Early Prediction of Periodontitis Risk Level with Saliva-based Advanced Learning biosensor (SAL 9000) Using Proteomic Biomarker MMP-9

H. Kanniyappan, S. Prithweeraj, L. Barba, S. Mamidi, R. Wang
University of Illinois, Illinois, United States

Poster stand number: T133

Keywords: Periodontitis, MMP-9, Saliva-Based, Electrochemistry, Machine Learning

Periodontitis is marked by the inflammation and peeling away of the gum tissue, with bone and tooth loss in advanced stages. It affects nearly half of all Americans, including over 70 percent of the elderly [1]. The enzyme MMP-9 (matrix metalloproteinase 9) is a core component of the causal pathway behind periodontitis and comprises the most important pathway in tissue destruction associated with periodontal disease [2]. To that end, the measurements of MMP levels may be useful markers for early detection of periodontitis and as a tool to assess prognostic follow-ups. Electrochemical biosensors, combined with machine learning, maybe a viable tool for the early detection of periodontitis risk, using the biomarker MMP-9. Hence, we hypothesize the development of a SAL 9000 Saliva-based Advanced Learning biosensor can predict periodontitis risk levels using MMP-9 as the prognostic biomarker. Firstly, we bound the MMP-9 antibody to the DSP, and protein to the antibody. Secondly, we developed a support-vector (SVM) machine learning program model and ran it using the electrochemical results. Therefore, detection and inhibition of MMPs be useful in periodontal disease prevention or be an essential part of periodontal disease therapy may greatly improve oral health globally.