Translation of near infrared brain imaging to assess children with cerebral palsy

Cerebral palsy (CP) is the most common motor disorder of central origin in childhood and affects at least 2 children per 1000 live births every year. Neuroimaging techniques are needed to study neuroplastic rearrangements in the human brain in vivo as a result of CP. Unfortunately, accurate imaging from currently available techniques often requires the patients complete body confinement, steadiness and minimal noise for a long period of time, which limits the success rate to less than 50% for normal children and worse for CP-affected ones. In this work we show that functional near infrared (fNIR) imaging is robust to motion artifacts and has excellent potential as a sensitive diagnostic tool for this motor disorder. We have analyzed data from pediatric normal and CP patients performing finger-tapping and handwaving motor cortex activation tasks. From these analyses we have identified both spatial and temporal metrics of NIRbased motor cortex activation patterns that can clearly distinguish between normal and CP patients. We also present data from additional patients where signal processing methods are applied to filter out concurrently recorded hemodynamic signals due to breathing and cardiac pulsation. It is shown that filtering can substantially improve the quality of activation data, thus enabling more accurate comparison of activation patterns between normal and CP-affected children.