Accuracy of smartphone based photography in screening for potentially malignant lesions among a rural population in Tamil Nadu: A cross-sectional study

Background and Objective: Oral cancer is a major public health problem which carries significant morbidity and mortality. A shift from treatment to prevention by screening is the key to reduce oral cancer lesion among population. Searching for an affordable and viable alternative to face-to-face screening that can expedite diagnosis of oral diseases among rural population with good accuracy is mandatory. One of the most realistic solutions to acknowledge this hurdle and the unavailability of dental professionals, is mobile teledentistry. Materials and Methods: Secondary data analysis was conducted, in which the data were derived from the project of “Oral Cancer Screening Program in Rural population” conducted by Thirumalai Mission Hospital, Ranipet. Ninety-six biopsies were taken for the patients who had visible oral lesions which had been provisionally diagnosed on clinical examination. Oral screening was carried out by unaided face-to-face screening method by a trained and calibrated dentist. In a separate subsequent visit, a trained teledental assistant took photographs of each participant's mouth by using a smartphone camera; the charting of the photographs was conducted independently by two dentists. Results: Intra-examiner reliability of Examiner 1 and Examiner 2 was 0.943 and 0.921, respectively. Inter examiner reliability score of 0.879 was obtained between both the examiners by the photographic method of diagnosis. Intraclass correlation coefficient between two methods of examination was 0.812. Agreement between the photographic examination (Examiner 1, Examiner 2) with the gold standard biopsy report was 0.791 and 0.855, respectively. Conclusion: Smartphone camera use offers a valid and reliable means of remote screening for oral lesions. Photographs of the oral lesions taken from the smart-phone camera with an acceptable diagnostic validity and reliability.