Artificial Intelligence-Driven Salivary Biomarkers for Detection and Risk Stratification of Oral Potentially Malignant Disorders and OSCC: A Protocol for a Systematic Review
Shazia Fathima J.H, Mohammed Sibgatullah, Mohmed Isaqali KarobariObjective
This systematic review protocol aims to synthesize evidence on artificial intelligence-driven analysis of salivary biomarkers for the detection, risk stratification and differentiation of OPMDs and OSCC, with particular attention to how AI models are integrated and validated rather than to their performance alone. Secondary objectives include examining study design features, internal and external validation approaches, risk of bias, recurring methodological shortcomings and evidence gaps and the clinical applicability of AI-assisted salivary diagnostics.
Methods
This systematic review protocol is registered in the International Prospective Register of Systematic Reviews (PROSPERO) under registration number XXX and has been developed following the guidelines of the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P). A comprehensive search will be performed across multiple electronic databases using prospectively developed and replicable search strategies. The review will include observational designs, diagnostic test accuracy investigations and model development or validation studies that apply artificial intelligence in patients with OPMDs or OSCC.
Expected Results
This systematic review will provide a systematic and scientifically valid integration of the available evidence on artificial intelligence-assisted salivary biomarker models in OPMDs and OSCC. Although there is a growing body of literature suggesting the promising performance of ML and DL models, the current literature is marked by considerable heterogeneity in terms of methodological quality, validation approaches and reporting.
Conclusion
By evaluating biomarker types, analysis methods, validation strategies and risk of bias, this systematic review is intended to enhance the existing knowledge base regarding the application of AI techniques to salivary biomarkers in OPMDs and OSCC, providing insight into the current evidence and the areas that need further exploration.