Abstract
Objectives
This study aimed to detect and segment target teeth and score on each individual teeth according to the Quigley-Hein plaque index (QHI) by using multi-view intraoral images and a deep learning approach.
Material and Methods
A dataset of intraoral images captured from both frontal and lateral views of permanent and deciduous dentitions was utilized. The dataset comprised of 210 photographs taken after applying a plaque disclosing agent. A three-stage method was employed, where the YOLOv8 model was first used to detect the target teeth, followed by the prompt-based SAM (Segment Anything Model) segmentation algorithm to segment teeth. A new single-tooth dataset consisting of 1400 photographs was obtained after applying a two-stage method. Finally, the multi-class classification model DeepPlaq we implemented was trained and evaluated on the accuracy of dental plaque indexing based on the QHI scoring system. Classification performance was measured using accuracy, recall, precision, and F1-score.
Results
The teeth detector exhibited an accuracy (mean average precision, mAP) of approximately 0.941 ± 0.005 in identifying teeth with plaque disclosing agents. The maximum accuracy attained in the plaque indexing through DeepPlaq was 0.84 (probability that DeepPlaq scored identical to an expert), and the average scoring error was less than 0.25 for a 0 to 5 scoring setting.
Conclusions
A three-stage approach demonstrated excellent performance in detecting and segmenting target teeth, and DeepPlaq model also showed strong performance in assessing dental plaque indices.
Clinical relevance
The evaluation of dental plaque indices using deep learning algorithms alleviated the burdensome and repetitive tasks of doctors, enabling quicker and more reliable decision-making.