AI Revolutionizes Early Detection and Diagnosis of Gastric Cancer
Artificial intelligence (AI) is rapidly transforming the landscape of early gastric cancer (EGC) detection and treatment, offering promising solutions to improve diagnostic accuracy, treatment planning, and patient outcomes, according to an article published in Annals of Medicine.
Gastric cancer is one of the most common malignancies worldwide, with aggressive characteristics that contribute to poor survival rates, especially in advanced stages. While early diagnosis can lead to a 5-year survival rate exceeding 90%, traditional endoscopic methods still face challenges in detecting early-stage lesions. Human error, variability in expertise among endoscopists, and limited access to advanced imaging technologies contribute to missed diagnoses and has led to growing interest in AI-assisted endoscopy.
Studies show that AI systems can achieve diagnostic accuracy comparable to, or even exceeding, that of experienced endoscopists using both white light endoscopy (WLE) and image-enhanced endoscopy (IEE) techniques, such as narrow-band imaging (NBI) and magnifying endoscopy (ME). These systems can assist endoscopists in identifying lesions that might otherwise be missed, particularly in cases of poorly characterized or small EGCs.
AI is also valuable in determining the depth of invasion of EGC, a critical factor in deciding between endoscopic submucosal dissection (ESD) and surgical resection. AI systems show promising accuracy in predicting submucosal infiltration, potentially reducing the need for additional invasive procedures. Moreover, AI can aid in determining the differentiation type of EGC (differentiated vs. undifferentiated), which influences treatment strategies and prognostic assessment. Studies have demonstrated AI's potential in accurately assessing the degree of EGC differentiation, addressing the limitations of traditional biopsy pathology.
Accurate delineation of EGC boundaries is crucial for ensuring complete resection during ESD. AI systems are being developed to precisely define these boundaries, minimizing the risk of incomplete resection and subsequent recurrence.
While AI offers numerous advantages in EGC management, it is not intended to replace human expertise. Instead, AI serves as a powerful tool to augment endoscopists' abilities, improving diagnostic accuracy, streamlining treatment planning, and enhancing patient care. As AI technology continues to evolve, it holds the potential to further revolutionize the field of gastroenterology and improve outcomes for individuals at risk of or diagnosed with early gastric cancer.
“Looking forward, with the inclusion of more high-quality and comprehensive research data, as well as more prospective and multicenter studies, we have reason to believe that AI will play a greater role in the diagnosis and treatment of endoscopists and bring more convenient and safer services to patients,” concluded the study authors.
Reference
Lei C, Sun W, Wang K, Weng R, Kan X, Li R. Artificial intelligence-assisted diagnosis of early gastric cancer: present practice and future prospects. Ann Med. 2025;57(1):2461679. doi:10.1080/07853890.2025.2461679
