AI app boosts gum disease patient compliance in new clinical trial

BACKGROUND: The purpose of this study is to report on the development of an artificial intelligence (AI) model designed to improve compliance in periodontal maintenance patients. METHODS: Fifty adult patients (32 females and 18 males; aged 25 to 78 years) with diagnosis and treatment of Generalized or Localized Periodontitis stages I, II, or III; grades A-C who were following a 3-month periodontal maintenance interval were provided an AI-based smartphone application (app) that provided personalized educational, motivational, and risk-awareness information every week between maintenance visits. Recession (REC), probing pocket depth (PPD), clinical attachment level (CAL), plaque index (PI), bleeding on probing (BOP), and gingival index (GI) captured at each maintenance visit out to 6 months were compared with baseline. In-office interviews were conducted to assess subjective data points, and cognitive interviews were conducted to validate patient-reported outcomes (PRO). RESULTS: REC, PPD, and CAL remained stable through all timepoints. PI, BOP, and GI demonstrated significant improvement compared with baseline (p < 0.01). PRO were validated and revealed that most subjects found the app to be easy to use, desired more interaction, perceived it as educational by raising their awareness of their oral hygiene, and would recommend the use of the app to a friend. CONCLUSIONS: An AI-based smartphone application was trained on a private practice maintenance population of three periodontists. The subjects demonstrated improvements in PI, BOP, and GI with no changes in REC, PPD, or CAL. Patients found the app to be educational and beneficial by improving their understanding of their periodontal disease risk. PLAIN LANGUAGE SUMMARY: An artificial intelligence model (AI) was developed to improve periodontal maintenance compliance. Compliance with periodontal maintenance in patients decreases dramatically after 1 year. This is a proof-of-concept study using an AI-based smartphone application (app) that communicates individualized periodontal disease risk awareness messages that were educational and motivational to effect a behavioral change in hopes of improving compliance. Measurements of disease stability/instability, such as gum inflammation, bleeding, bone loss, and assessment of brushing effectiveness, were captured at baseline and at subsequent office visits every 3 months for a total of 6 months. Subjective patient information was captured via in-office interviews to supplement clinical data and provide the AI model with multimodal data points. Patient-reported outcome surveys were developed via cognitive interviews. Results showed improvements in brushing effectiveness, which translated to less gum bleeding and inflammation. Other clinical measurements of disease activity showed no change, which could translate to a lack of disease progression. Patient-reported outcome development revealed that most of the subjects found the interaction with the app to be positive and beneficial. This proof-of-concept study demonstrates a novel approach using an AI-based smartphone app to be worthy of further investigation on efficacy and possibly defines an objective measure of patient compliance.