Radiomics in the Diagnosis of Gastric Cancer: Current Status and Future Perspectives.

Abstract

Gastric cancer is a malignant cancerous lesion with high morbidity and mortality. Preoperative diagnosis of gastric cancer is challenging owing to the presentation of atypical symptoms and the diversity of occurrence of focal gastric lesions. Therefore, an endoscopic biopsy is used to diagnose gastric cancer in combination with imaging examination for a comprehensive evaluation of the local tumor range (T), lymph node status (N), and distant metastasis (M). The resolution of imaging examinations has significantly improved with the technological advancement in this sector. However, imaging examinations can barely provide valuable information. In clinical practice, an examination method that can provide information on the biological behavior of the tumor is critical to strategizing the treatment plan. Artificial intelligence (AI) allows for such an inspection procedure by reflecting the histological features of lesions using quantitative information extracted from images. Currently, AI is widely employed across various medical fields, especially in the processing of medical images. The basic application process of radiomics has been described in this study, and its role in clinical studies of gastric cancer has been discussed.