Above: Long (Jason) Lu, PhD
Below: Scott Holland, PhD
A new computer program may be able to predict whether hearing-impaired children will develop language skills after cochlear implant surgery.
A study published Oct. 12, 2015, in Brain and Behavior, details how the program analyzes functional magnetic resonance images (fMRI) to show how regions of infants’ brains respond to auditory stimulus tests given before surgery. The computer model was produced by a team led by Long (Jason) Lu, PhD, and Scott Holland, PhD, of the Pediatric Neuroimaging Consortium.
“This study identifies two potential biomarkers for predicting cochlear implant outcomes,” Lu says.
The model detects heightened activity in the left hemisphere’s speech-recognition and language-association areas, and physical variations in the right cerebellar structures. After fMRI data is collected, the computer algorithm uses a process called Bag-of-Words to predict which children are good candidates for cochlear implants.
“This is one of the first successful methods for translating data from fMRI of hearing-impaired children into something with potential for practical clinical use,” says Lu.