A deep learning framework for automated dental segmentation and diagnostic report generation from cone-beam computed tomography

Micro–Computed Tomography–Guided Artificial Intelligence for Pulp Cavity and Tooth Segmentation on Cone-beam Computed Tomography

Objective: This study aims to develop a publicly accessible dataset for mandibular canal segmentation in cone beam computed tomography (CBCT) scans and to propose a framework for automated mandibular canal segmentation. Methods: A total of 236 CBCT scans were collected from the Stomatology Hospital of the Shantou University Medical College, and the mandibular canals in these files were finely annotated. A custom designed 3D UNet, named MyResUNet, along with two commonly used UNet models, were used as candidate models. Soft dice similarity coefficient (DSC) loss was used as the loss function. A post-processing step involving connected components analysis and removing small objects was applied during inference. Model performance was assessed using voxel accuracy (ACC), sensitivity (SEN), specificity (SPE),DSC(1), Hausdorff distance (HD), the 95th percentile Hausdorff distance (HD95), average surface distance (ASD), and average symmetric surface distance (ASSD). Results: The MCSTU dataset, which contains a development dataset and an independent test dataset comprising 218 and 18 CBCT images with fine-grained annotations, respectively, has been made publicly available. The validation loss of MyResUNet was lower than that of two commonly used models. The inclusion of post-processing significantly enhanced the performance, especially by reducing the HD metric. On the hold-out test dataset, the MyResUNet model achieved ACC, SEN, SPE, DSC, HD, HD95, ASD, ASSD with 95% confidence interval of 1 (1-1), 0.86 (0.83-0.87), 1 (1-1), 0.85 (0.83-0.86), 10.1 (8.67-13.6), 1.8 (1.6-2.2), 0.69 (0.58-0.85), and 0.72 (0.6-0.83), respectively. On the test dataset, the MyResUNet model obtained ACC, SEN, SPE, DICE, HD, HD95, ASD, ASSD(2, 3)with 95% confidence interval of 1 (1-1), 0.93 (0.91-0.95), 1 (1-1), 0.80 (0.79-0.81), 21.3 (11.7-53.9), 2.59 (2.33-3), 1 (0.96-1.21), and 0.92 (0.861-1), respectively. Both the code and trained models are publicly available. Conclusion: The proposed segmentation framework achieved strong performance on both the hold-out and independent test datasets. In the future, after further validation of the model’s generalization ability, it may be applied in real clinical settings for oral surgery planning.

A three-dimensional (3D) model of the teeth provides important information for orthodontic diagnosis and treatment planning. Tooth segmentation is an essential step in generating the 3D digital model from computed tomography (CT) images. The aim of this study is to develop an accurate and efficient tooth segmentation method from CT images. The 3D dental CT volumetric images are segmented slice by slice in a two-dimensional (2D) transverse plane. The 2D segmentation is composed of a manual initialization step and an automatic slice by slice segmentation step. In the manual initialization step, the user manually picks a starting slice and selects a seed point for each tooth in this slice. In the automatic slice segmentation step, a developed hybrid level set model is applied to segment tooth contours from each slice. Tooth contour propagation strategy is employed to initialize the level set function automatically. Cone beam CT (CBCT) images of two subjects were used to tune the parameters. Images of 16 additional subjects were used to validate the performance of the method. Volume overlap metrics and surface distance metrics were adopted to assess the segmentation accuracy quantitatively. The volume overlap metrics were volume difference (VD, mm(3)) and Dice similarity coefficient (DSC, %). The surface distance metrics were average symmetric surface distance (ASSD, mm), RMS (root mean square) symmetric surface distance (RMSSSD, mm), and maximum symmetric surface distance (MSSD, mm). Computation time was recorded to assess the efficiency. The performance of the proposed method has been compared with two state-of-the-art methods. For the tested CBCT images, the VD, DSC, ASSD, RMSSSD, and MSSD for the incisor were 38.16 ± 12.94 mm(3), 88.82 ± 2.14%, 0.29 ± 0.03 mm, 0.32 ± 0.08 mm, and 1.25 ± 0.58 mm, respectively; the VD, DSC, ASSD, RMSSSD, and MSSD for the canine were 49.12 ± 9.33 mm(3), 91.57 ± 0.82%, 0.27 ± 0.02 mm, 0.28 ± 0.03 mm, and 1.06 ± 0.40 mm, respectively; the VD, DSC, ASSD, RMSSSD, and MSSD for the premolar were 37.95 ± 10.13 mm(3), 92.45 ± 2.29%, 0.29 ± 0.06 mm, 0.33 ± 0.10 mm, and 1.28 ± 0.72 mm, respectively; the VD, DSC, ASSD, RMSSSD, and MSSD for the molar were 52.38 ± 17.27 mm(3), 94.12 ± 1.38%, 0.30 ± 0.08 mm, 0.35 ± 0.17 mm, and 1.52 ± 0.75 mm, respectively. The computation time of the proposed method for segmenting CBCT images of one subject was 7.25 ± 0.73 min. Compared with two other methods, the proposed method achieves significant improvement in terms of accuracy. The presented tooth segmentation method can be used to segment tooth contours from CT images accurately and efficiently.

Towards clinically applicable automated mandibular canal segmentation on CBCT

This study developed and evaluated a two-stage deep learning-based system for automatic segmentation of mandibular cortical bone, mandibular cancellous bone, maxillary cortical bone and maxillary cancellous bone on cone beam computed tomography (CBCT) images. A dataset containing 155 CBCT scans acquired with different parameters was obtained. A two-stage deep learning-based system was developed for automatically segmenting jawbone structures. The Dice similarity coefficient (DSC) and average symmetric surface distance (ASSD) were used to assess the segmentation performance of the system by comparing the automatic segmentation results with the ground truth. The impact of dental and quality abnormalities on segmentation performance was analysed, and a comparison of automatic segmentation (AS) with manually refined segmentation (MRS) was reported. The system achieved promising segmentation performance, with average DSC values of 93.69%, 96.83%, 86.14% and 95.57% and average ASSD values of 0.13mm, 0.16mm, 0.29mm and 0.41mm for the mandibular cortical bone, mandibular cancellous bone, maxillary cortical bone and maxillary cancellous bone, respectively. Quality abnormalities had a negative impact on segmentation performance. The performance metrics (DSCs > 98.8% and ASSDs < 0.1mm) indicated high overlap between the AS and MRS. The proposed system offers an accurate and time-efficient method for segmenting jawbone structures on CBCT images. Automatically segmenting jawbone structures is essential in most digital dental workflows. The proposed system has considerable potential for application in digital clinical workflows to assist dentists in making more accurate diagnoses and developing patient-specific treatment plans.

Fully automated method for three-dimensional segmentation and fine classification of mixed dentition in cone-beam computed tomography using deep learning

To develop and validate a modified deep learning (DL) model based on nnU-Net for classifying and segmenting five-class jaw lesions using cone-beam CT (CBCT). A total of 368 CBCT scans (37 168 slices) were used to train a multi-class segmentation model. The data underwent manual annotation by two oral and maxillofacial surgeons (OMSs) to serve as ground truth. Sensitivity, specificity, precision, F1-score, and accuracy were used to evaluate the classification ability of the model and doctors, with or without artificial intelligence assistance. The dice similarity coefficient (DSC), average symmetric surface distance (ASSD), and segmentation time were used to evaluate the segmentation effect of the model. The model achieved the dual task of classifying and segmenting jaw lesions in CBCT. For classification, the sensitivity, specificity, precision, and accuracy of the model were 0.871, 0.974, 0.874, and 0.891, respectively, surpassing oral and maxillofacial radiologists (OMFRs) and OMSs, approaching the specialist. With the model's assistance, the classification performance of OMFRs and OMSs improved, particularly for odontogenic keratocyst (OKC) and ameloblastoma (AM), with F1-score improvements ranging from 6.2% to 12.7%. For segmentation, the DSC was 87.2% and the ASSD was 1.359 mm. The model's average segmentation time was 40 ± 9.9 s, contrasting with 25 ± 7.2 min for OMSs. The proposed DL model accurately and efficiently classified and segmented five classes of jaw lesions using CBCT. In addition, it could assist doctors in improving classification accuracy and segmentation efficiency, particularly in distinguishing confusing lesions (eg, AM and OKC).

Cerebral palsy is a neurological condition that is known to affect muscle growth. Detailed investigations of muscle growth require segmentation of muscles from MRI scans, which is typically done manually. In this study, we evaluated the performance of 2D, 3D, and hybrid deep learning models for automatic segmentation of 11 lower leg muscles and two bones from MRI scans of children with and without cerebral palsy. All six models were trained and evaluated on manually segmented T1 -weighted MRI scans of the lower legs of 20 children, six of whom had cerebral palsy. The segmentation results were assessed using the median Dice similarity coefficient (DSC), average symmetric surface distance (ASSD), and volume error (VError) of all 13 labels of every scan. The best performance was achieved by H-DenseUNet, a hybrid model (DSC 0.90, ASSD 0.5 mm, and VError 2.6 cm3 ). The performance was equivalent to the inter-rater performance of manual segmentation (DSC 0.89, ASSD 0.6 mm, and VError 3.3 cm3 ). Models trained with the Dice loss function outperformed models trained with the cross-entropy loss function. Near-optimal performance could be attained using only 11 scans for training. Segmentation performance was similar for scans of typically developing children (DSC 0.90, ASSD 0.5 mm, and VError 2.8 cm3 ) and children with cerebral palsy (DSC 0.85, ASSD 0.6 mm, and VError 2.4 cm3 ). These findings demonstrate the feasibility of fully automatic segmentation of individual muscles and bones from MRI scans of children with and without cerebral palsy.

This retrospective study aimed to investigate the prevalence of pre-eruptive intracoronal resorption (PIR) in unerupted permanent teeth in a Turkish population using cone beam computed tomography (CBCT). A total of 1,317 CBCT images were screened. In all the images, the following were recorded: the number of unerupted teeth, number of teeth with intracoronal resorption, affected tooth type and number, and location of radiolucent defects. Demographic data were also obtained. The prevalence of intracoronal resorption in the study group was 15.1%, with a prevalence in teeth of 3.5%. The prevalence of intracoronal resorption using CBCT was much higher than that recorded previously using panoramic or bitewing radiographs intracoronal resorption was more common in males (57%) than females (43%). Twenty-three cases were located in the maxilla (48%), and 25 were located in the mandible (52%). The mandibular third molar was the most affected tooth type, followed by maxillary third molars and supernumerary teeth. CBCT can be useful for diagnosing PIR defects because it provides an accurate representation of internal dental anatomy. SCANNING 38:442-447, 2016. © 2016 Wiley Periodicals, Inc.

Preoperative radiological identification of mandibular canals is essential for maxillofacial surgery. This study demonstrates the reproducibility of a deep learning system (DLS) by evaluating its localisation performance on 165 heterogeneous cone beam computed tomography (CBCT) scans from 72 patients in comparison to an experienced radiologist’s annotations. We evaluated the performance of the DLS using the symmetric mean curve distance (SMCD), the average symmetric surface distance (ASSD), and the Dice similarity coefficient (DSC). The reproducibility of the SMCD was assessed using the within-subject coefficient of repeatability (RC). Three other experts rated the diagnostic validity twice using a 0–4 Likert scale. The reproducibility of the Likert scoring was assessed using the repeatability measure (RM). The RC of SMCD was 0.969 mm, the median (interquartile range) SMCD and ASSD were 0.643 (0.186) mm and 0.351 (0.135) mm, respectively, and the mean (standard deviation) DSC was 0.548 (0.138). The DLS performance was most affected by postoperative changes. The RM of the Likert scoring was 0.923 for the radiologist and 0.877 for the DLS. The mean (standard deviation) Likert score was 3.94 (0.27) for the radiologist and 3.84 (0.65) for the DLS. The DLS demonstrated proficient qualitative and quantitative reproducibility, temporal generalisability, and clinical validity.

Preoperative radiological identification of mandibular canals is essential for maxillofacial surgery. This study demonstrates the reproducibility of a deep learning system (DLS) by evaluating its localisation performance on 165 heterogeneous cone beam computed tomography (CBCT) scans from 72 patients in comparison to an experienced radiologist's annotations. We evaluated the performance of the DLS using the symmetric mean curve distance (SMCD), the average symmetric surface distance (ASSD), and the Dice similarity coefficient (DSC). The reproducibility of the SMCD was assessed using the within-subject coefficient of repeatability (RC). Three other experts rated the diagnostic validity twice using a 0-4 Likert scale. The reproducibility of the Likert scoring was assessed using the repeatability measure (RM). The RC of SMCD was 0.969 mm, the median (interquartile range) SMCD and ASSD were 0.643 (0.186) mm and 0.351 (0.135) mm, respectively, and the mean (standard deviation) DSC was 0.548 (0.138). The DLS performance was most affected by postoperative changes. The RM of the Likert scoring was 0.923 for the radiologist and 0.877 for the DLS. The mean (standard deviation) Likert score was 3.94 (0.27) for the radiologist and 3.84 (0.65) for the DLS. The DLS demonstrated proficient qualitative and quantitative reproducibility, temporal generalisability, and clinical validity.

Cone beam computed tomography (CBCT) image segmentation is crucial in prostate cancer radiotherapy, enabling precise delineation of the prostate gland for accurate treatment planning and delivery. However, the poor quality of CBCT images poses challenges in clinical practice, making annotation difficult due to factors such as image noise, low contrast, and organdeformation. The objective of this study is to create a segmentation model for the label-free target domain (CBCT), leveraging valuable insights derived from the label-rich source domain (CT). This goal is achieved by addressing the domain gap across diverse domains through the implementation of a cross-modality medical image segmentationframework. Our approach introduces a multi-scale domain adaptive segmentation method, performing domain adaptation simultaneously at both the image and feature levels. The primary innovation lies in a novel multi-scale anatomical regularization approach, which (i) aligns the target domain feature space with the source domain feature space at multiple spatial scales simultaneously, and (ii) exchanges information across different scales to fuse knowledge from multi-scaleperspectives. Quantitative and qualitative experiments were conducted on pelvic CBCT segmentation tasks. The training dataset comprises 40 unpaired CBCT-CT images with only CT images annotated. The validation and testing datasets consist of 5 and 10 CT images, respectively, all with annotations. The experimental results demonstrate the superior performance of our method compared to other state-of-the-art cross-modality medical image segmentation methods. The Dice similarity coefficients (DSC) for CBCT image segmentation results is %, and the average symmetric surface distance (ASSD) is . Statistical analysis confirms the statistical significance of the improvements achieved by ourmethod. Our method exhibits superiority in pelvic CBCT image segmentation compared to its counterparts.

LG-nnU-net for multilabel anal sphincter segmentation on MRI: quantitative evaluation in patients with anal fistula.

The aim of the present study was to morphometrically analyze the mandibular canal through the mandibular ramus by cone beam computed tomography (CBCT) and to relate the findings to performing sagittal split ramus osteotomy. CBCT of 200 patients were analyzed. Five parameters were measured at the axial scan, from the mandibular foramen to 21mm below it (3-mm intervals). The canal was classified according to the position within the bone marrow space. Variations were evaluated according to age, sex, side, and number of mandibular teeth. The following measurements increased gradually towards the most inferior level of measurement: the total thickness of the mandibular ramus through the center of the mandibular canal, the width of the bone marrow space (both buccal and lingual), and the narrowest width from the mandibular canal inner cortical to the mandibular ramus external cortical. The inner diameter of the mandibular canal slightly decreased to the same direction. Concerning the mandibular canal position within the bone marrow space, the percentage of the separate type increased towards the most inferior level of measurement, and the contact and fusion types decreased. Age, number of teeth, and sex had no significant influence on the total thickness of the mandibular ramus and on the narrowest width from the mandibular canal inner cortical to the mandibular ramus external cortical.

ObjectivesThe current study aimed to automatically detect tooth presence, tooth numbering, and types of periodontal bone defects from cone-beam CT (CBCT) images using a segmentation method with an advanced artificial intelligence (AI) algorithm.MethodsThis study utilized a dataset of CBCT volumes collected from 502 individual subjects. Initially, 250 CBCT volumes were used for automatic tooth segmentation and numbering. Subsequently, CBCT volumes from 251 patients diagnosed with periodontal disease were employed to train an AI system to identify various periodontal bone defects using a segmentation method in web-based labelling software. In the third stage, CBCT images from 251 periodontally healthy subjects were combined with images from 251 periodontally diseased subjects to develop an AI model capable of automatically classifying patients as either periodontally healthy or periodontally diseased. Statistical evaluation included receiver operating characteristic curve analysis and confusion matrix model.ResultsThe area under the receiver operating characteristic curve (AUC) values for the models developed to segment teeth, total alveolar bone loss, supra-bony defects, infra-bony defects, perio-endo lesions, buccal defects, and furcation defects were 0.9594, 0.8499, 0.5052, 0.5613 (with cropping, AUC: 0.7488), 0.8893, 0.6780 (with cropping, AUC: 0.7592), and 0.6332 (with cropping, AUC: 0.8087), respectively. Additionally, the classification CNN model achieved an accuracy of 80% for healthy individuals and 76% for unhealthy individuals.ConclusionsThis study employed AI models on CBCT images to automatically detect tooth presence, numbering, and various periodontal bone defects, achieving high accuracy and demonstrating potential for enhancing dental diagnostics and patient care.