Related Experiment Video
Updated: Jun 13, 2025
06:28
Biomechanical Changes Related to Low Back Pain: An Innovative Tool for Movement Pattern Assessment and Treatment Evaluation in Rehabilitation
Published on: December 13, 2024
441
Identifying gait subgroups in low back pain patients with artificial intelligence: implications for individualized
Leonardo Metsavaht1,2, Felipe F Gonzalez3,4,5, Eliane Celina Guadagnin1
1Instituto Brasil de Tecnologias da Saúde (IBTS), Rio de Janeiro, Brazil.
Summary
Researchers identified five distinct gait profiles in patients with low back pain (LBP), revealing unique kinematic and physical characteristics for each group. These findings offer insights for targeted LBP management and understanding associated risks.
Area of Science:
- Biomechanics and Movement Science
- Orthopedics and Musculoskeletal Health
- Artificial Intelligence in Healthcare
Background:
- Low back pain (LBP) affects a significant portion of the population, often leading to altered movement patterns.
- Understanding diverse gait characteristics in LBP patients is crucial for effective clinical assessment and treatment.
- Previous research has not fully elucidated the distinct gait profiles and their associated clinical features in LBP.
Purpose of the Study:
- To investigate the existence of different gait profiles among individuals experiencing low back pain (LBP).
- To assess the clinical characteristics associated with each identified gait profile in LBP patients.
- To explore the potential of artificial intelligence in classifying LBP-related gait abnormalities.
Main Methods:
- A cross-sectional retrospective study analyzed 3D gait kinematics (trunk/pelvis motion, coordination) in 111 LBP patients.
- An AI algorithm, employing PCA, SOMs, and K-means clustering, was used to identify distinct gait profiles.
- Clinical data including demographics, passive range of motion (ROM), and hip strength were compared across profiles.
Main Results:
- Five distinct gait profiles were identified: Flexed Trunk, Lumbar Rectification, Pelvic Impairment, Trunk Extension/Excessive Rotation, and Tight Axial Control.
- Profiles showed significant differences in trunk and pelvic kinematics, with specific profiles linked to sex and body mass.
- Certain profiles exhibited altered hip passive ROM, while age, hip strength, and passive trunk ROM showed no significant inter-profile differences.
Conclusions:
- The study successfully identified five unique gait profiles in LBP patients, each with distinct kinematic and physical attributes.
- These profiles provide valuable insights into potential clinical implications, anatomical structures at risk, and tailored management strategies for LBP.
- The findings underscore the heterogeneity of gait in LBP and highlight the utility of AI in uncovering these patterns.
