Upper Limb Recovery Prediction Based on Multilevel Mixed Effect EMG Synergy and Biomarker Values

Ghada Muneer Bani Musa, Fady Alnajjar, Shingo Shimoda, Adel Al-Jumaily

Research output: Chapter in Book/Conference paperChapterpeer-review


Stroke is one of the most permanent causes of disability worldwide. The Stroke Impairment Assessment Set (SIAS) biomarker or Functional Independence Measure (FIM) biomarker is used to correct stroke deficiencies pre- and post-stroke and determine muscle rehabilitation strength. Clinical experts measure these clinical biomarker scores. However, the usefulness of biomarker values in predicting patients’ future rehabilitation recovery is limited. In contrast, electromyography signal data could help predict future rehabilitation recovery better. This chapter presents a prediction recovery technique for upper limb severe, moderate, and mild stroke patients using the improved multilevel mixed effect prediction method based on clinical biomarker scores. This technique analyses the electromyography signal during stroke rehabilitation activities. The results of this analysis are then used to predict the recovery rehabilitation for a few months using time series prediction and synergy electromyography data. We compare our results with the SIAS value that was scored during the examination. This assists the patient in finding out how much the success of our prediction is based on the biomarker value. We also examine how the synergy EMG of the affected or unaffected side of the body could predict the patient’s potential rehabilitation and performing daily activities of living and represent the level of recovery.

Original languageEnglish
Title of host publicationNon-Invasive Health Systems based on Advanced Biomedical Signal and Image Processing
PublisherCRC Press
Number of pages23
ISBN (Electronic)9781003838104
ISBN (Print)9781032386942
Publication statusPublished - 1 Jan 2024


Dive into the research topics of 'Upper Limb Recovery Prediction Based on Multilevel Mixed Effect EMG Synergy and Biomarker Values'. Together they form a unique fingerprint.

Cite this