The goal of this observational study is to develop and pre-validate a machine learning algorithm to predict early recovery of mobility in patients undergoing hip or knee joint replacement surgery. The primary research question is:
Can a machine learning model accurately classify patients with faster versus slower recovery of autonomous mobility in the first days after joint replacement surgery?
Patients who have undergone elective hip or knee arthroplasty and received post-operative physiotherapy will have their clinical and perioperative data collected retrospectively (2020-2023) and prospectively (March 2026-December 2027). The algorithm will be trained on retrospective data and tested prospectively to evaluate its predictive performance for early mobilization and length of hospital stay.
Adults aged 18 years or older
Patients underwent elective hip or knee arthroplasty.
Patients for whom postoperative physiotherapy was initiated.
Patients who underwent surgery for oncologic disease, femoral fracture, or revision joint arthroplasty.
Patients for whom postoperative physiotherapy was not provided due to postoperative complications
clinical data are unavailable.
For the retrospective phase, data will be extracted from existing clinical and research databases at IRCCS Istituto Ortopedico Rizzoli. Pre-existing data collected between 2020 and 2023 will be used to develop automated predictive models. For the prospective phase, patients admitted to IRCCS Istituto Ortopedico Rizzoli for elective hip or knee arthroplasty between March 2026 and December 2027 will be consecutively enrolled. Newly collected data will be used to externally validate the predictive model, which will be applied without any modification (locked model).
Cohort
Prospective
Predictive Model for Early Mobility Recovery and Length of Stay
Other interventionApplication of a machine learning-based predictive algorithm to retrospectively and prospectively analyze clinical and perioperative data in patients undergoing hip or knee arthroplasty, without influencing clinical decision-making.
Area under the receiver operating characteristic curve (AUROC) for discrimination ability of the machine learning predictive model
The discrimination ability of the machine learning predictive model will be assessed using the area under the receiver operating characteristic curve (AUROC). AUROC summarizes the trade-off between sensitivity and specificity across all possible classification thresholds. AUROC values range from 0.5 (no discrimination) to 1.0 (perfect discrimination). Higher values indicate better model performance. Values above 0.8 will be considered indicative of good discriminatory performance.
Calibration of the machine learning predictive model assessed by calibration plots
Agreement between predicted probabilities and observed outcomes will be evaluated using calibration plots. Calibration will be visually assessed by plotting predicted versus observed event probabilities.
Predictive performance of the machine learning model assessed by precision and F1-score
Predictive performance will be evaluated using precision and F1-score derived from the confusion matrix by comparing predicted class labels with observed outcomes. Precision reflects the proportion of correctly predicted positive cases among all predicted positives. The F1-score represents the harmonic mean of precision and recall. Higher values indicate better predictive performance.
Click on the lead sponsor to view all of their trials.
Istituto Ortopedico Rizzoli
Lead sponsor
Azienda U.S.L. - IRCCS di Reggio Emilia
Collaborator
