Convolutional neural network‐based automated maxillary alveolar bone segmentation on cone‐beam computed tomography images

Abstract

Objectives To develop and assess the performance of a novel artificial intelligence (AI)‐driven convolutional neural network (CNN)‐based tool for automated three‐dimensional (3D) maxillary alveolar bone segmentation on cone‐beam computed tomography (CBCT) images. Materials and Methods A total of 141 CBCT scans were collected for performing training (n = 99), validation (n = 12), and testing (n = 30) of the CNN model for automated segmentation of the maxillary alveolar bone and its crestal contour. Following automated segmentation, the 3D models with under‐ or overestimated segmentations were refined by an expert for generating a refined‐AI (R‐AI) segmentation. The overall performance of CNN model was assessed. Also, 30% of the testing sample was randomly selected and manually segmented to compare the accuracy of AI and manual segmentation. Additionally, the time required to generate a 3D model was recorded in seconds (s). Results The accuracy metrics of automated segmentation showed an excellent range of values for all accuracy metrics. However, the manual method (95% HD: 0.20 ± 0.05 mm; IoU: 95% ± 3.0; DSC: 97% ± 2.0) showed slightly better performance than the AI segmentation (95% HD: 0.27 ± 0.03 mm; IoU: 92% ± 1.0; DSC: 96% ± 1.0). There was a statistically significant difference of the time‐consumed among the segmentation methods (p < .001). The AI‐driven segmentation (51.5 ± 10.9 s) was 116 times faster than the manual segmentation (5973.3 ± 623.6 s). The R‐AI method showed intermediate time‐consumed (1666.7 ± 588.5 s). Conclusion Although the manual segmentation showed slightly better performance, the novel CNN‐based tool also provided a highly accurate segmentation of the maxillary alveolar bone and its crestal contour consuming 116 times less than the manual approach.

Applications of artificial intelligence in the utilisation of imaging modalities in dentistry: A systematic review and meta-analysis of in-vitro studies

“…Within this subset, 67 literature reviews, 46 editorials, and 37 seminar articles were excluded, aligning with the study's focus on primary research articles. This comprehensive curation and scrutiny culminated in the inclusion of nine in-vitro papers [ [27] , [28] , [29] , [30] , [31] , [32] , [33] , [34] , [35] , [36] , [37] ] in the review, which met the specified eligibility criteria. Table 3 presents the overview of the included in-vitro papers [ [19] , [20] , [21] , [22] , [23] , [24] , [25] , [26] , [27] ].…”

“… Bui et al [ 28 ] 533 CBCT scans Detection of caries using dental radiographs Deep pre-trained model, SVM, KNN, DT, NB, RF Not specified Accuracy: 91.70 %, Sensitivity: 90.43 %, Specificity: 92.67 % AI demonstrated higher accuracy and specificity in caries detection than conventional methods. Fontenele et al [ 29 ] 141 CBCT scans Automated 3D maxillary alveolar bone segmentation on CBCT images CNN model, manual segmentation Not specified Manual: 95 % HD: 0.20 ± 0.05 mm, IoU: 95 % ± 3.0, DSC: 97 % ± 2.0, AI: 95 % HD: 0.27 ± 0.03 mm, IoU: 92 % ± 1.0, DSC: 96 % ± 1.0 AI provided fast and highly accurate segmentation, offering a substantial time-saving advantage. Gerhardt et al [ 30 ] 175 CBCT scans Automated detection and labelling of teeth and edentulous regions on CBCT images AI-Driven Tool (Virtual Patient Creator, Relu BV, Leuven, Belgium) Not specified Detection Accuracy: 99.7 %, Labelling Accuracy: 99 %, Segmentation Accuracy (IoU): 0.96/0.97 AI achieved near-perfect accuracy in detecting and labeling teeth and edentulous regions.…”

“…The AI achieved an accuracy of 91.70 %, sensitivity of 90.43 %, and specificity of 92.67 %, indicating its superior performance in caries detection compared to conventional methods. Fontenele et al [ 29 ] aimed to automate 3D maxillary alveolar bone segmentation on CBCT images using a CNN model. While their manual segmentation was highly accurate, AI provided a significant time-saving advantage, with only a slight reduction in accuracy.…”

Comparison of ex vivo periodontal defects with their respective 3D models generated by artificial intelligence on CBCT images

“…The study by Fontenele et al [21] assessed the accuracy of the AI-driven tool used in this study to segment the maxillary alveolar bone and its crestal contour. The authors compared the timing of automated segmentation (AI), manually re ned segmentation (R-AI), and fully manual segmentation that was performed on 30% of the testing sample.…”

Automatic segmentation of mandibular canal using transformer based neural networks

“…With the advent of advanced deep learning techniques, neural network-based segmentation of oral structures has shown significant progress ( Cui et al, 2022 ; Fontenele et al, 2023 ). However, the segmentation of the mandibular canal still falls short when compared to other anatomical structures.…”