Layered deep learning for automatic mandibular segmentation in cone-beam computed tomography.

OBJECTIVE: To develop and validate a layered deep learning algorithm which automatically creates three-dimensional (3D) surface models of the human mandible out of cone-beam computed tomography (CBCT) imaging. MATERIALS & METHODS: Two convolutional networks using a 3D U-Net architecture were combined and deployed in a cloud-based artificial intelligence (AI) model. The AI model was trained in two phases and iteratively improved to optimize the segmentation result using 160 anonymized full skull CBCT scans of orthognathic surgery patients (70 preoperative scans and 90 postoperative scans). Finally, the final AI model was tested by assessing timing, consistency, and accuracy on a separate testing dataset of 15 pre- and 15 postoperative full skull CBCT scans. The AI model was compared to user refined AI segmentations (RAI) and to semi-automatic segmentation (SA), which is the current clinical standard. The time needed for segmentation was measured in seconds. Intra- and inter-operator consistency were assessed to check if the segmentation protocols delivered reproducible results. The following consistency metrics were used: intersection over union (IoU), dice similarity coefficient (DSC), Hausdorff distance (HD), absolute volume difference and root mean square (RMS) distance. To evaluate the match of the AI and RAI results to those of the SA method, their accuracy was measured using IoU, DSC, HD, absolute volume difference and RMS distance. RESULTS: On average, SA took 1218.4s. RAI showed a significant drop (p<0.0001) in timing to 456.5s (2.7-fold decrease). The AI method only took 17s (71.3-fold decrease). The average intra-operator IoU for RAI was 99.5% compared to 96.9% for SA. For inter-operator consistency, RAI scored an IoU of 99.6% compared to 94.6% for SA. The AI method was always consistent by default. In both the intra- and inter-operator consistency assessments, RAI outperformed SA on all metrics indicative of better consistency. With SA as the ground truth, AI and RAI scored an IoU of 94.6% and 94.4%, respectively. All accuracy metrics were similar for AI and RAI, meaning that both methods produce 3D models that closely match those produced by SA. CONCLUSION: A layered 3D U-Net architecture deep learning algorithm, with and without additional user refinements, improves time-efficiency, reduces operator error, and provides excellent accuracy when benchmarked against the clinical standard. CLINICAL SIGNIFICANCE: Semi-automatic segmentation in CBCT imaging is time-consuming and allows user-induced errors. Layered convolutional neural networks using a 3D U-Net architecture allow direct segmentation of high-resolution CBCT images. This approach creates 3D mandibular models in a more time-efficient and consistent way. It is accurate when benchmarked to semi-automatic segmentation.

Post-traumatic trigeminal neuropathy: correlation between objective and subjective assessments and a prediction model for neurosensory recovery.

BACKGROUND: Post-traumatic trigeminal neuropathy (PTN) can have a substantial effect on patient well-being. However, the relation between the neuropathic symptoms and their effect on psychosocial functioning remains a matter of debate. The purpose of this study was to evaluate the association between objective and subjective assessments of neurosensory function in PTN and predict neurosensory outcome using baseline measurements. METHODS: This prospective observational cohort study included patients diagnosed with PTN at the Department of Oral and Maxillofacial Surgery, University Hospital Leuven, Belgium, between April 2018 and May 2020. Standardized objective and subjective neurosensory examinations were recorded simultaneously on multiple occasions during the follow-up period. Correlation analyses and principal component analysis were conducted, and a prediction model of neurosensory recovery was developed. RESULTS: Quality of life correlated significantly (P < 0.05) with percentage of affected dermatome (ρ = - 0.35), the presence of brush stroke allodynia (ρ = - 0.24), gain-of-function sensory phenotype (ρ = - 0.41), Medical Research Council Scale (ρ = 0.36), and Sunderland classification (ρ = - 0.21). Quality of life was not significantly correlated (P > 0.05) with directional discrimination, stimulus localization, two-point discrimination, or sensory loss-of-function. The prediction model showed a negative predictive value for neurosensory recovery after 6 months of 87%. CONCLUSIONS: We found a strong correlation of subjective well-being with the presence of brush stroke allodynia, thermal and/or mechanical hyperesthesia, and the size of the neuropathic area. These results suggest that positive symptoms dominate the effect on affect. In patients reporting poor subjective well-being in the absence of positive symptoms or a large neuropathic area, additional attention towards psychosocial triggers might enhance treatment outcome. The prediction model could contribute to establishing realistic expectations about the likelihood of neurosensory recovery but remains to be validated in future studies.