Regression-Based Machine Learning for Predicting Lifting Movement Pattern Change in People with Low Back Pain.

Machine learning (ML) algorithms are crucial within the realm of healthcare applications. However, a comprehensive assessment of the effectiveness of regression algorithms in predicting alterations in lifting movement patterns has not been conducted. This research represents a pilot investigation using regression-based machine learning techniques to forecast alterations in trunk, hip, and knee movements subsequent to a 12-week strength training for people who have low back pain (LBP). The system uses a feature extraction algorithm to calculate the range of motion in the sagittal plane for the knee, trunk, and hip and 12 different regression machine learning algorithms. The results show that Ensemble Tree with LSBoost demonstrated the utmost accuracy in prognosticating trunk movement. Meanwhile, the Ensemble Tree approach, specifically LSBoost, exhibited the highest predictive precision for hip movement. The Gaussian regression with the kernel chosen as exponential returned the highest prediction accuracy for knee movement. These regression models hold the potential to significantly enhance the precision of visualisation of the treatment output for individuals afflicted with LBP.

Machine Learning Derived Lifting Technique in People without Low Back Pain.

This paper presents a method for determining the number of lifting techniques used by healthy individuals through the analysis of kinematic data collected from 115 participants utilizing an motion capture system. The technique utilizes a combination of feature extraction and Ward's method to analyse the range of motion in the sagittal plane of the knee, hip, and trunk. The findings identified five unique lifting techniques in people without low back pain. The multivariate analysis of variance statistical analysis reveals a significant difference in the range of motion in the trunk, hip and knee between each cluster for healthy people (F (12, 646) = 125.720, p < 0.0001).Clinical Relevance- This information can assist healthcare professionals in choosing effective treatments and interventions for those with occupational lower back pain by focusing rehabilitation on specific body parts associated with problematic lifting techniques, such as the trunk, hip, or knee, which may lead to improved pain and disability outcomes, exemplifying precision medicine.

Machine Learning Derived Lifting Techniques and Pain Self-Efficacy in People with Chronic Low Back Pain.

This paper proposes an innovative methodology for finding how many lifting techniques people with chronic low back pain (CLBP) can demonstrate with camera data collected from 115 participants. The system employs a feature extraction algorithm to calculate the knee, trunk and hip range of motion in the sagittal plane, Ward's method, a combination of K-means and Ensemble clustering method for classification algorithm, and Bayesian neural network to validate the result of Ward's method and the combination of K-means and Ensemble clustering method. The classification results and effect size show that Ward clustering is the optimal method where precision and recall percentages of all clusters are above 90, and the overall accuracy of the Bayesian Neural Network is 97.9%. The statistical analysis reported a significant difference in the range of motion of the knee, hip and trunk between each cluster, F (9, 1136) = 195.67, p < 0.0001. The results of this study suggest that there are four different lifting techniques in people with CLBP. Additionally, the results show that even though the clusters demonstrated similar pain levels, one of the clusters, which uses the least amount of trunk and the most knee movement, demonstrates the lowest pain self-efficacy.