Thanks for your feedback, this will help us improve our content for you!
AD/PD 2023 | Digital markers detect changes in neurocognitive performance in those with vs without hearing loss
M. Florencia Iulita, PhD, Altoida, Inc., shares the results of a pilot study investigating the performance of a digital biomarker platform in characterizing the neurocognitive profiles of individuals with/without hearing loss. The study looked at the cognitive profiles of three groups: cognitively normal (CN) with hearing loss, CN without hearing loss, and mild cognitive impairment (MCI) with hearing loss. The Altoida digital neuro-signature (DNS) assessment was conducted on a smart device, where the individual had to perform speech and motor tasks and augmented reality exercises reflective of activities of daily living. The study found that individuals with MCI and hearing loss had the lowest Altoida DNS scores, indicating lower cognitive performance. Those who were CN with hearing loss performed similarly to those with MCI and hearing loss, but significantly differently compared to those who were CN without hearing loss. These data demonstrate the presence of distinct cognitive profiles in those with and without hearing loss, and that these differences can be captured using digital biomarker signatures. This interview took place at the AD/PD™ 2023 congress in Gothenburg, Sweden.
These works are owned by Magdalen Medical Publishing (MMP) and are protected by copyright laws and treaties around the world. All rights are reserved.
Transcript (edited for clarity)
First, I just want to say that I am very excited to be here in Gothenburg for this edition of AD/PD. It’s a conference that I have been attending for many years and I can’t wait to hear and learn about all the progress that has been made in the field from other fellow scientists, clinicians, and industry delegates. So the work we brought to the conference are the results of a pilot study where we evaluated and applied a digital biomarker assessment, developed by Altoida, to characterize the cognitive performance of individuals with and without hearing loss in a group that was identified as cognitively normal by clinical and neuropsychological assessment and a group that already started the study with mild cognitive impairment...
First, I just want to say that I am very excited to be here in Gothenburg for this edition of AD/PD. It’s a conference that I have been attending for many years and I can’t wait to hear and learn about all the progress that has been made in the field from other fellow scientists, clinicians, and industry delegates. So the work we brought to the conference are the results of a pilot study where we evaluated and applied a digital biomarker assessment, developed by Altoida, to characterize the cognitive performance of individuals with and without hearing loss in a group that was identified as cognitively normal by clinical and neuropsychological assessment and a group that already started the study with mild cognitive impairment.
So perhaps you might wonder what is the interest in hearing loss in this context? And several years of research have shown that hearing loss is associated with accelerated cognitive decline and the two last reports from the Lancet commission have actually flagged hearing loss as a key modifiable factor for dementia prevention. So we are interested in Alzheimer’s disease so of course in this context, this topic was terribly important. So as I was explaining, this study was about characterizing the cognitive profile of these three groups with a digital biomarker. We compared the cognitive performance of these three groups, cognitively normal without hearing loss, cognitively normal with hearing loss and mild cognitive impairment with hearing loss, with a digital biomarker called Altoida DNS. DNS stands for Digital Neuro Signature. And what this is, this is a research device based on machine learning that simulates conducting activities of daily living. So it’s an assessment that can be done on a smart device. It takes approximately ten minutes, and the user has to perform a series of motoric exercises and augmented reality exercise. So for example, tapping and tracing shapes or placing and finding objects in a virtual environment. And with this, the device captures a series of digital and multimodal features to assess the performance of the individual. And what I was explaining is that through machine learning, these models can be trained to identify distinct phenotypes. And in this case the model that we applied is a machine learning model that was trained with clinical cohort data to identify mild cognitive impairment.
So what we found was that individuals with mild cognitive impairment and hearing loss had the lowest Altoida scores, so indicating a lower cognitive performance. And what was interesting is that the performance of individuals who were classified as cognitively normal by clinical assessment but who had hearing loss was not different from those with mild cognitive impairment and hearing loss. And we also detected a small but significant difference between cognitively normal without hearing loss and cognitively normal with hearing loss. And what this tells us is that these digital biomarkers are able to capture these subtle differences in cognitive performance between the three groups.
Within a subset of these populations, we segregated them by amyloid positivity. So individuals who were amyloid positive or amyloid negative by CSF assessment. And what we found is that those who were amyloid positive had a lower cognitive performance or lower scores on the Altoida assessment, compared with those who were amyloid negative. And the last part of the work, what we did was to explore some correlations between the digital biomarker and current neuropsychological scales that are used. And we found significant associations between the digital biomarkers of the Altoida platform and these common neuropsychological scales like the MMSE or for example, a test of episodic memory and so on.
Florencia Iulita is an employee of Altoida Inc. and holds stock options in the company.