Study shows artificial intelligence can predict language success after cochlear implants

An artificial intelligence model using deep transfer learning – the most advanced form of machine learning – predicts spoken language outcomes one to three years after cochlear implants (implanted electronic hearing aid) with 92% accuracy, according to a large international study published in JAMA Otolaryngology-Head & Neck Surgery.

Although cochlear implantation is the only effective treatment for improving hearing and facilitating speech for children with severe to profound hearing loss, speech development after early implantation is more variable compared to children born with normal hearing. If children who are likely to have more difficulties with speech are identified before implantation, intensive treatment can be offered earlier to improve their speech.

The researchers trained artificial intelligence models to predict outcomes based on preimplantation brain MRI scans from 278 children in Hong Kong, Australia and the US, who spoke three different languages ​​(English, Spanish and Cantonese). The three centers in the study also used different protocols for brain scanning and different outcome measures.

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

“Our results support the feasibility of a single AI model as a powerful predictor of language outcomes for children served by cochlear implant programs worldwide. This is an exciting advance for the field,” said senior author Nancy M. Young, MD, Medical Director of Audiology and Cochlear Implant Programs at Luririe at Child’s Hospital in the US. study.

This AI-powered tool enables a ‘predict-to-prescribe’ approach to optimizing language development by identifying which child may benefit from more intensive treatment.”

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

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

Dr. Young holds the Lillian S. Wells Professorship in Pediatric Otolaryngology at Lurie Children’s. She is also Professor of Otolaryngology at Northwestern University Feinberg School of Medicine, and Professor and Fellow at the Knowles Hearing Center, Department of Communication Sciences and Disorders at Northwestern University School of Communication.

The Lurie Children’s Cochlear Implant Program is one of the largest and most experienced in the world, with more than 2,000 cochlear implant procedures performed since its inception in 1991.