Experimental validation of computer-vision methods for the successful detection of endodontic treatment obturation and progression from noisy radiographs

Abstract

Purpose (1) To evaluate the effects of denoising and data balancing on deep learning to detect endodontic treatment outcomes from radiographs. (2) To develop and train a deep-learning model and classifier to predict obturation quality from radiomics. Methods The study conformed to the STARD 2015 and MI-CLAIMS 2021 guidelines. 250 deidentified dental radiographs were collected and augmented to produce 2226 images. The dataset was classified according to endodontic treatment outcomes following a set of customized criteria. The dataset was denoised and balanced, and processed with YOLOv5s, YOLOv5x, and YOLOv7 models of real-time deep-learning computer vision. Diagnostic test parameters such as sensitivity (Sn), specificity (Sp), accuracy (Ac), precision, recall, mean average precision (mAP), and confidence were evaluated. Results Overall accuracy for all the deep-learning models was above 85%. Imbalanced datasets with noise removal led to YOLOv5x’s prediction accuracy to drop to 72%, while balancing and noise removal led to all three models performing at over 95% accuracy. mAP saw an improvement from 52 to 92% following balancing and denoising. Conclusion The current study of computer vision applied to radiomic datasets successfully classified endodontic treatment obturation and mishaps according to a custom progressive classification system and serves as a foundation to larger research on the subject matter.

Clinical Annotation and Segmentation Tool (CAST) Implementation for Dental Diagnostics

“…The study also examined the performance of YOLO as an object detection model in radiographic datasets. Previous investigations revealed that the greatest limitations in this context were noise and class imbalances across the imaging groups, which was also experienced in the current study with the intraoral image dataset [18]. Although having radiographs aligned with clinical photographs was preferred, the authors faced challenges in collecting an adequate number of oral images for the design implementation.…”

“…Similarly, restorative status was subclassified into ‘amalgam restoration’ and ‘tooth-coloured restorations.’ The second dataset featured 424 digital peri-apical radiographs of teeth that had undergone complete or partial endodontic treatment. These images were categorized as ‘no endodontic treatment,’ ‘incomplete,’ ‘complete,’ or ‘endodontic mishap.’ It is worth noting that the datasets underwent classification using bounding boxes, balancing, and denoising procedures based on well-documented methodologies from a previous report [ 18 ]. Figure 1 demonstrates the different methods applied for labelling.…”

“…These images were categorized as 'no endodontic treatment,' 'incomplete,' 'complete,' or 'endodontic mishap.' It is worth noting that the datasets underwent classification using bounding boxes, balancing, and denoising procedures based on well-documented methodologies from a previous report [18]. Figure 1 demonstrates the different methods applied for labelling.…”

Influence of Intraoral Scanners, Operators, and Data Processing on Dimensional Accuracy of Dental Casts for Unsupervised Clinical Machine Learning: An In Vitro Comparative Study

“…Even with the most reliable operator-driven imaging modalities and 3D scanning tools, inter-rater reliability and agreements are adversely affected by operator-dependent biases [3][4][5]. As dental research attempts to integrate various forms of 2D datasets, including caries diagnoses from oral photographs and microscopy [6] and endodontic treatment status monitoring from radiographs [7], for AI implementation, it is imperative to determine whether 3D data derived from different clinics and disparate intraoral scanners would impact the dimensional accuracy of the datasets intended to develop AI models.…”

The role of radiomics in dentistry and oral radiology

“…technologies' potential to significantly enhance clini cal treatment [6][7][8][9]. This article highlighted the potential role of radiomics in dentistry and oral radiology.…”