Natural history of CMT disease: A 18 months follow-up study through gait analysis

s / Gait & Posture 37 (2013) S1–S31 S11Fig. 1. Materialsandmethods:FunctionalMagneticResonanceImag-ing (fMRI) task, consisting in observation of spatial targettrajectories and their recall from memory, was administered to 22healthysubjects.Theobservedmovementsconsistedinupperlimbreaching tasks (Fig. 1A) and computer generated dot movements(Fig. 1B) visualized in subsequent random sequences. The subjecthad to decide if trajectories were congruent with the given target.Control on responses was tested introducing trials with congruentand incongruent targets.Results: The observation of upper limb reaching and dot move-ments induced similar patterns of activations according to acaudo-rostral pathway (including specific areas of the occipital,temporal,parietal,andfrontallobes),seeorangeoverlappingareasin Fig. 2. Similarly, evidence of visual processing of movementswerefoundforthesubsequentretrievaloftheobservedmovement.Furthermore, activations of frontal cortical areas were evidentwhenthesubjectwasdecidingwhetherthetargettrajectorieswerecongruent or not.Discussion: We identified the neural pathways associatedwith visual processing of movement stimuli used in upper limbrobot-mediated training. In both study conditions, activationswere elicited in cerebral areas involved in visual perception, sen-sory integration, recognition of movement, re-mapping on thesomatosensory and motor cortex, memory recall and responsecontrol. Furthermore, we documented the brain’s ability to assim-ilate abstract object movements trajectories with human motorgestures. This results suggest the appropriateness of the visualfeedback provided during robotic motor training. The results ofour study highlight the role of fMRI in improving the understand-ing of visual motor analysis and decision processing during brainreorganisation after training.