Algorithm predicts speech outcomes after cochlear implant

An algorithm developed by researchers at the Chinese University of Hong Kong has accurately predicted speech improvement in children who use cochlear implants (CIs). Findings were published in the Proceedings of the National Academy of Sciences of the United States of America.

This study described how the algorithm could improve personalized therapy for children with CIs to improve language development. Led by Gangyi Feng, PhD, from the Chinese University of Hong Kong, a team of researchers collected data using pre-surgical magnetic resonance images of pediatric CI patients to predict the variability in perceived speech improvement.

A machine learning model, trained on information of neuroanatomical networks affected or unaffected by auditory deprivation, was evaluated in comparisons of CI patients and a control group. Results showed the algorithm was most accurate, specific and sensitive for patient classification. It was precise in prediction when the region of the brain being evaluation was unaffected by auditory deprivation.

“These findings suggest that brain areas unaffected by auditory deprivation are critical to developing closer to typical speech outcomes," wrote Feng and colleagues. "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."