Related Experiment Video
Updated: Aug 19, 2025
10:31
Intraoperative Gastroscopy for Tumor Localization in Laparoscopic Surgery for Gastric Adenocarcinoma
Published on: August 9, 2016
12.8K
Artificial intelligence in gastric cancer: applications and challenges
Runnan Cao1,2, Lei Tang3, Mengjie Fang1,2,4
1School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, P. R. China.
Gastroenterology Report
|December 2, 2022
Summary
Artificial intelligence (AI) aids in gastric cancer (GC) screening, diagnosis, and treatment by analyzing medical images. Challenges like data scarcity require improved AI models for better accuracy and robustness in GC management.
Area of Science:
- Oncology
- Medical Imaging
- Artificial Intelligence
Background:
- Gastric cancer (GC) presents high mortality, with diagnosis and treatment decisions heavily reliant on expert interpretation of medical images.
- Limitations in imaging quality, experience, objective criteria, and inter-observer variability hinder diagnostic accuracy in GC.
- Machine learning, particularly deep learning, offers advanced capabilities for automatic information extraction from data, with growing applications in clinical settings.
Purpose of the Study:
- To provide a comprehensive overview of current research on artificial intelligence (AI) applications in gastric cancer (GC).
- To summarize AI's role in GC screening, diagnosis, and treatment decision-making.
- To identify challenges and future directions for AI development in GC.
Main Methods:
- Review of current research literature on AI in gastric cancer.
- Analysis of AI applications in GC screening, including precancerous disease identification and early cancer detection.
- Evaluation of AI's role in GC diagnosis (TNM staging, subtype classification) and treatment (surgical margin determination, prognosis prediction).
Main Results:
- AI demonstrates potential in identifying precancerous conditions and assisting early detection of GC through endoscopic and pathological data.
- AI tools can support tumor-node-metastasis (TNM) staging and GC subtype classification.
- AI shows promise in aiding surgical margin assessment and predicting patient prognosis in GC cases.
Conclusions:
- AI offers significant potential to enhance accuracy and efficiency across the spectrum of gastric cancer management, from screening to treatment.
- Current AI approaches face challenges including data scarcity and limited interpretability, necessitating further research and development.
- Development of more regulated data, standardized procedures, and advanced algorithms is crucial for creating robust and accurate AI models for GC.
