web of Science Core Collection from January 2022 through July 2026, supplemented by backward and forward citation tracking. Earlier landmark studies were retained when necessary to explain the development of major applications. Priority was given to prospective, multicenter, externally validated, and clinically actionable studies. Evidence maturity was assessed according to study design, validation breadth, workflow relevance, calibration and clinical-utility assessment, and implementation evidence. The main constraint on gastric cancer AI is no longer pattern-recognition capability but the absence of reproducible clinical benefit under real-world conditions. Priorities include multiregional prospective validation, standardized calibration and decision curve assessment, comparison of clinicians with and without AI, integration into existing workflows, evaluation of economic and patient-level outcomes, and post-deployment monitoring.
Related
- Clinical Translation of Artificial Intelligence in Oncology: Real-World Validation, Workflow Integration, and Precision Medicine Applications
- Artificial Intelligence, Multi-Omics, and Multimodal Integration in Precision Radiotherapy for Head and Neck Cancer: From Biological Insights to Treatment Decisions
- Artificial intelligence for ophthalmic vessel segmentation: from multimodal imaging to clinical translation
- Artificial Intelligence for Thyroid Nodule Diagnosis: From Multimodal Integration to Clinical Utility.
- Advances in Multimodal Artificial Intelligence in Radiology: Data Integration, Foundation Models, and Clinical Applications-A Narrative Review.
BibTeX
@article{yang2026artificial,
title = {{Artificial intelligence across the gastric cancer care continuum: Clinical evidence, multimodal integration, and translation}},
author = {Junsong Yang and Jin Tang and Jinsong He},
journal = {iScience},
year = {2026},
month = oct,
doi = {10.1016/j.isci.2026.117725},
url = {https://www.cell.com/iscience/fulltext/S2589-0042(26)03106-8},
note = {CC BY-NC 4.0},
}
RIS
TY - JOUR
TI - Artificial intelligence across the gastric cancer care continuum: Clinical evidence, multimodal integration, and translation
AU - Junsong Yang
AU - Jin Tang
AU - Jinsong He
PY - 2026
DA - 2026/10/22
JO - iScience
DO - 10.1016/j.isci.2026.117725
UR - https://www.cell.com/iscience/fulltext/S2589-0042(26)03106-8
AB - Artificial intelligence (AI) is increasingly being applied across the gastric cancer care continuum, but its clinical maturity differs substantially among endoscopy, cross-sectional imaging, digital pathology, prognostic modeling, treatment selection, and surgical support. Existing reviews have often focused on individual modalities or summarized algorithmic performance without systematically distinguishing technical feasibility from translational readiness. We conducted a structured narrative review of studies indexed in PubMed/MEDLINE and the Web of Science Core Collection from January 2022 through July 2026, supplemented by backward and forward citation tracking. Earlier landmark studies were retained when necessary to explain the development of major applications. Priority was given to prospective, multicenter, externally validated, and clinically actionable studies. Evidence maturity was assessed according to study design, validation breadth, workflow relevance, calibration and clinical-utility assessment, and implementation evidence. The main constraint on gastric cancer AI is no longer pattern-recognition capability but the absence of reproducible clinical benefit under real-world conditions. Priorities include multiregional prospective validation, standardized calibration and decision curve assessment, comparison of clinicians with and without AI, integration into existing workflows, evaluation of economic and patient-level outcomes, and post-deployment monitoring.
SN - 2589-0042
ER -