Machine learning predicts improvement of functional outcomes in spinal cord injury patients after inpatient rehabilitation

Aug 5, 2026

Retrospective analysis of 589 spinal cord injury (SCI) patients admitted to a single acute rehabilitation facility trains advanced machine learning models to predict rehabilitation outcome at discharge. The primary endpoint is the Functional Independence Measure (FIM) score, reflecting independence after comprehensive inpatient rehabilitation. Tree-based approaches, especially Random Forest and XGBoost, outperform traditional statistical models and generalized linear models, achieving substantially higher predictive performance and identifying key predictors such as initial FIM and injury-related demographic factors.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages19

Reading Time48 min

AccessFree to download

Introduction

Methods

Results

Discussion

TYPE Original Research PUBLISHED 25 August 2025 DOI 10.3389/fresc.2025.1594753 EDITED BY Long Wang, University of Science and Technology Beijing, China REVIEWED BY Taslim Uddin, Bangabandhu Sheikh Mujib Medical University (BSMMU), Bangladesh Dewa Putu Wisnu Wardhana, Udayana University Hospital, Indonesia *CORRESPONDENCE Daniel C. Lu [dclu@mednet.ucla.edu](dclu@mednet.ucla.edu) †These authors have contributed equally to this work RECEIVED 17 March 2025 ACCEPTED 24 July 2025 PUBLISHED 25 August 2025 CITATION Rasoolinejad M, Say I, Wu PB, Liu X, Zhou Y, Zhang N, Rosario ER and Lu DC (2025) Machine learning predicts improvement of functional outcomes in spinal cord injury patients after inpatient rehabilitation. Front. Rehabil. Sci. 6:1594753 . doi: 10.3389/fresc.2025.1594753 COPYRIGHT © 2025 Rasoolinejad, Say, Wu, Liu, Zhou, Zhang, Rosario and Lu. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Machine learning predicts improvement of functional outcomes in spinal cord injury patients after inpatient rehabilitation Mohammad Rasoolinejad1†, Irene Say1†, Peter B. Wu1, Xinran Liu2, Yan Zhou1,3, Nathan Zhang4, Emily R. Rosario5 and Daniel C. Lu1,6,7* 1Department of Neurosurgery, David Geffen School of Medicine, University of California, Los Angeles, CA, United States, 2Department of Molecular, Cell and Developmental Biology, University of California, Los Angeles, CA, United States, 3School of Medicine, University of Colorado Anschutz Medical Campus, Aurora, CO, United States, 4Department of Computer Science, University of California, Los Angeles, CA, United States, 5Research Institute, Casa Colina Hospital and Centers for Healthcare, Pomona, CA, United States, 6Neuromotor Recovery and Rehabilitation Center, David Geffen School of Medicine, University of California, Los Angeles, CA, United States, 7Brain Research Institute, University of California, Los Angeles, CA, United States Introduction: Spinal cord injury (SCI) presents a signiﬁcant burden to patients, families, and the healthcare system. The ability to accurately predict functional outcomes for SCI patients is essential for optimizing rehabilitation strategies, guiding patient and family decision making, and improving patient care. Methods: We conducted a retrospective analysis of 589 SCI patients admitted toa single acute rehabilitation facility and used the dataset to train advanced machine learning algorithms to predict patients’ rehabilitation outcomes. The primary outcome was the Functional Independence Measure (FIM) score at discharge, reﬂecting the level of independence achieved by patients after comprehensive inpatient rehabilitation. Results: Tree-based algorithms, particularly Random Forest (RF) and XGBoost, signiﬁcantly outperformed traditional statistical models and Generalized Linear Models (GLMs) in predicting discharge FIM scores. The RF model exhibited the highest predictive accuracy, with an R-squared value of 0.90 and a Mean Squared Error (MSE) of 0 . 29 on the training dataset, while achieving 0 . 52 Rsquared and 1 .37 MSE on the test dataset. The XGBoost model also demonstrated strong performance, with an R-squared value of 0.74 and an MSE of 0.75 on the training dataset, and 0.51 R-squared with 1.39 MSE on the test dataset. Our analysis identiﬁed key predictors of rehabilitation outcomes, including the initial FIM scores and speciﬁc demographic factors such as level of injury and prehospital living settings. The study also highlighted the superior ability of tree-based models to capture the complex, non-l

FAQs

What outcome does the study predict for spinal cord injury patients?

The study predicts the Functional Independence Measure (FIM) score at discharge, representing the independence achieved after inpatient rehabilitation.

Which machine learning models performed best?

Tree-based models, particularly Random Forest and XGBoost, significantly outperformed traditional statistical models and generalized linear models in predicting discharge FIM scores.

What key factors were identified as predictors of rehabilitation outcomes?

Initial FIM scores and demographic factors such as level of injury and prehospital living settings were identified as important predictors of rehabilitation outcomes.

Aug 5, 2026

Upload by Seraphina