Effect Of Novel Brain-computer Interfaces Combined With Functional Electrical Stimuiation(BCI-FES) System On The Brain Plasticity In Subjects With Chronic Stroke

Objective: To study the feasibility of a new rehabilitation training system--braincomputer interfaces combined functional electrical stimulation(BCI-FES) in subjectswith chronic strokeand its neural mechanisms of the training effects.Methods: Foursubjects with chronic stroke,were assigned to BCI-FES group or FESgroup for four weeks(20days),5days per week, and40minutes per day. Allmeasurements were done pre and post training while only upper limb functionalassessments were done in4weeks follow-up stage while subjects received notraining.Onesubject in the BCI-FES group received FES training for another4weeksafter follow-up training and20-week elution period. Measurements included upper limbfunction tests and functional magnetic resonance imaging(fMRI) sessions and diffusiontensor imaging(DTI) scans. Upper limb function tests included the maximumgrasp-relax speed of paralyzed hand, the upper limb Fugl-Meyer scores, the manualmuscle test(MMT), the abalone column test, the Box&Block test and the Jebson-Taylertest. The fMRI sessions were composed by “motor”,“imagery” and “rest” blocks. ThefMRI data were analyzed by AFNI software and depicted the images of activatedclusters. The fractional anisotropy(FA) values, the FA index, and theapparent diffusioncoefficient(ADC) values of regions of interest(ROIs) were calculated from the DTI data.Results:1. Both groups showed no improvement after training in MMT scores.Themaximum grasp-relax speeds of paralyzed hand in BCI-FES group increased aftertraining, but decreased in FES group. Other hand motorfunctional tests improved in allpatients after training; however, the improvement in BCI-FES group was moresignificant than in FES group.2. Upper limb motor function improved at follow-up incomparison withpre-training, while decreased when compared with post-training.3. Inpresent study all subjects exhibited abnormal fMRI activation patterns when task performed using theirparetic hand. In comparison to the normal adults, theyshowedmore activity in ipsilateral primary motor cortex(M1) area and in secondary motor areas,while less activity in contralateral M1area.4. After training, BCI-FES group showedmore brain areas activated thanpre-training,while not shown in FES group whensubjects were asked to perform the “motor” or“imagery” task.5. When comparedbetween tasks,”imagery” task activated more secondary motor areas and less M1areawhen compared with “motor” task.6. The volume of activated brain for“motor” and“imagery” task of the affected hand was increased in both groups after training, and theincrease was more significant in BCI-FES group.7. The average laterality index(LI)value of performing “motor” and “imagery” task with the affected hand increased inBCI-FES group after training compared to pre-training, while it decreased in FES group.8. Less white matter fiber tracts were observed in the peri-lesional area and theipsilesional posterior limb of internal capsule compared to the contralesionalhemisphere in all subjects.9. After training, more white matter fiber tracts wereobserved in the peri-lesional area and in the ipsilesional posterior limb of internalcapsule in BCI-FES groupbut not in FES group.10. The average FA values and FAindex in peri-lesional area increased in both groups after training, and the increase wasmore significant in BCI-FES group. The average FA values and FA index in theipsilesional posterior limb of internal capsule increased after training in BCI-FES group,but decreased in FES group.11. The average ADC values in the ipsilesional ROIsdecreased after training in BCI-FES group but increased in FES group.12. The increaseof LI values of performing“motor” task with the affected hand showed a significantpositive correlation with the increase of FA values in the peri-lesional areas (P=0.017，r=0.9412) and in the ipsilesional posterior limb of internal capsule(P=0.034，r=0.9065).Conclutions:1. BCI-FES is a feasible rehabilitation therapy for stroke survivors.2.BCI-FES training can promote the motor function recovery of upper extremities forpatients with chronic stroke.3. The mechanism of the recovery by using BCI-FES isproposed to modulate neuro-plasticity both in cortical level and subcortical level.