Analysis of Force Profile Features in Spinal Manipulation Therapy

Spinal manipulation therapy (SMT) is widely used as an intervention for musculoskeletal conditions. However, the automated detection and analysis of force profile features in SMT have received limited attention. This study aims to address this research gap by developing a toolbox for the automatic detection and annotation of force-time profile features in SMT. For validation purposes, we will investigate the correlation between these features and characteristics of patient vignettes. Force data was collected from 1233 SMT interventions using a commercially available pressure sensor. With the aggregation of three feature selection methods (Chi squared, MRMR, and ReliefF), the results indicate a significant increase in maximum thrust speed for mentally envisioned athletic male patients compared to elderly females (
$p< 0.01$
). To the best of our knowledge, this study is the first of its kind, representing a pioneering exploration of automated force profile analysis in SMT. The findings hold immense potential to advance technology, support training of manual interventions, and facilitate the development of objective treatment feedback tools. The observed correlations between the extracted features and patient characteristics provide valuable insights for personalized SMT approaches.