Utility of Artificial Intelligence in Detecting Lymph Node Metastasis in Oral Cancer: A Systematic Review

Oral squamous cell carcinoma (OSCC) is associated with considerable morbidity because of occult lymph node metastasis (OLNM). Conventional imaging has limitations in detecting OLNM in oral cancer. AI-based technologies have been proposed as potential aids in diagnosis. This systematic review was carried out following PRISMA 2020 guidelines and registered in the International Prospective Register of Systematic Reviews (PROSPERO) (February 2025). PubMed, Scopus, Embase, Web of Science, and Google Scholar were searched. Studies utilizing artificial intelligence (AI)-based techniques, such as machine learning (ML), deep learning (DL), and convolutional neural networks (CNNs), in the detection of lymph node metastasis in oral cancer with reported results of diagnostic accuracy, sensitivity, specificity, or area under the curve (AUC) were included. Methodological quality was assessed by Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2); evidence certainty was assessed by Grading of Recommendations, Assessment, Development, and Evaluation (GRADE). A total of 15 articles (n = 5,758) fulfilled the inclusion criteria. The AI-based models yielded AUC values of 0.729-0.961, sensitivity of 0.65-1.00, and specificity of 0.576-0.98. DL-based CNN architectures and radiomics-DL hybrid models showed the highest diagnostic accuracy. Moderate overall risk of bias was reported with the QUADAS-2 tool, while GRADE provided certainty of evidence as low to very low due to study heterogeneity and lack of external validation. The techniques employing AI appear to have great potential for diagnosing OLNM in patients with mouth cancer because they have more accurate capabilities than imaging. There is a need for future multicenter validation, standardization of algorithms, and ethical frameworks for their application.