Intra-oral scan segmentation using deep learning.

OBJECTIVE: Intra-oral scans and gypsum cast scans (OS) are widely used in orthodontics, prosthetics, implantology, and orthognathic surgery to plan patient-specific treatments, which require teeth segmentations with high accuracy and resolution. Manual teeth segmentation, the gold standard up until now, is time-consuming, tedious, and observer-dependent. This study aims to develop an automated teeth segmentation and labeling system using deep learning. MATERIAL AND METHODS: As a reference, 1750 OS were manually segmented and labeled. A deep-learning approach based on PointCNN and 3D U-net in combination with a rule-based heuristic algorithm and a combinatorial search algorithm was trained and validated on 1400 OS. Subsequently, the trained algorithm was applied to a test set consisting of 350 OS. The intersection over union (IoU), as a measure of accuracy, was calculated to quantify the degree of similarity between the annotated ground truth and the model predictions. RESULTS: The model achieved accurate teeth segmentations with a mean IoU score of 0.915. The FDI labels of the teeth were predicted with a mean accuracy of 0.894. The optical inspection showed excellent position agreements between the automatically and manually segmented teeth components. Minor flaws were mostly seen at the edges. CONCLUSION: The proposed method forms a promising foundation for time-effective and observer-independent teeth segmentation and labeling on intra-oral scans. CLINICAL SIGNIFICANCE: Deep learning may assist clinicians in virtual treatment planning in orthodontics, prosthetics, implantology, and orthognathic surgery. The impact of using such models in clinical practice should be explored.

Positional assessment of lower third molar and mandibular canal using explainable artificial intelligence.

OBJECTIVE: The aim of this study is to automatically assess the positional relationship between lower third molars (M3i) and the mandibular canal (MC) based on the panoramic radiograph(s) (PR(s)). MATERIAL AND METHODS: A total of 1444 M3s were manually annotated and labeled on 863 PRs as a reference. A deep-learning approach, based on MobileNet-V2 combination with a skeletonization algorithm and a signed distance method, was trained and validated on 733 PRs with 1227 M3s to classify the positional relationship between M3i and MC into three categories. Subsequently, the trained algorithm was applied to a test set consisting of 130 PRs (217 M3s). Accuracy, precision, sensitivity, specificity, negative predictive value, and F1-score were calculated. RESULTS: The proposed method achieved a weighted accuracy of 0.951, precision of 0.943, sensitivity of 0.941, specificity of 0.800, negative predictive value of 0.865 and an F1-score of 0.938. CONCLUSION: AI-enhanced assessment of PRs can objectively, accurately, and reproducibly determine the positional relationship between M3i and MC. CLINICAL SIGNIFICANCE: The use of such an explainable AI system can assist clinicians in the intuitive positional assessment of lower third molars and mandibular canals. Further research is required to automatically assess the risk of alveolar nerve injury on panoramic radiographs.

Deep learning for automated segmentation of the temporomandibular joint.

OBJECTIVE: Quantitative analysis of the volume and shape of the temporomandibular joint (TMJ) using cone-beam computed tomography (CBCT) requires accurate segmentation of the mandibular condyles and the glenoid fossae. This study aimed to develop and validate an automated segmentation tool based on a deep learning algorithm for accurate 3D reconstruction of the TMJ. MATERIALS AND METHODS: A three-step deep-learning approach based on a 3D U-net was developed to segment the condyles and glenoid fossae on CBCT datasets. Three 3D U-Nets were utilized for region of interest (ROI) determination, bone segmentation, and TMJ classification. The AI-based algorithm was trained and validated on 154 manually segmented CBCT images. Two independent observers and the AI algorithm segmented the TMJs of a test set of 8 CBCTs. The time required for the segmentation and accuracy metrics (intersection of union, DICE, etc.) was calculated to quantify the degree of similarity between the manual segmentations (ground truth) and the performances of the AI models. RESULTS: The AI segmentation achieved an intersection over union (IoU) of 0.955 and 0.935 for the condyles and glenoid fossa, respectively. The IoU of the two independent observers for manual condyle segmentation were 0.895 and 0.928, respectively (p<0.05). The mean time required for the AI segmentation was 3.6 s (SD 0.9), whereas the two observers needed 378.9 s (SD 204.9) and 571.6 s (SD 257.4), respectively (p<0.001). CONCLUSION: The AI-based automated segmentation tool segmented the mandibular condyles and glenoid fossae with high accuracy, speed, and consistency. Potential limited robustness and generalizability are risks that cannot be ruled out, as the algorithms were trained on scans from orthognathic surgery patients derived from just one type of CBCT scanner. CLINICAL SIGNIFICANCE: The incorporation of the AI-based segmentation tool into diagnostic software could facilitate 3D qualitative and quantitative analysis of TMJs in a clinical setting, particularly for the diagnosis of TMJ disorders and longitudinal follow-up.

Estimation of Alveolar Bone Loss in Periodontitis Using Machine Learning.

AIM: The objective of this research was to perform a pilot study to develop an automatic analysis of periapical radiographs from patients with and without periodontitis for the percentage alveolar bone loss (ABL) on the approximal surfaces of teeth using a supervised machine learning model, that is, convolutional neural networks (CNN). MATERIAL AND METHODS: A total of 1546 approximal sites from 54 participants on mandibular periapical radiographs were manually annotated (MA) for a training set (n = 1308 sites), a validation set (n = 98 sites), and a test set (n = 140 sites). The training and validation sets were used for the development of a CNN algorithm. The algorithm recognised the cemento-enamel junction, the most apical extent of the alveolar crest, the apex, and the surrounding alveolar bone. RESULTS: For the total of 140 images in the test set, the CNN scored a mean of 23.1 ± 11.8 %ABL, whilst the corresponding value for MA was 27.8 ± 13.8 %ABL. The intraclass correlation (ICC) was 0.601 (P < .001), indicating moderate reliability. Further subanalyses for various tooth types and various bone loss patterns showed that ICCs remained significant, although the algorithm performed with excellent reliability for %ABL on nonmolar teeth (incisors, canines, premolars; ICC = 0.763). CONCLUSIONS: A CNN trained algorithm on radiographic images showed a diagnostic performance with moderate to good reliability to detect and quantify %ABL in periapical radiographs.

Automated chart filing on panoramic radiographs using deep learning.

OBJECTIVE: The aim of this study is to automatically detect, segment and label teeth, crowns, fillings, root canal fillings, implants and root remnants on panoramic radiographs (PR(s)). MATERIAL AND METHODS: As a reference, 2000 PR(s) were manually annotated and labeled. A deep-learning approach based on mask R-CNN with Resnet-50 in combination with a rule-based heuristic algorithm and a combinatorial search algorithm was trained and validated on 1800 PR(s). Subsquently, the trained algorithm was applied onto a test set consisting of 200 PR(s). F1 scores, as a measure of accuracy, were calculated to quantify the degree of similarity between the annotated ground-truth and the model predictions. The F1-score considers the harmonic mean of precison (positive predictive value) and recall (specificity). RESULTS: The proposes method achieved F1 scores up to 0.993, 0.952 and 0.97 for detection, segmentation and labeling, respectivley. CONCLUSION: The proposed method forms a promising foundation for the further development of automatic chart filing on PR(s). CLINICAL SIGNIFICANCE: Deep learning may assist clinicians in summarizing the radiological findings on panoramic radiographs. The impact of using such models in clinical practice should be explored.

Immediate Nonsubmerged Custom Root Analog Implants: A Prospective Pilot Clinical Study.

PURPOSE: To evaluate the feasibility of a commercially available immediate root analog implant system Replicate (Natural Dental Implants). MATERIALS AND METHODS: Five consecutive patients in need of an implant in the premolar region were recruited for this pilot study. Following clinical examination, a cone beam computed tomography scan was made and the dental impressions digitized. On the basis of the superimposition of these datasets, a three-dimensional (3D) envelope was created for the selected tooth. Subsequently, the tooth root at the prospective implant site was segmented to create a 3D surface, and the obtained mesh data were used as the basis for designing a single-piece root analog implant within the 3D envelope. The designed root analog implant was fabricated using a five-axis computer-aided manufacturing machine. The root analog implants were inserted following flapless minimally invasive root extraction. Following 3 months of uninterrupted healing, definitive restorations were fabricated. Peri-implant clinical and radiographic measurements were obtained up to 12 months follow-up. RESULTS: All patients functioned well following 12 months of functional loading. Within one patient, one of the two root analog implants failed early. Peri-implant clinical and radiographic measurements demonstrated a stable situation after 12 months of functional loading. CONCLUSION: A novel digital approach for immediately restoring single teeth using root analog implants was introduced. In the future, long-term evaluation of the root analog implant technique is necessary to evaluate the success and survival of implants that were inserted using this technique.

Autotransplantation of Premolars With a 3-Dimensional Printed Titanium Replica of the Donor Tooth Functioning as a Surgical Guide: Proof of Concept.

PURPOSE: Autotransplantation of premolars is a good treatment option for young patients who have missing teeth. This study evaluated the use of a preoperatively 3-dimensional (3D)-printed replica of the donor tooth that functions as a surgical guide during autotransplantation. MATERIALS AND METHODS: Five consecutive procedures were prospectively observed. Transplantations of maxillary premolars with optimal root development were included in this study. A 3D-printed replica of the donor tooth was used to prepare a precisely fitting new alveolus at the recipient site before extracting the donor tooth. Procedure time, extra-alveolar time, and number of attempts needed to achieve a good fit of the donor tooth in the new alveolus were recorded. RESULTS: For each transplantation procedure, the surgical time was shorter than 30 minutes. An immediate good fit of the donor tooth in the new alveolus was achieved with an extra-alveolar time shorter than 1 minute for all transplantations. CONCLUSION: These results show that the extra-alveolar time is very short when the surgical guide is used; therefore, the chance of iatrogenic damage to the donor tooth is minimized. The use of a replica of the donor tooth makes the autotransplantation procedure easier for the surgeon and facilitates optimal placement of the transplant.

Accuracy of preemptively constructed, cone beam CT-, and CAD/CAM technology-based, individual Root Analogue Implant technique: an in vitro pilot investigation.

OBJECTIVES: The aim of this in vitro pilot investigation is to assess the accuracy of the preemptive individually fabricated root analogue implant (RAI) based on three-dimensional (3D) root surface models obtained from a cone beam computed tomography (CBCT) scan, computer-aided designing (CAD), and computer-aided manufacturing (CAM) technology and to measure the discrepancy in congruence with the alveolar socket subsequent to placement of the RAI. MATERIALS AND METHODS: Eleven single-rooted teeth from nine human cadaver mandibles were scanned with the 3D Accuitomo 170 CBCT system. The 3D surface reconstructions of the teeth acquired from the CBCT scans were used as input for fabrication of the RAIs in titanium using rapid manufacturing technology. The teeth were then carefully extracted. The teeth and RAIs were consequently optically scanned. The mandibles with the empty extraction sockets were scanned with CBCT using identical settings to the first scan. Finally, the preemptively made RAIs were implanted into their respective sockets, and the mandibles were again scanned with CBCT using the same scan settings as previous scans. All 3D surface reconstructions (CBCT 3D surface models and optical scan 3D models) were saved for further analysis. 3D models of original teeth and optical scans of the RAIs were superimposed onto each other; differences were quantified as root mean square (RMS) and Hausdorff surface distance. To obtain an estimate of the fit (congruence) of the RAIs in their respective sockets, the volumetric data sets of the sockets were compared with those of the root part of RAIs congruent with the sockets. RESULTS: Superimposed surfaces of the RAIs and the original tooth reveal discrepancy for RMS, volumetric geometry, and surface area varying from 0.08 mm to 0.35 mm, 0.1% to 7.9%, and 1.1% to 3.8%, respectively. Comparing volume differences of the alveolus with the socket corresponding part of the RAI resulted in every case the volume of the socket being greater than the root part of the RAI ranging from 0.6% to 5.9% volume difference. CONCLUSION: The preemptive CAD/CAM-based RAI technique might offer promising features for immediate implant placement. However, due to the lack of prospective clinical data, further research is needed to fine-tune and evaluate this technique.

Designing a novel dental root analogue implant using cone beam computed tomography and CAD/CAM technology.

OBJECTIVES: The study aim is to introduce a novel preemptively constructed dental root analogue implant (RAI) based on three-dimensional (3D) root surface models obtained from a cone beam computed tomography (CBCT) scan, computer aided designing and computer aided manufacturing technology. MATERIALS & METHODS: One partially edentulous mandibular human cadaver was scanned with the Accuitomo 170 CBCT system. The scan volumes and datasets were used to create 3D surface models of the tooth. A 3D surface mesh of the tooth was stored as a standard triangulation language (STL) file. A high-end selective laser melting technology was used to fabricate the RAI from the STL file. The RAI was produced in a biocompatible titanium alloy (Ti6Al4V). Optical scanning technology was used to measure the RAI, as well as the natural tooth that was extracted. To validate the accuracy of the CBCT 3D root surface and the manufactured Titanium RAI, both surfaces were superimposed on the optical scan of the tooth, which served as the gold "reference" standard. RESULTS: The differences between the RAI and the optical scan of the original tooth are most noticeable at the level of the apex and the cementenamel junction areas on the buccal and lingual side (divergence of more than 0.15 mm). Surface area measurements show an overall decrease in surface area of 6.33% for the RAI in comparison with the original tooth and an increase of 0.27% when comparing the 3D surface model with optical scan of the original tooth. CONCLUSION: With the use of currently available technology it is very well feasible to preemptively create a custom RAI in titanium. However, clinical evidence evaluating the success of this novel dental implant approach is needed.