Evaluating the Artificial Intelligence Driven Monitoring System Notifications: Clear Aligner and Oral Hygiene Perspectives

INTRODUCTION: The widespread use of smartphones presents a remarkable opportunity for real-time management of orthodontic treatment. The development of orthodontic smartphone applications, such as Dental Monitoring (DM), epitomises the use of Artificial Intelligence Driven Remote Monitoring (AIDRM) for sophisticated image analysis. The primary aim of this study is to evaluate the validity of DM's AI image analysis notification software in detecting unfavorable clinical conditions regarding clear aligner and oral hygiene concerns. METHODS: Twenty-three patients undergoing clear aligner treatment with the implementation of DM were recruited for this study. To assess the validity of DM's AI image analysis software in detecting unfavourable clinical conditions, all notifications were compared to the 'gold standard' orthodontist visual exam of uploaded DM scans. A range of 533-1425 observations were assessed for identifying generalised and localised detection of clinical conditions regarding clear aligner and oral hygiene parameters. Then, the accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated. RESULTS: Dental Monitoring AI image analysis software demonstrated a higher accuracy for generalised detection of clear aligner and oral hygiene parameters, while a lower accuracy was demonstrated for localised detection of stated parameters. CONCLUSIONS: Dental monitoring is a viable option for clinical use in remotely monitoring clear aligner and oral hygiene parameters during clear aligner orthodontic treatment. However, the clinician should remain aware that the AI notification software has trouble identifying specific areas of clinical issues, and therefore, the monitoring is not fully automated and must be closely followed by the overseeing orthodontist.