As groundbreaking as cochlear implants (CIs) are, these devices still have some limitations, particularly in configuring a CI specifically for an individual. A recent study from the University of Sydney School of Biomedical Engineering addressed this challenge and examined the use of output signal to noise ratio (OSNR) in predicting how a CI might fare for a recipient (Ear Hear. Sep/Oct 2020;41:1270-1281). It concluded that the OSNR model “could accurately predict individual recipient scores for a range of algorithms and noise types” and may be used as “a tool to assist researchers and clinicians in the development or fitting of CI sound processors.”
A cochlear implant uses sound processing algorithms to convert external sounds into electric signals, which are then sent directly to the auditory nerve. A vast number of algorithm parameter combinations could be programmed into the device, and evaluating their efficacy requires extensive testing.
Brett Swanson, PhD, who supervised the research, described the meticulous work it takes to test the viability of CI algorithms.
“A cochlear implant stimulates the auditory nerve directly, so if you're a researcher with normal hearing, you can't listen to it yourself,” he noted in a press release. “Instead, we rely on dedicated volunteers with cochlear implants who spend hours in soundproof rooms listening to sentences in noise and telling us what they hear. It is vital work, but mentally draining.”
A reliable prediction model such as the OSNR streamlines this testing process by calculating the speech intelligibility that a CI user would have across various listening conditions.
“This research has the potential to drastically reduce the amount of time that we need from our volunteers,” said Swanson.
More than reducing volunteer testing time, the OSNR method could mean that personalized configurations of CIs are soon possible for hearing loss patients.
Currently, because of the difficulty of CI testing, only a few algorithm parameter combinations are typically tested for each CI recipient. Finding a recipient's optimal configuration is a challenge.
“For practical reasons, few parameter settings are set for a specific recipient,” lead author Greg Watkins, PhD, told The Hearing Journal. “Threshold and maximum levels are set, and a small number of other parameters might be changed.”
But if ideal configurations could be accurately calculated, each patient's CI algorithms could be specifically configured to suit them best.
“Although the prediction methodology was designed as a lab tool, it does provide potential to evaluate different sound processor configurations for an individual ‘on the bench’ rather than in a sound booth,” Watkins added. “Potentially, this metric could be used to develop configurations, which are customized to an individual recipient's unique hearing capabilities.”
How did his team confirm the accuracy of the OSNR method?
“The prediction method was designed to predict how accurately individuals would understand speech if their sound processor algorithm was changed or reconfigured,” Watkins explained.
“The method takes existing ‘reference’ speech scores in one condition and maps them to a metric, the OSNR, to provide a map from OSNR to a speech score. Using computer simulation, the OSNR value for other listening conditions or sound processor configurations are calculated, and the reference score map is used to predict speech scores in the new condition.
“The method has been shown to be accurate for a range of noise types and sound processing conditions. It seems best suited to predicting performance when significant non-linear processing causes changes to the instantaneous gains applied during sound processing.”
An accurate prediction model would be highly valuable for clinicians hoping to maximize the benefit of CIs for their patients. Watkins himself knows the value of this from his own hearing loss history.
“I am an electrical engineer by training and worked in the telecommunications industry, in product development and project management, for many years. About 15 years ago, I started to experience hearing loss and this motivated me to return to university as a part-time PhD student at the University of Sydney, Australia,” he shared.
“My motivation was the idea that my technical experience and living with a disability might bring a new perspective to some of the challenges with CIs. During my studies, I received bilateral CIs and began the process of learning to hear with them—which is a process in itself and is not something that happens overnight.”
This study provides an opportunity for customized configurations or perhaps additional sets of “preferred configurations” that could be selected by an audiologist, said Watkins. “When? That depends on interest from [CI] manufacturers.”