AI improves Cochlear implant placement

Auditory specialists placing Cochlear implants in hearing-impaired patients may position the devices better with an assist from AI than with conventional methods, according to researchers in Belgium who have published two case studies in Cochlear Implants International.

Neuroscience doctoral candidate Justine Wathour of the Catholic University of Louvain and colleagues selected two adults who were experiencing disappointing results with their manually fitted Cochlear implants (CIs), according to the study abstract.

One patient had undergone 19 fitting sessions over 17 months yet still could not recognize words spoken in open sets.

The other patient had developed good open-set word recognition over a nine-month period of manual adjustments but struggled to understand speech at soft and loud volume levels.

It took clinicians two tries to get AI-proposed placement just right on the first subject. But on the second try, they used an AI approach based on universal, population-based statistics. This time the patient succeeded with open-set word recognition, and in just three months.

The second patient’s word-recognition scores quickly improved with the AI placement proposals, even at the volume extremes that previously were problematic.

Moreover, the AI-suggested modifications in both cases seemed to be atypical, suggesting the placements probably wouldn’t have been tried without the aid of AI.

Wathour et al. concluded the two case studies “illustrate that adults implanted with manual Cochlear implant fitting may experience an improvement in their auditory results with AI-assisted fitting.”