Signal processing and machine learning analysis of IMU-based lower-limb kinematics in spinal sagittal imbalance
- 1Department of Biomedical Engineering, Amirkabir University of Technology
- 2Department of Computer and Data Sciences, Faculty of Mathematical Sciences, Shahid Beheshti University
- 3Bone and Joint Reconstruction Research Center, Shafa Orthopedic Hospital, Iran University of Medical Sciences
Abstract
Gait problems in patients with spinal sagittal imbalance (SSI) pose significant challenges that often persist even after surgical intervention. Identifying movement disorders preoperatively and addressing them postoperatively is crucial for improving patient outcomes. Despite advances in surgical techniques, many patients continue to experience gait abnormalities, highlighting the need for objective assessment methods. This study had three primary aims: (1) to compare the walking patterns of SSI patients before surgery with those of controls; (2) to compare pre-surgical and post-surgical gait within the same patients; and (3) to determine whether lower-extremity gait variability differs between patients and controls. Our results show that SSI patients exhibit significantly higher stride-to-stride variability than controls, and that this variability persists—though with some improvement—at three months post-surgery. The results demonstrated that the SVM model, with an accuracy above 97%, effectively differentiated pre-surgical and post-surgical gait data, revealing that while some improvements were observed, many patients retained gait abnormalities. These findings suggest the need for targeted rehabilitation focused on limb-specific recovery, emphasizing the importance of using ML-based gait analysis for more effective post-surgical interventions.
Related
- Spinopelvic compensation and functional outcomes after short-segment lumbar fusion in patients with residual sagittal imbalance
- Predicting One-Year Psychiatric Rehospitalization Using Natural Language Processing and Machine Learning Methods Based on Clinical Texts
- Hierarchical affective predictive imbalance in adolescent borderline personality disorder: task-evoked fMRI evidence from multimodal machine learning
- Localization-Aware Multiscale Deep Learning for Lumbar Foraminal Stenosis in Multi-Scanner Sagittal MRI: A Leakage-Controlled Evaluation
- Incorporating natural language processing-derived features from pathology reports into machine learning models to predict differentiated thyroid cancer recurrence
BibTeX
@article{madadi2026signal,
title = {{Signal processing and machine learning analysis of IMU-based lower-limb kinematics in spinal sagittal imbalance}},
author = {Sadegh Madadi and Mostafa Rostami and Hadi Farahani and Farshad Nikouee},
journal = {Scientific Reports},
year = {2026},
month = jul,
doi = {10.1038/s41598-026-63130-w},
url = {https://doi.org/10.1038/s41598-026-63130-w},
note = {CC BY},
}
RIS
TY - JOUR
TI - Signal processing and machine learning analysis of IMU-based lower-limb kinematics in spinal sagittal imbalance
AU - Sadegh Madadi
AU - Mostafa Rostami
AU - Hadi Farahani
AU - Farshad Nikouee
PY - 2026
DA - 2026/07/01
JO - Scientific Reports
DO - 10.1038/s41598-026-63130-w
UR - https://doi.org/10.1038/s41598-026-63130-w
AB - Gait problems in patients with spinal sagittal imbalance (SSI) pose significant challenges that often persist even after surgical intervention. Identifying movement disorders preoperatively and addressing them postoperatively is crucial for improving patient outcomes. Despite advances in surgical techniques, many patients continue to experience gait abnormalities, highlighting the need for objective assessment methods. This study had three primary aims: (1) to compare the walking patterns of SSI patients before surgery with those of controls; (2) to compare pre-surgical and post-surgical gait within the same patients; and (3) to determine whether lower-extremity gait variability differs between patients and controls. Our results show that SSI patients exhibit significantly higher stride-to-stride variability than controls, and that this variability persists—though with some improvement—at three months post-surgery. The results demonstrated that the SVM model, with an accuracy above 97%, effectively differentiated pre-surgical and post-surgical gait data, revealing that while some improvements were observed, many patients retained gait abnormalities. These findings suggest the need for targeted rehabilitation focused on limb-specific recovery, emphasizing the importance of using ML-based gait analysis for more effective post-surgical interventions.
SN - 2045-2322
ER -