Machine learning approaches applied in spinal pain research

Published: Posted on

The latest review article from CPR Spine team members discusses the advantages and disadvantages of current analytical strategies in contemporary basic science and epidemiological research. The review considers how analytical machine learning approaches could provide the platform to harness the variability of patient presentations to enhance clinical prediction of pain persistence or recurrence.

Reference: Falla D, Devecchi V, Jiménez-Grande D, Rügamer D, Liew BXW. Machine learning approaches applied in spinal pain research. J Electromyogr Kinesiol. 2021 Dec;61:102599. doi: 10.1016/j.jelekin.2021.102599.

Link to the publication:  https://www.sciencedirect.com/science/article/pii/S1050641121000869?via%3Dihub

Mapping electromyography (EMG) alterations in individuals with LBP compared to controls in a lifting task, onto resultant class probabilities. FDboost first identifies the time-varying β-coefficient of each functional predictor, which represents the change in log odds for a unit change in predictor value from the control group. Second, the cumulative change over time in log-odds is determined for each functional predictor, and the cumulative change over predictors are combined additively and transformed to class probabilities. * reflects the instance where the EMG differences between groups are maximally different, which corresponds to the instance where the β-coefficient has the highest magnitude.

Leave a Reply

Your email address will not be published.