 The study proposes a classification and prediction system for gastric cancer diagnosis using saliva samples, which utilizes machine learning techniques and computer-aided diagnostic systems to improve survival rates by detecting early gastric cancer. The study uses high-performance liquid chromatography mass spectrometry, HPLCMS, to analyze 220 saliva samples and identifies 14 amino acid biomarkers for binary classification using support vector machine SVM. The proposed method achieves an overall accuracy of 97.18%, specificity of 97.44%, and sensitivity of 96.88%, with the potential for clinical translation. This article was authored by Mohamed Iqil Oslam, Tsu-Aili Shwey, Ken Wang, and others.