Detection of the separated root canal instrument on panoramic radiograph: a comparison of LSTM and CNN deep learning methods

Abstract

Objectives: A separated endodontic instrument is one of the challenging complications of root canal treatment. The purpose of this study was to compare two deep learning methods that are convolutional neural network (CNN) and long short-term memory (LSTM) to detect the separated endodontic instruments on dental radiographs. Methods: Panoramic radiographs from the hospital archive were retrospectively evaluated by two dentists. A total of 915 teeth, of which 417 are labeled as “separated instrument” and 498 are labeled as “healthy root canal treatment” were included. A total of six deep learning models, four of which are some varieties of CNN (Raw-CNN, Augmented-CNN, Gabor filtered-CNN, Gabor-filtered-augmented-CNN) and two of which are some varieties of LSTM model (Raw-LSTM, Augmented-LSTM) were trained based on several feature extraction methods with an applied or not applied an augmentation procedure. The diagnostic performances of the models were compared in terms of accuracy, sensitivity, specificity, positive and negative predictive value using ten-fold cross-validation. A McNemar’s tests was employed to figure out if there is a statistically significant difference between performances of the models. Receiver Operating Characteristic (ROC) curves were developed to assess the quality of the performance of the most promising model (Gabor filtered-CNN model) by exploring different cut-off levels in the last decision layer of the model. Results: The Gabor filtered-CNN model showed the highest accuracy (84.37 ± 2.79), sensitivity (81.26 ± 4.79), positive predictive value (84.16 ± 3.35) and negative predictive value (84.62 ± 4.56 with a confidence interval of 80.6 ± 0.0076. McNemar’s tests yielded that the performance of the Gabor filtered-CNN model significantly different from both LSTM models (p < 0.01). Conclusions: Both CNN and LSTM models were achieved a high predictive performance on to distinguish separated endodontic instruments in radiographs. The Gabor filtered-CNN model without data augmentation gave the best predictive performance.

Detection of the separated endodontic instrument on periapical radiographs using a deep learning‐based convolutional neural network algorithm

“…According to a recent study, Mask R‐CNN could detect pulp stones on bite‐wing radiographs detection with approximately 90% sensitivity [37]. Another recent study showed that CNN models are successful at detecting separated endodontic instruments in panoramic radiographs [13]. The relatively lower accuracy rates compared to our study may be due to the use of panoramic radiographs instead of periapical radiographs and the use of different deep‐learning methods.…”

“…[8] Periapical lesion detection, evaluation of tooth morphology and root fracture were evaluated using a CNNbased algorithm in dentistry [9][10][11][12]. A recent study also utilised CNN to detect separated endodontic instruments on panoramic radiographs [13].…”

Artificial intelligence in endodontics: relevant trends and practical perspectives

“…Convolutional neural network was characterized with high levels of accuracy and sensitivity (84.37 ± 2.79 and 81.26 ± 4.79 correspondingly) for the detection of broken endodontic fi le within the root canal based on the panoramic images analysis [21]. Analogically AI helped to predict presence of C-shaped distal canal within mandibular molar based on the analysis of panoramic images with the accuracy of 95.1% and sensitivity of 92.7% [22].…”

Current Progress and Challenges of Using Artificial Intelligence in Clinical Dentistry—A Narrative Review

“…Analysis of panoramic pictures using a convolutional neural network has been shown to be highly accurate and sensitive (84.37 ± 2.79 and 81.26 ± 4.79, respectively) for detecting a damaged endodontic file within the root canal [40].…”