Analysis of the feasibility of using deep learning for multiclass classification of dental anomalies on panoramic radiographs

Abstract

The aim of the feasibility study was to construct deep learning models for the classification of multiple dental anomalies in panoramic radiographs. Panoramic radiographs with single supernumerary teeth and/or odontomas were considered the "case" group; panoramic radiographs with no dental anomalies were considered the "control" group. The dataset comprised 150 panoramic radiographs: 50 each of no dental anomalies, single supernumerary teeth, and odontomas. To classify the panoramic radiographs into case and control categories, we employed AlexNet, which is a convolutional neural network model. AlexNet was able to classify whole panoramic radiographs into two or three classes, according to the presence or absence of supernumerary teeth or odontomas. The performance metrics of the three-class classification were 70%, 70.8%, 70%, and 69.7% for accuracy, precision, sensitivity, and F1 score, respectively, in the macro average. These results support the feasibility of using deep learning to detect multiple dental anomalies in panoramic radiographs.

The diagnostic performance of impacted third molars in the mandible: A review of deep learning on panoramic radiographs

“…Image segmentation is needed to detect and separate objects of interest in an image ( Ronneberger et al, 2015 ). Deep learning can detect and classify impacted teeth, such as canine ( Aljabri et al, 2022 ), mesiodens ( Jeon et al, 2022 ), supernumerary in the maxilla ( Kuwada et al, 2020 ), and dental anomalies ( Okazaki et al, 2022 ). Based on the results of the literature search, one article discussed deep learning using the U-Net model for ITM detection in the mandible, which showed a high performance with an average dice coefficient score of 93.6 % and a Jaccard index of 88.1 % ( Vinayahalingam et al, 2019 ).…”

Tooth and Supporting Tissue Anomalies Detection from Panoramic Radiography Using Integrating Convolution Neural Network with Batch Normalization

“…Previous research by Okazaki et al [24] used the pre-trained model AlexNet to categorize three dental anomalies, including cases without dental abnormalities, with single supernumerary teeth, and with odontoma, from panoramic radiographs. This study demonstrates a macro matrix performance of 70% for accuracy, 70.8% for precision, 70% for sensitivity, and 69.7% for the F1 score.…”

“…While pre-trained CNNs have demonstrated their effectiveness in dental anomaly classification, enhancing classification accuracy is imperative through the utilization of custom datasets and deep transfer learning techniques. The comprehensive categorization of panoramic radiographs into three groups-those without dental abnormalities, those with single supernumerary teeth, and those with odontoma-has been explored using CNN transfer learning, specifically with AlexNet [24]. The outcomes affirm the potential of employing deep learning for the identification of various dental abnormalities in panoramic radiographs.…”

A Basic Study for Predicting Dysphagia in Panoramic X-ray Images Using Artificial Intelligence (AI) Part 2: Analysis of the Position of the Hyoid Bone on Panoramic Radiographs

“…Regarding the number of groupings, a previous paper by Okazaki et al [12] may be recalled. These authors investigated whether abnormal images of different teeth could be correctly diagnosed; however, the targets of this study were single supernumerary teeth and odontomas.…”

“…Mao et al [9] detected furcation involvement on molar teeth, and Son et al [10] discussed automatic fracture detection in the maxillofacial area. Ha et al [11] evaluated the detection of supernumerary teeth, and Okazaki et al [12] investigated diagnostic accuracy in the detection of odontoma and impacted teeth. Other studies that have examined multiple diseases, including the detection of cysts and tumors by Yang et al [13].…”