journal article·2023-03-29Open access

Preoperative Prediction of Optimal Femoral Implant Size by Regularized Regression on 3D Femoral Bone Shape

Adriaan Lambrechts, Christophe Van Dijck, Roel Wirix-Speetjens, Jos Vander Sloten, Frederik Maes, Sabine Van Huffel

Applied Sciences· Volume 13· Issue 7

Abstract

Preoperative determination of implant size for total knee arthroplasty surgery has numerous clinical and logistical benefits. Currently, surgeons use X-ray-based templating to estimate implant size, but this method has low accuracy. Our study aims to improve accuracy by developing a machine learning approach that predicts the required implant size based on a 3D femoral bone mesh, the key factor in determining the correct implant size. A linear regression framework imposing group sparsity on the 3D bone mesh vertex coordinates was proposed based on a dataset of 446 MRI scans. The group sparse regression method was further regularized based on the connectivity of the bone mesh to enforce neighbouring vertices to have similar importance to the model. Our hypergraph regularized group lasso had an accuracy of 70.1% in predicting femoral implant size while the initial implant size prediction provided by the instrumentation manufacturer to the surgeon has an accuracy of 23.1%. Furthermore, our method was capable of predicting the implant size up to one size smaller or larger with an accuracy of 99.1%, thereby surpassing other state-of-the-art methods. The hypergraph regularized group lasso was able to obtain a significantly higher accuracy compared to the implant size prediction provided by the instrumentation manufacturer.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Orthopaedic implants and arthroplasty · Total Knee Arthroplasty Outcomes · Orthopedic Infections and Treatments

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

Materialise NV, 3001 Leuven, Belgium

Biomechanics Section, Mechanical Engineering, KU Leuven, 3001 Leuven, Belgium

Processing Speech and Images (PSI), Department of Electrical Engineering (ESAT), KU Leuven, 3001 Leuven, Belgium

Medical Imaging Research Center, UZ Leuven, 3000 Leuven, Belgium

STADIUS Center for Dynamical Systems, Signal Processing and Data Analytics, Department of Electrical Engineering (ESAT), KU Leuven, 3001 Leuven, Belgium

References (37)

Show source-provided references
1. Patient Satisfaction after Total Knee Arthroplasty: Who is Satisfied and Who is Not?10.1007/s11999-009-1119-9 ↗
2. Demographic data is more predictive of component size than digital radiographic templating in total knee arthroplasty10.1186/s43019-020-00075-y ↗
3. The reliability and accuracy of digital templating in total knee replacement10.1302/0301-620X.91B7.21476 ↗
4. Accuracy of digital templating in total knee arthroplasty
5. The accuracy of preoperative templating in total knee arthroplasty
6. Preoperative CT-Based Three-Dimensional Templating in Robot-Assisted Total Knee Arthroplasty More Accurately Predicts Implant Sizes than Two-Dimensional Templating10.1055/s-0038-1666829 ↗
7. 2D versus 3D templating in total knee arthroplasty10.1016/j.knee.2015.08.014 ↗
8. Patient-specific instruments for total knee arthroplasty can accurately predict the component size as used peroperative10.1007/s00167-016-4345-1 ↗
9. Overhang of the Femoral Component in Total Knee Arthroplasty: Risk Factors and Clinical Consequences10.2106/JBJS.H.00434 ↗
10. Tibesku, C.O. (2015). The Unhappy Total Knee Replacement: A Comprehensive Review and Management Guide, Springer.
11. Total Knee Replacement Sizing: Shoe Size Is a Better Predictor for Implant Size than Body Height
12. Patient-specific Instruments for Total Knee Arthroplasty
13. A comparison of alignment using patient specific guides, computer navigation and conventional instrumentation in total knee arthroplasty10.1016/j.knee.2013.11.004 ↗
14. Investigating Pose Representations and Motion Contexts Modeling for 3D Motion Prediction10.1109/TPAMI.2021.3139918 ↗
15. Liu, H., Liu, M., Li, D., Zheng, W., Yin, L., and Wang, R. (2022). Recent Advances in Pulse-Coupled Neural Networks with Applications in Image Processing. Electronics, 11.10.3390/electronics11203264 ↗
16. Validation and performance of a machine-learning derived prediction guide for total knee arthroplasty component sizing10.1007/s00402-021-04041-5 ↗
17. Can Demographic Variables Accurately Predict Component Sizing in Primary Total Knee Arthroplasty?10.1016/j.arth.2017.05.007 ↗
18. Prospective Validation of a Demographically Based Primary Total Knee Arthroplasty Size Calculator10.1016/j.arth.2019.02.048 ↗
19. Accurately Predicting Total Knee Component Size without Preoperative Radiographs
20. Using Patient Demographics and Statistical Modeling to Predict Knee Tibia Component Sizing in Total Knee Arthroplasty10.1016/j.arth.2018.01.031 ↗
21. Naylor, B.H., Butler, J.T., Kuczynski, B., Bohm, A.R., and Scuderi, G.R. (2022). Can Component Size in Total Knee Arthroplasty Be Predicted Preoperatively?—An Analysis of Patient Characteristics. J. Knee Surg.10.1055/s-0042-1748902 ↗
22. Predicting implant size in total knee arthroplasty using demographic variables10.1302/0301-620X.102B6.BJJ-2019-1620.R1 ↗
23. Artificial Intelligence Based Patient-Specific Preoperative Planning Algorithm for Total Knee Arthroplasty10.3389/frobt.2022.899349 ↗
24. Sparse Discriminant Analysis10.1198/TECH.2011.08118 ↗
25. Marching cubes: A high resolution 3D surface construction algorithm10.1145/37402.37422 ↗
26. Statistical shape model-based prediction of tibiofemoral cartilage10.1080/10255842.2018.1495711 ↗
27. Amberg, B., Romdhani, S., and Vetter, T. (2007, January 17–22). Optimal Step Nonrigid ICP Algorithms for Surface Registration. Proceedings of the 2007 IEEE Conference on Computer Vision and Pattern Recognition, Minneapolis, MI, USA.10.1109/CVPR.2007.383165 ↗
28. Roth, V., and Fischer, B. (2008, January 5–9). The Group-Lasso for generalized linear models: Uniqueness of solutions and efficient algorithms. Proceedings of the 25th International Conference on Machine Learning, ICML ’08, Helsinki, Finland.10.1145/1390156.1390263 ↗
29. Hastie, T.R., and Wainwright, M. (2015). Statistical Learning with Sparsity: The Lasso and Generalizations, Chapman & Hall/CRC.10.1201/b18401 ↗
30. Hypergraph p-Laplacian Regularization for Remotely Sensed Image Recognition10.1109/TGRS.2018.2867570 ↗
31. Regularization and Variable Selection via the Elastic Net10.1111/j.1467-9868.2005.00503.x ↗
32. A method for solving the convex programming problem with convergence rate O (1/k⌃ 2)
33. Sparse matrix test problems10.1145/62038.62043 ↗
34. Anterior femoral notching ≤ 3 mm is associated with increased risk for supracondylar periprosthetic femoral fracture after total knee arthroplasty: A systematic review and meta-analysis10.1007/s00590-021-02989-z ↗
35. Automated segmentation of knee bone and cartilage combining statistical shape knowledge and convolutional neural networks: Data from the Osteoarthritis Initiative10.1016/j.media.2018.11.009 ↗
36. High intra- and inter-observer reliability of planning implant size in MRI-based patient-specific instrumentation for total knee arthroplasty10.1007/s00167-020-05946-1 ↗
37. Algorithm based automatic templating is less accurate than manual digital templating in total knee arthroplasty10.1002/jor.24696 ↗