Oral cancer is the most common cancer in India, accounting for 40% of all cancers overall and accounting for one-third of the world burden. The poor survival rate in India is mainly due to late diagnosis and the resultant progression of disease to an advanced stage at diagnosis. Therefore, there exists an urgent need for low-cost, easy to use imaging device that enables oral cancer screening and triage patients in Low- Resource Settings (LRS). The major goal of the study is to validate a low cost dual mode mobile intraoral imaging system for oral cancer detection in LRS. This project provides early detection of oral lesions and timely patient referral to specialists can avoid disease progression, reduce morbidity and mortality. The successful demonstration of this mobile dual mode intra-oral imaging screening system will lead to a sustainable solution for early detection of oral cancers in community settings, eventually improving oral cancer detection rates, treatment outcomes, and quality of life of patients in LRS.

The project team has developed mobile imaging platform that specifically addresses critical barriers to improve oral cancer screening and management in LRS. The mobile imaging platform consists of a smart phone, mobile App, intraoral imaging probe, cloud computing, and web application. More than 5,000 patients in three locations in India (cancer center, dental school, and remote region) were screened to demonstrate the performance. Remote specialists using the mobile oral cancer screening data achieved diagnostic sensitivities and specificities of 92% and 85% respectively vs. the standard-of-care (SOC). A convolutional neural network (CNN) based deep learning method for classifying oral images and achieved sensitivity of 87% and specificity of 87%, for delineating oral cancer and pre-cancer lesions. The aim of this research

is to modify the dual mode mobile imaging platform (prototype) for real time clinical use in low resource setting.