Artificial intelligence-driven workflow synchronizing interdisciplinary dentistry: Narrative review.
Authors
Affiliations (5)
Affiliations (5)
- Ashman Department of Periodontology and Implant Dentistry, New York University College of Dentistry, New York, USA.
- Touro College of Dental Medicine at New York Medical College, New York, USA.
- Department of Operative Dentistry, Tokyo Dental College, Tokyo, Japan.
- AI Research Unit, Graduate School of Dentistry, The University of Osaka, Osaka, Japan.
- Department of Conservative Dentistry, Periodontology and Digital Dentistry, LMU University Hospital, Munich, Germany.
Abstract
This narrative review summarizes an artificial intelligence (AI)-integrated digital workflow that enables a seamless, multidisciplinary approach to dental diagnosis and treatment planning, aiming to improve diagnostic coordination, efficiency, and interdisciplinary communication. All studies selected for this narrative review were extracted from the PubMed database. Forty studies were selected, of which 24 were used for general information and 16 were used for the statistical analysis of accuracy of AI software within data acquisition, treatment planning, and clinical implementation. The digital workflow was divided into three phases: data acquisition, treatment planning, and AI data analysis for decision support. Six articles reported the accuracy outcomes for AI-integrated data acquisition tools, such as intraoral scanners (IOS), cone-beam computed tomography (CBCT), facial scanners (FS), and jaw motion trackers (JMT); seven reported AI performance in assisting treatment planning; and three assessed the clinician acceptance rate of AI-supported decisions. IOS, CBCT, FS, and JMT achieved 88-97% overall accuracies. The integrated convolutional neural network and recurrent neural network models obtained 87-98% overall accuracy for treatment planning. Finally, the AI-generated treatment plan obtained 75-95% clinician acceptance rate. By integrating the IOS, CBCT, FS, and JMT, a comprehensive virtual patient can be created to facilitate seamless communication and effective treatment planning. This AI-integrated workflow may enhance interdisciplinary coordination and treatment planning. However, prospective clinical studies are required to validate their effects on patient outcomes and satisfaction.