A deep learning approach to automatic teeth detection and numbering based on object detection in dental periapical films

Hu Chen,Ludan Zhang,Ji Wu,Kailai Zhang,Chin-Hui Lee,Peijun Lyu,Hong Li

doi:10.1038/s41598-019-40414-y

Abstract

We propose using faster regions with convolutional neural network features (faster R-CNN) in the TensorFlow tool package to detect and number teeth in dental periapical films. To improve detection precisions, we propose three post-processing techniques to supplement the baseline faster R-CNN according to certain prior domain knowledge. First, a filtering algorithm is constructed to delete overlapping boxes detected by faster R-CNN associated with the same tooth. Next, a neural network model is implemented to detect missing teeth. Finally, a rule-base module based on a teeth numbering system is proposed to match labels of detected teeth boxes to modify detected results that violate certain intuitive rules. The intersection-over-union (IOU) value between detected and ground truth boxes are calculated to obtain precisions and recalls on a test dataset. Results demonstrate that both precisions and recalls exceed 90% and the mean value of the IOU between detected boxes and ground truths also reaches 91%. Moreover, three dentists are also invited to manually annotate the test dataset (independently), which are then compared to labels obtained by our proposed algorithms. The results indicate that machines already perform close to the level of a junior dentist.

Highlights

Human teeth are generally hard substances and do not damage ; their shapes can remain unchanged after a person’s death without being eroded
In this study, we propose a deep learning approach for automatic teeth detection and numbering based on faster R-convolutional neural network (CNN) with improved efficiencies and reduced workloads
The train and validation datasets were used to train our previous fast Regions with Convolutional Neural Network features (R-CNN) network[26], and the performances regarding test images are shown in Table 3, in the column “Prior work”, demonstrating a lower accuracy

Summary

Introduction

Human teeth are generally hard substances and do not damage ; their shapes can remain unchanged after a person’s death without being eroded. The work burden of a dentist and the occurrences of misdiagnosis may be reduced if intelligent dental X-ray film interpretation tools are developed to improve the quality of dental care. From this perspective, automatic teeth identification using digitized films is an important task for smart healthcare. Deep learning has developed in recent years, and is capable of automatically extracting image features using the original pixel information as input These new algorithms significantly reduce the workload of human experts, and can extract certain features that are difficult for humans to recognize. Prior domain knowledge is utilized to improve algorithm performance of the baseline faster R-CNN model, which is only a generic tool for general image recognition tasks but does not consider known tooth configuration information

Methods

Results

Discussion

Conclusion