Comprehensive signal interpretation of functional hand strength for activities of daily living (ADL) rehabilitation via multivariate data analysis (MVA)

The objective of this work is to study the functional hand strength for ADL rehabilitation tasks based on EMG signals and gripping/pinch force measurement via multivariate data analysis (MVA): Principal Components Analysis (PCA) and Projections to Latent Structures (PLS). The study allows us to identify weak muscles of patients with motor weakness, such as spinal cord injury (SCI) and post-stroke patients. We can then focus on rehabilitation activities to strengthen specific muscles. This analysis can also provide useful data for objective and quantitative assessment, towards control applications on the hand rehabilitation device being developed. From the study, subjects´ group and age are shown as dominant factors compared with other factors, such as gender, BMI and frequency of exercise. The PLS model is created to predict the outcome of the EMG results from the gripping/pinch force for the subjects showing with difficulties in EMG measurement.