Artificial Intelligence Based Patient-Specific Preoperative Planning Algorithm for Total Knee Arthroplasty

Abstract

Previous studies have shown that the manufacturer’s default preoperative plans for total knee arthroplasty with patient-specific guides require frequent, time-consuming changes by the surgeon. Currently, no research has been done on predicting preoperative plans for orthopedic surgery using machine learning. Therefore, this study aims to evaluate whether artificial intelligence (AI) driven planning tools can create surgeon and patient-specific preoperative plans that require fewer changes by the surgeon. A dataset of 5409 preoperative plans, including the manufacturer’s default and the plans corrected by 39 surgeons, was collected. Features were extracted from the preoperative plans that describe the implant sizes, position, and orientation in a surgeon- and patient-specific manner. Based on these features, non-linear regression models were employed to predict the surgeon’s corrected preoperative plan. The average number of corrections a surgeon has to make to the preoperative plan generated using AI was reduced by 39.7% compared to the manufacturer’s default plan. The femoral and tibial implant size in the manufacturer’s plan was correct in 68.4% and 73.1% of the cases, respectively, while the AI-based plan was correct in 82.2% and 85.0% of the cases, respectively, compared to the surgeon approved plan. Our method successfully demonstrated the use of machine learning to create preoperative plans in a surgeon- and patient-specific manner for total knee arthroplasty.

Reliability of artificial intelligence in predicting total knee arthroplasty component sizes: a systematic review

“…Patient sex appears to contribute an important role in predicting TKA size Kunze [ 9 ] 31.2 ± 5.6 (13.7–59.8) 5306:6472 Machine learning 5 (stochastic gradient boosting, random forest, support vector machine, extreme gradient boosting, and elastic-net penalized logistic regression) Training set 80% Test set 20% XR Actual sizing from OR records Two large academic and three community centres Machine learning algorithms demonstrated good accuracy for predicting within one size of the final tibial and femoral components used for TKA. Patient height and sex were the most important factors for prediction femoral and tibial component size, respectively Lambrechts[ 2 ] NR NR Machine learning 3 (multi-task LASSO (MTL), LASSO, and group LASSO) Training set 70% Test set 30% MRI Actual sizing from OR records 39 experienced surgeons from 38 hospitals A machine learning-based preoperative plan, which captures surgical preferences in a patient- and surgeon-specific manner, has the potential to reduce the time needed to modify the preoperative plan prior to approval Burge [ 14 ] NR 33:45 Machine learning 1 (ML-based 2D–3D pipeline) Training set 90% Validation set 10% Test set (additional subjects 78) 44% XR & MRI Actual sizing from OR records Osteoarthritis Initiative (OAI) and KISTI Higher prediction accuracies than generally reported for manual templating techniques LoE Level of evidence, BMI Body mass index, ML Machine learning, and ( Y ) Years …”

“…Lambrechts et al [ 2 ] tested the performance of ML models in the prediction of preoperative planning corrections for the femur–tibia joint interface. The ML models were validated on a dataset comprised of 5409 patients undergoing TKA.…”

“…In this era of rapid technological evolution, artificial intelligence (AI) and machine learning (ML) have surfaced as game changers in diverse medical domains, including orthopaedic surgery [ 1 , 2 ]. These advancements are notably apparent in total knee arthroplasty (TKA), an effective orthopaedic procedure that enhances patient quality of life and functional recovery [ 2 ]. TKA is one of the most frequently performed orthopaedic surgeries globally, with a projected annual surge of 85%, reaching 1.26 million by 2030 [ 3 ].…”

“…Several recent studies have reflected the growing interest in the application of AI-based tools in knee arthroplasty and the broader orthopaedic field [ 2 , 9 , 13 , 14 ]. Despite this, the literature reveals limitations and gaps.…”

Applications of artificial intelligence in orthopaedic surgery

“…One growing trend in orthopaedic arthroplasty has been the use of patient specific implants, allowing for potentially superior outcomes, though involves more work during pre-operative planning ( 24 ). Lambrechts et al developed a novel application for ML in patient-specific joint replacement by using ML to automate patient- and surgeon-specific preoperative planning ( 25 ). Combining LASSO and SVM approaches, the AI-based preoperative plans were significantly improved as compared to the manufacturer's plans by requiring fewer manual corrections by the surgeon, thus streamlining the surgeons' preoperative workflow and reducing time needed to make corrections.…”