Detection of dental caries in oral photographs taken by mobile phones based on the YOLOv3 algorithm

Baichen Ding; Zhuo Zhang; Yiran Liang; Weiwei Wang; Siwei Hao; Ze Meng; Lian Guan; Ying Hu; Bin Guo; Runlian Zhao; Yan Lv

doi:10.21037/atm-21-4805

Detection of dental caries in oral photographs taken by mobile phones based on the YOLOv3 algorithm

Ann Transl Med. 2021 Nov;9(21):1622. doi: 10.21037/atm-21-4805.

Authors

Baichen Ding^#^{1

2}, Zhuo Zhang^#³, Yiran Liang⁴, Weiwei Wang³, Siwei Hao², Ze Meng⁵, Lian Guan⁶, Ying Hu⁷, Bin Guo², Runlian Zhao³, Yan Lv⁷

Affiliations

¹ Medical School of Chinese People's Liberation Army (PLA), Beijing, China.
² Department of Stomatology, the First Medical Centre, Chinese PLA General Hospital, Beijing, China.
³ The College of Information Science and Technology, Beijing University of Chemical Technology, Beijing, China.
⁴ Department of Endodontics, Beijing Stomatological Hospital and School of Stomatology, Capital Medical University, Beijing, China.
⁵ Department of Stomatology, Handan First Hospital, Handan, China.
⁶ Department of Stomatology, Beijing Hospital of Integrated Traditional Chinese and Western Medicine, Beijing, China.
⁷ Beijing Institute of Dental Research, Beijing Stomatological Hospital and School of Stomatology, Capital Medical University, Beijing, China.

^# Contributed equally.

Abstract

Background: The aim of this study was to explore the value of using the YOLOv3 algorithm for detection and diagnosis of dental caries in oral photographs taken with mobile phones.

Methods: Oral photographs taken with the mobile phones of 570 patients were used as 3 datasets: the augmented images (n=3,990), the enhanced images (n=3,990), and the combined augmented and enhanced images (n=7,980). Oral photographs taken by mobile phones from another 70 patients were used as an independent test set. We used the YOLOv3 network for migration learning for modelling. Diagnostic precision, recall, F1-score, and mean average precision (mAP) were calculated to obtain the detection and diagnostic performance of the YOLOv3 algorithm.

Results: After 3 independent training, the mAP value of the original group YOLOv3 algorithm was 56.20%, in which the precision for primary caries recognition was 76.92%, recall was 49.59%, and F1-score was 0.60; the precision for secondary caries recognition was 91.67%, recall was 52.38%, and F1-score was 0.67. The mAP value of the enhance group algorithm was 66.69%, in which the precision for primary caries identification was 81.82%, recall was 52.07%, and F1-score was 0.64, and the precision for secondary caries identification was 100%, recall was 33.33%, and F1-score was 0.50. The mAP value of the comprehensive group algorithm was 85.48%, in which the precision for primary caries identification was 93.33%, recall was 69.42%, F1-score was 0.80, and the F1-score for secondary caries identification was 0.50; precision was 100%, recall was 52.38%, and F1-score was 0.69.

Conclusions: The caries detection capability based on the YOLOv3 algorithm highlights the potential utility of deep learning in caries detection and diagnosis. Comparing the 3 experiments, the detection of the model trained after using image augmented and enhancement techniques was significantly improved.

Keywords: Dental caries; YOLOv3 algorithm; image recognition; oral images.