AI in gastric cancer imaging: current advances, barriers, and strategies

AI using deep learning and radiomics shows potential in gastric cancer imaging for detection, staging, treatment response and prognosis, but clinical translation is limited by data heterogeneity, poor generalizability, and lack of prospective evidence.

• Deep learning and radiomics are applied to lesion detection, staging, treatment response, and prognosis in gastric cancer.
• Key barriers to clinical translation include data heterogeneity/silos, limited generalizability, poor interpretability, and insufficient prospective evidence and regulatory frameworks.
• Proposed next steps: data governance, technological innovation, clinical translation, and physician-AI collaboration.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.