Evaluating the Effectiveness of Transtibial Prosthetic Socket Shape Design using Artificial Intelligence: A Clinical Comparison with Traditional Plaster Cast Socket Designs.

OBJECTIVE: This study investigates the feasibility of creating an AI algorithm to enhance prosthetic socket shapes for transtibial prostheses, aiming for a less operator-dependent, standardized approach. DESIGN: The study comprised two phases: first, developing an AI algorithm in a cross-sectional study to predict prosthetic socket shapes. Second, testing the AI-predicted Digitally Measured and Standardized Designed (DMSD-)prosthetic socket against a Manually Measured and Designed (MMD-)prosthetic socket in a two-week within-subject cross-sectional study. SETTING: The study was done at the rehabilitation department of the Radboud University Medical Center in Nijmegen, the Netherlands. PARTICIPANTS: The AI algorithm was developed using retrospective data from 116 patients from a Dutch orthopedic company: OIM Orthopedie, and tested on ten randomly selected participants from Papenburg Orthopedie. INTERVENTIONS: Utilization of an AI algorithm to enhance the shape of a transtibial prosthetic socket. MAIN OUTCOME MEASURES: The algorithm was optimized to minimize the error in the test set. Participants' Socket Comfort Score (SCS) and fitting ratings from an independent physiotherapist and prosthetist were collected. RESULTS: Predicted prosthetic shapes deviated by 2.51 mm from the actual designs. 8/10 DMSD and all 10 MMD-prosthetic sockets were satisfactory for home testing. Participants rated DMSD prosthetic sockets at 7.1 ± 2.2 (n=8) and MMD prosthetic sockets at 6.6 ± 1.2 (n=10) on average. CONCLUSION: The study demonstrates promising results for using an AI algorithm in prosthetic socket design, but long-term effectiveness and refinement for improved comfort and fit in more deviant cases are necessary.

Automated condylar seating assessment using a deep learning-based three-step approach.

OBJECTIVES: In orthognatic surgery, one of the primary determinants for reliable three-dimensional virtual surgery planning (3D VSP) and an accurate transfer of 3D VSP to the patient in the operation room is the condylar seating. Incorrectly seated condyles would primarily affect the accuracy of maxillary-first bimaxillary osteotomies as the maxillary repositioning is dependent on the positioning of the mandible in the cone-beam computed tomography (CBCT) scan. This study aimed to develop and validate a novel tool by utilizing a deep learning algorithm that automatically evaluates the condylar seating based on CBCT images as a proof of concept. MATERIALS AND METHODS: As a reference, 60 CBCT scans (120 condyles) were labeled. The automatic assessment of condylar seating included three main parts: segmentation module, ray-casting, and feed-forward neural network (FFNN). The AI-based algorithm was trained and tested using fivefold cross validation. The method's performance was evaluated by comparing the labeled ground truth with the model predictions on the validation dataset. RESULTS: The model achieved an accuracy of 0.80, positive predictive value of 0.61, negative predictive value of 0.9 and F1-score of 0.71. The sensitivity and specificity of the model was 0.86 and 0.78, respectively. The mean AUC over all folds was 0.87. CONCLUSION: The innovative integration of multi-step segmentation, ray-casting and a FFNN demonstrated to be a viable approach for automating condylar seating assessment and have obtained encouraging results. CLINICAL RELEVANCE: Automated condylar seating assessment using deep learning may improve orthognathic surgery, preventing errors and enhancing patient outcomes in maxillary-first bimaxillary osteotomies.

Fully automated landmarking and facial segmentation on 3D photographs.

Three-dimensional facial stereophotogrammetry provides a detailed representation of craniofacial soft tissue without the use of ionizing radiation. While manual annotation of landmarks serves as the current gold standard for cephalometric analysis, it is a time-consuming process and is prone to human error. The aim in this study was to develop and evaluate an automated cephalometric annotation method using a deep learning-based approach. Ten landmarks were manually annotated on 2897 3D facial photographs. The automated landmarking workflow involved two successive DiffusionNet models. The dataset was randomly divided into a training and test dataset. The precision of the workflow was evaluated by calculating the Euclidean distances between the automated and manual landmarks and compared to the intra-observer and inter-observer variability of manual annotation and a semi-automated landmarking method. The workflow was successful in 98.6% of all test cases. The deep learning-based landmarking method achieved precise and consistent landmark annotation. The mean precision of 1.69 ± 1.15 mm was comparable to the inter-observer variability (1.31 ± 0.91 mm) of manual annotation. Automated landmark annotation on 3D photographs was achieved with the DiffusionNet-based approach. The proposed method allows quantitative analysis of large datasets and may be used in diagnosis, follow-up, and virtual surgical planning.

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.

Quantification of muscle mass in the legs of patients with peripheral arterial occlusive disease: associations between volumetric and cross-sectional single-slice measurements for identification of atrophy and focal sarcopenia.

BACKGROUND: Sarcopenia, commonly determined by measuring skeletal muscle mass index (SMI) at the third lumbar level, has been identified as a predictor of clinical outcome in a variety of diseases. For patients with peripheral arterial occlusive disease (PAOD), we hypothesized that lower extremity SMI (LESMI) might be a more precise predictor of outcome and the extent of chronic ischemia than the systemic muscle mass at the L3 level. We investigated the association between complete muscle volume and muscle area derived with single-slice 2-dimensional measurements in the legs to identify at which level cross-sectional single-slice measurements are most representative of the muscle volume and investigated whether LESMI is associated with systemic sarcopenia and PAOD severity. METHODS: Muscle volumes and areas were semiautomatically segmented from computed tomography (CT) scans of the affected and contralateral legs of 50 PAOD patients with Fontaine stage IIb and 50 PAOD patients with Fontaine stage IV. The muscle mass was determined for the complete volumes of the upper and lower legs and for cross-sectional slices at 40%, 50%, and 60% of the length of the femur and tibia. Patients were determined as sarcopenic based on sex-specific cut-off values at the L3 spinal segment. Two observers segmented 20 randomly selected patients to determine the interobserver reliability with the intraclass correlation coefficient. RESULTS: The correlation between the LESMI of the complete muscle volume and the three cross-sectional slices in all 200 upper and 200 lower legs was moderately strong to strong. Interobserver reliability of cross-sectional slice segmentation was excellent. The LESMI, both measured volumetrically and cross-sectionally, were significantly lower in patients with sarcopenia compared to patients without sarcopenia. The LESMI was lower in patients with Fontaine stage IV compared to patients with Fontaine stage IIb for both volumetric and cross-sectional measurements. CONCLUSIONS: Segmentation of skeletal muscle mass from cross-sectional single-slice CT in the upper and lower leg can accurately and precisely substitute complete volume segmentations. These findings warrant the use of measurements based on cross-sectional single-slice CT for assessing the LESMI. Patients with systemic sarcopenia are also at increased risk for muscle mass loss in the lower extremities. In the current study, LESMI was lower in patients with Fontaine class IV PAOD compared to patients with Fontaine class IIb PAOD. Future studies should assess the predictive value of the LESMI on clinical outcomes in PAOD patients.