A Web-based Dental Caries Detection System Using Image Recognition and Deep Learning with SUS and UTAUT Analysis

Abstract

The vital processes of a clinic system are the diagnostic process as well as storing patient dental records, forms, reports and other medical notes. This study created a web-based application that has an automatic dental caries detection and assessment tool, along with an online storage for patient dental records. The dental caries detection tool enabled the user to upload dental images, and output the severity class level, based on the ICDAS II criteria. This tool incorporated deep learning algorithms and uses the ImageAI Python library. Moreover, the web app works as an online dental diagnostic service and is able to store the patient's dental records and information in a database. Furthermore, SUS and UTAUT questionnaires were used to assess the usability. The SUS resulted with an average standard deviation of 0.7765, that interpreted the respondents’ answers to be consistent and a mean score of 76.25 which shows that most of the users found the system usable. The Unified Theory of Acceptance and Use of Technology (UTAUT) raw scores revealed high marks towards the questions under attitude toward using technology, social influence, and behavioral intention.