Presurgical MRI May Predict Speech Outcomes Following Cochlear Implant

This article originally appeared here.
Regions of brain that were unaffected by auditory deprivation provide most precise prediction results
Regions of brain that were unaffected by auditory deprivation provide most precise prediction results

HealthDay News — Brain areas unaffected by auditory deprivation can predict speech outcomes after cochlear implant (CI) in children, according to a study published online January 15 in the Proceedings of the National Academy of Sciences of the United States of America.

Gangyi Feng, PhD, from the Chinese University of Hong Kong, and colleagues examined the neurobiological basis of variability in outcomes using pre-surgical neural morphological data obtained from magnetic resonance imaging of individual pediatric CI candidates implanted below age 3.5 years to predict variability of improvement in speech perception after surgery. Neuroanatomical density and spatial pattern similarity of CI candidates was compared with that of age-matched children with normal hearing. 

Related Articles

The researchers detailed neuroanatomical networks that were affected or unaffected by auditory deprivation based on the comparison of CI candidates and control children. Using this information, machine learning models were built to predict the individual children's speech development following CI. The highest accuracy, specificity, and sensitivity in patient classification and most precise prediction results were seen for the regions of the brain that were unaffected by auditory deprivation, in particular the auditory association and cognitive brain regions.

"These findings suggest that brain areas unaffected by auditory deprivation are critical to developing closer to typical speech outcomes," the authors write. "Moreover, they suggest that determination of the type of neural reorganization caused by auditory deprivation before implantation is valuable for predicting post-CI language outcomes for young children."

Abstract
Full Text (subscription or payment may be required)