Study Shows AI Can Predict Speech Success After Cochlear Implants

In line with a serious international study published in , an AI model using deep transfer learning – probably the most advanced type of machine learning – predicted spoken language outcomes one to a few years after cochlear implants (implanted electronic hearing aid) with 92% accuracy

Although cochlear implantation is the one effective treatment for improving hearing and enabling spoken language in children with severe to profound hearing loss, spoken language development is more variable after early implantation in comparison with children with normal hearing. If it is set prior to implantation that children are prone to have greater difficulty with spoken language, intensified therapy could also be offered earlier to enhance their language.

Researchers trained AI models to predict outcomes based on pre-implantation brain MRI scans of 278 children in Hong Kong, Australia and the US who spoke three different languages ​​(English, Spanish and Cantonese). The three centers within the study also used different brain scanning protocols and different consequence measures.

Such complex, heterogeneous data sets are problematic for traditional machine learning, however the study showed excellent results with the deep learning model. It outperformed traditional machine learning models in all consequence measures.

“Our results reveal the feasibility of a single AI model as a sturdy prognostic tool for the language outcomes of kids served by cochlear implant programs worldwide. That is an exciting advance for the sphere,” said lead creator Nancy M. Young, medical director of audiology and cochlear implant programs at Ann & Robert H. Lurie Kid’s Hospital in Chicago – the U.S. center of the study.

This AI-powered tool enables a “predict-to-prescribe” approach to optimizing language development by identifying which child might profit from more intensive therapy.”

Nancy M. Young, Ann & Robert H. Lurie Kid’s Hospital of Chicago

This work was supported by Research Grants Council of Hong Kong Grant GRF14605119, the National Institutes of Health R21DC016069 and R01DC019387.

Dr. Young holds the Lillian S. Wells Professorship in Pediatric Otolaryngology at Lurie Kid’s. She can also be a professor of otolaryngology on the Feinberg School of Medicine at Northwestern University and a professor and fellow on the Knowles Hearing Center, Department of Communication Sciences and Disorders on the Northwestern University School of Communication.

Lurie’s pediatric cochlear implant program is considered one of the most important and most experienced on the earth. Since its introduction in 1991, greater than 2,000 cochlear implant procedures have been performed.