Abstract
BACKGROUND: To develop a new deep learning model for detecting white spot lesions (WSLs), which are commonly observed in patients undergoing orthodontic treatment, and assess its accuracy. METHODS: A total of 653 intra-oral photographs of WSLs were collected and annotated. Our novel model, tooth-to-WSL You Only Look Once (TW-YOLO), and the original YOLOv5 model were fine-tuned and evaluated, with 457 photographs used for training; 130, for validation; and 66, for hold-out testing. Cohen's kappa coefficient between model prediction and orthodontist annotation was used as the primary evaluation metric, and mean average precision (mAP@0.5:0.95), average precision (mAP@0.5) and F1 score were also evaluated. The score-CAM technique was used for explainability analysis. RESULTS: Cohen's kappa coefficient values were 0.76 and 0.62 for TW-YOLO and YOLOv5, respectively. The mAP@0.5 and mAP@0.5:0.95 were 0.78, 0.51 for TW-YOLO and 0.69, 0.45 for YOLOv5, respectively. Explainability analysis suggested that the TW-YOLO model could implicitly learn the distribution pattern of WSLs by shifting more attention toward these regions. CONCLUSION: Compared to original YOLO model, our novel TW-YOLO model demonstrated improved accuracy. Smaller proportion of small sized object and examine tooth enamel at original resolution contributed to this improvement.