Assessment of Neuroplasticity Using EEG Signal in Rehabilitation of Brain Stem Stroke Patients

Abstract

Robot-assisted motor training provides an efficient alternative to conventional rehabilitation methods used for poststroke patients. The re-learning of lost motor functions happens through neuroplasticity in the brain. Electroencephalogram (EEG) provides an effective method for assessing neuroplasticity. Movement-related cortical potential (MRCP), an EEG-derived time-domain pattern, indicates changes due to motor skills gained as a result of the training. This study aims to perform a two-stage robot-assisted rehabilitation program on brain stem stroke patients consisting of a total of 24 training sessions and to assess whether significant motor recovery and neuroplasticity induction are achieved after the first stage or after completing both stages of the designed rehabilitation program. Three brain stem stroke patients were recruited for hand motor training on AMADEO rehabilitation robot for 8 weeks consisting of two stages of 4 weeks each. Three assessments methods which include standard clinical tests, hand strength and range of movement measurements using AMADEO assessment tool, as well as EEG signal acquisition, were performed at the beginning of all the training sessions (week 0), after completion of the first stage of rehabilitation (week 4) and after completion of both stages of the training sessions (week 8). The experimental results demonstrate that all brain stem stroke patients show significant functional hand motor recovery, as indicated by clinical tests, hand strength, and range of movement measurements, after completing 8 weeks of the training. Moreover, MRCP signal negative peak showed a significant decrease in its amplitude when the patients completed two phases of rehabilitation training, indicating neuroplasticity induction.