Rapid and Noninvasive Early Detection of Lung Cancer by Integration of Machine Learning and Salivary Metabolic Fingerprints Using MS LOC Platform: A Large‐Scale Multicenter Study

Abstract Most lung cancer (LC) patients are diagnosed at advanced stages due to the lack of effective screening tools. This multicenter study analyzes 1043 saliva samples (334 LC cases and 709 non‐LC cases) using a novel high‐throughput platform for metabolic fingerprint acquisition. Machine learning identifies 35 metabolic features distinguishing LC from non‐LC subjects, enabling the development of a classification model named SalivaMLD. In the validation set and test set, SalivaMLD demonstrates strong diagnostic performance, achieving an area under the curve of 0.849‐0.850, a sensitivity of 81.69–83.33%, and a specificity of 74.23–74.39%, outperforming conventional tumor biomarkers. Notably, SalivaMLD exhibits superior accuracy in distinguishing early stage LC patients. Hence, this rapid and noninvasive screening method may be widely applied in clinical practice for LC detection.