A pilot study of a deep learning approach to detect marginal bone loss around implants

Min Liu,Hu Chen,Yunsong Liu,Shimin Wang

doi:10.1186/s12903-021-02035-8

Abstract

BackgroundRecently, there has been considerable innovation in artificial intelligence (AI) for healthcare. Convolutional neural networks (CNNs) show excellent object detection and classification performance. This study assessed the accuracy of an artificial intelligence (AI) application for the detection of marginal bone loss on periapical radiographs.MethodsA Faster region-based convolutional neural network (R-CNN) was trained. Overall, 1670 periapical radiographic images were divided into training (n = 1370), validation (n = 150), and test (n = 150) datasets. The system was evaluated in terms of sensitivity, specificity, the mistake diagnostic rate, the omission diagnostic rate, and the positive predictive value. Kappa (κ) statistics were compared between the system and dental clinicians.ResultsEvaluation metrics of AI system is equal to resident dentist. The agreement between the AI system and expert is moderate to substantial (κ = 0.547 and 0.568 for bone loss sites and bone loss implants, respectively) for detecting marginal bone loss around dental implants.ConclusionsThis AI system based on Faster R-CNN analysis of periapical radiographs is a highly promising auxiliary diagnostic tool for peri-implant bone loss detection.

Highlights

Dental implants are important for restoring biological function in patients with missing teeth [1, 2] and have become increasingly popular since the 1980s [3]
The inclusion criteria were as follows: periapical radiographs of dental implants, appropriate radiation exposure, and radiographs of dental implants acquired in parallel
The average precision [35] (AP) for marginal bone loss gradually increased with an increasing number of iterations

Summary

Introduction

Dental implants are important for restoring biological function in patients with missing teeth [1, 2] and have become increasingly popular since the 1980s [3]. Monitoring and maintenance are critical for long-term stability after implantation [4]. Marginal bone resorption is an important parameter that should be monitored. Bone loss of < 1.5 mm at 1-year post-loading is generally considered acceptable, followed by the loss of 0.2 mm annually thereafter [5, 6]. In cases where bone loss exceeds this amount, careful investigation is needed, including in. Bone loss is usually evaluated on radiographs. For general practitioners, evaluating marginal bone loss around implants can be difficult. Detection of the peri-implant bone level relies on imaging findings. This study assessed the accuracy of an artificial intelligence (AI) application for the detection of marginal bone loss on periapical radiographs

Objectives

Methods

Results

Discussion

Conclusion