Brain Activation in Primary Motor and Somatosensory Cortices during Motor Imagery Correlates with Motor Imagery Ability in Stroke Patients

Your doctor should be able to translate this into a stroke protocol.
Full article at the link.
http://www.hindawi.com/isrn/neurology/2012/613595/
Abstract
Aims. While studies on healthy subjects have shown a partial overlap between the motor execution and motor imagery neural circuits, few have investigated brain activity during motor imagery in stroke patients with hemiparesis. This work is aimed at examining similarities between motor imagery and execution in a group of stroke patients. Materials and Methods. Eleven patients were asked to perform a visuomotor tracking task by either physically or mentally tracking a sine wave force target using their thumb and index finger during fMRI scanning. MIQ-RS questionnaire has been administered. Results and Conclusion. Whole-brain analyses confirmed shared neural substrates between motor imagery and motor execution in bilateral premotor cortex, SMA, and in the contralesional inferior parietal lobule. Additional region of interest-based analyses revealed a negative correlation between kinaesthetic imagery ability and percentage BOLD change in areas 4p and 3a; higher imagery ability was associated with negative and lower percentage BOLD change in primary sensorimotor areas during motor imagery.1.


Introduction
The residual disability after stroke is substantial, with about 65% of patients at 6 months unable to effectively incorporate the paretic hand into daily activities [1, 2]. In turn, the degree of functional deficit contributes to a reduced quality of life after stroke [2–6]. Studies on healthy volunteers have shown that mental practice with motor imagery can improve the performance of motor skill behaviours [7, 8]. Motor imagery can be defined as a dynamic state during which the representation of a specific motor action is internally reactivated within working memory without any overt motor output [9]. According to the motor simulation hypothesis [10], and the grounded cognition perspective [11], represented actions correspond to covert, quasi-executed actions, involving a partial reenactment of the mechanisms that normally participate in various stages of action generation [10]. Therefore, motor imagery is a cognitive process based on sensorimotor simulation mechanisms, where individuals implicitly reenact actions without producing an overt motor output [10]. A certain degree of similarity between brain areas activated during motor imagery and execution has indeed been demonstrated [10, 12], providing empirical support for such theoretical framework.