AI-assisted screening of oral potentially malignant disorders using smartphone-based photographic images.

Simple Summary : The early detection of oral cancer is essential for improving patient outcomes. A conventional oral examination by specialists is the clinical standard for detecting oral lesions. However, many high-risk individuals in middle and low-income countries lack access to specialists. Therefore, there is a need to develop an easy-to-use, non-invasive oral screening tool that enhances the existing system for detecting precancerous lesions. This study explores artificial intelligence (AI)-based techniques to identify precancerous lesions using photographic images of oral cavities in the Indian population. The high performance of deep learning models suggests that an AI-based solution can be deployed for community screening programs in low-resource settings after further improvement and validation.