This clinical study aims to develop and evaluate an artificial intelligence (AI)-driven virtual biopsy technology for the diagnosis of gastric cancer with positive peritoneal exfoliative cytology (PEC). Gastric cancer with peritoneal metastasis often presents a challenge for early detection and diagnosis, with traditional diagnostic methods such as imaging and histopathology being limited in sensitivity and specificity. In this study, we propose the use of AI algorithms to analyze non-invasive biomarkers, including transcriptomic profiles and imaging data, to predict the presence of peritoneal exfoliative cytology-positive gastric cancer. Virtual biopsy leverages AI to integrate multiple datasets, providing a comprehensive diagnostic tool that could potentially replace or supplement current invasive diagnostic procedures. By developing this technology, we aim to improve the early diagnosis and monitoring of gastric cancer, particularly in cases with occult peritoneal metastasis, and ultimately enhance patient outcomes through more timely and accurate treatment strategies. The study will involve the collection of clinical samples from gastric cancer patients with suspected peritoneal metastasis. The AI model will be trained on these samples to identify relevant biomarkers for PEC-positive gastric cancer. Clinical validation will be conducted to assess the performance of this AI-driven virtual biopsy system compared to conventional diagnostic methods. This study has the potential to provide a novel, non-invasive diagnostic approach for gastric cancer with peritoneal involvement, offering a significant advancement in the field of early cancer detection and personalized medicine.