Joint Track Machine Learning: An Autonomous Method of Measuring Total Knee Arthroplasty Kinematics From Single-Plane X-Ray Images.

BACKGROUND: Dynamic radiographic measurements of 3-dimensional (3-D) total knee arthroplasty (TKA) kinematics have provided important information for implant design and surgical technique for over 30 years. However, current methods of measuring TKA kinematics are too cumbersome, inaccurate, or time-consuming for practical clinical application. Even state-of-the-art techniques require human-supervision to obtain clinically reliable kinematics. Eliminating human supervision could potentially make this technology practical for clinical use. METHODS: We demonstrate a fully autonomous pipeline for quantifying 3D-TKA kinematics from single-plane radiographic imaging. First, a convolutional neural network (CNN) segmented the femoral and tibial implants from the image. Second, those segmented images were compared to precomputed shape libraries for initial pose estimates. Lastly, a numerical optimization routine aligned 3D implant contours and fluoroscopic images to obtain the final implant poses. RESULTS: The autonomous technique reliably produces kinematic measurements comparable to human-supervised measures, with root-mean-squared differences of less than 0.7 mm and 4° for our test data, and 0.8 mm and 1.7° for external validation studies. CONCLUSION: A fully autonomous method to measure 3D-TKA kinematics from single-plane radiographic images produces results equivalent to a human-supervised method, and may soon make it practical to perform these measurements in a clinical setting.

Determining the relationship between tibiofemoral geometry and passive motion with partial least squares regression.

Tibiofemoral geometry influences knee passive motion and understanding their relationship can provide insight into knee function and mechanisms of injury. However, the complexity of the geometric constraints has made characterizing the relationship challenging. The aim of this study was to determine the tibiofemoral bone geometries that explain the variation in passive motion using a partial least squares regression (PLSR) model. The PLSR model was developed for 29 healthy cadaver specimens (10 female, 19 male) with femur and tibia geometries retrieved from MRI images and six degree-of-freedom tibiofemoral kinematics determined during a flexion cycle with minimal medial pressure. The first 13 partial least squares (PLS) components explained 90% of the variation in the kinematics and accounted for 89% of the variation in geometry. The first three PLS components which shared geometric changes to particular surface congruencies of the tibial and femoral condyles explained the most amount of variation in the kinematics, primarily in anterior-posterior translation. Meanwhile, variations in femoral condyle width and the intercondylar space, tibia plateau size and conformity, and tibia eminences heights in PLS 2 and 4 explained the greatest amount of variation in internal-external rotation. PLS 4 exhibiting variation in overall size of the knee accounted for greatest amount of variation in geometry (50%) and had the greatest influence on the abduction-adduction motion and some on internal-external rotation but, overall, explained only a small proportion of the kinematics (10%). Elucidating the complex relationship between tibiofemoral bone geometry and passive kinematics may help personalize treatments for improved functional outcomes in patients.

Automatic tracking of healthy joint kinematics from stereo-radiography sequences.

Kinematic tracking of healthy joints in radiography sequences is frequently performed by maximizing similarities between computed perspective projections of 3D computer models and corresponding objects' appearances in radiographic images. Significant human effort associated with manual tracking presents a major bottleneck in biomechanics research methods and limits the scale of target applications. The current work introduces a method for fully-automatic tracking of tibiofemoral and patellofemoral kinematics in stereo-radiography sequences for subjects performing dynamic activities. The proposed method involves the application of convolutional neural networks for annotating radiographs and a multi-stage optimization pipeline for estimating bone pose based on information provided by neural net predictions. Predicted kinematics are evaluated by comparing against manually-tracked trends across 20 distinct trials. Median absolute differences below 1.5 millimeters or degrees for 6 tibiofemoral and 3 patellofemoral degrees of freedom demonstrate the utility of our approach, which improves upon previous semi-automatic methods by enabling end-to-end automation. Implementation of a fully-automatic pipeline for kinematic tracking will benefit evaluation of human movement by enabling large-scale studies of healthy knee kinematics.

Machine learning for rapid estimation of lower extremity muscle and joint loading during activities of daily living.

Joint contact and muscle forces estimated with musculoskeletal modeling techniques offer useful metrics describing movement quality that benefit multiple research and clinical applications. The expensive processing of laboratory data associated with generating these outputs presents challenges to researchers and clinicians, including significant time and expertise requirements that limit the number of subjects typically evaluated. The objective of the current study was to develop and compare machine learning techniques for rapid, data-driven estimation of musculoskeletal metrics from derived gait lab data. OpenSim estimates of patient joint and muscle forces during activities of daily living were simulated using laboratory data from 70 total knee replacement patients and used to develop 4 different machine learning algorithms. Trained machine learning models predicted both trend and magnitude of estimated joint contact (mean correlation coefficients ranging from 0.93 to 0.94 during gait) and muscle forces (mean correlation coefficients ranging from 0.83 to 0.91 during gait) based on anthropometrics, ground reaction forces, and joint angle data. Patient mechanics were accurately predicted by recurrent neural networks, even after removing dependence on key subsets of predictor features. The ability to quickly estimate patient mechanics from derived measurements of movement has the potential to broaden the impact of musculoskeletal modeling by enabling faster assessment in both clinical and research settings.

Assessment of scapular morphology and bone quality with statistical models.

The design of total shoulder arthroplasty implants are guided by anatomy. The objective of this study was to develop statistical models to quantify shape and material property variation in the scapula. Material-mapped models were reconstructed from CT scans for a training set of subjects. Statistical shape (SSM) and intensity (SIM) models were created; SSM modes described scaling, changes in the medial border and acromial process, and elongation of the scapular blade. SIM modes captured bone quality changes in the anterior and inferior glenoid. Bone quality was independent of scapular morphology. Variation described by the statistical representations can inform implant design and sizing.

Investigating gender and ethnicity differences in proximal humeral morphology using a statistical shape model.

Morphological variability in the shoulder influences the joint biomechanics and is an important consideration in arthroplasty and implant design. The objectives of this study were to quantify cortical and cancellous proximal humeral morphology and to assess whether shape variation was influenced by gender and ethnicity, with the overarching goal of informing implant design and treatment. A statistical shape model of the proximal humeral cortical and cancellous regions was developed for a training set of 84 subjects of both genders and different ethnicities. Cortical and cancellous bone geometries were reconstructed from CT scans, meshed with triangular elements, and registered to a template. Principal component analysis was applied to quantify modes of variation. Anatomical measurements were computed on the registered geometries to assess correlation with modes of variation. Parallel analysis identified six significant modes of variation, which accounted for 93% of variation in the training set and described scaling (Mode 1), inclination of the head (Modes 2 and 5), and shape of the greater tuberosity and neck region (Modes 3, 4, and 6). Size differences as described by Mode 1 were statistically significant for gender and ethnicity, where female and Asian subjects were smaller than male and Caucasian subjects, respectively; however, differences in other modes were not significant. Cortical thickness of the shaft after normalization by outer diameter was significantly larger for Asian subjects compared to Caucasian subjects. The statistical shape model quantified cortical and cancellous humeral morphology considering gender and ethnicity, providing descriptive data to support surgical planning, and implant design. © 2018 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 36:3043-3052, 2018.