RP402 - Objective Measurement of Tinnitus Using Functional Near-Infrared Spectroscopy

Abstract:
Lack of an objective measure of tinnitus presence and severity remains a significant obstacle to patient care and development of treatments. In an initial study we showed feasibility of a non-invasive brain imaging technique called functional near-infrared spectroscopy (fNIRS) and machine learning data analytics to differentiate data from tinnitus patients from controls as well as tinnitus patients at different severity levels as rated subjectively. We have now validated our findings in a larger study.

Summary:
Background
Chronic tinnitus, hearing sounds not present externally, is a debilitating condition, which affects 10–20% of adults. Tinnitus in its severe form is accompanied by symptoms such as anxiety, depression and cognitive dysfunction. Currently there is no objective measure of tinnitus that can be used clinically. Clinical assessment of the condition uses subjective feedback from individuals, which is not always reliable and is prone to placebo effects. In an initial study, we showed sensitivity of functional near-infrared spectroscopy (fNIRS) to tinnitus presence and severity. We have now validated our initial findings in a larger confirmatory study.

Objective
We investigated the sensitivity of fNIRS to differentiate individuals with and without tinnitus and to identify fNIRS features associated with subjective ratings of tinnitus severity.

Methods
fNIRS is a non-invasive brain imaging technique which uses near-infrared light to monitor changes in blood oxygen levels allowing imaging and analysis of the oxygen demands of active brain regions. fNIRS measures changes in oxygenated and de-oxygenated hemoglobin concentrations using a montage of light sources and detectors placed on a cap. We recorded fNIRS signals in the resting state and in response to auditory or visual stimuli from 100 individuals with chronic tinnitus and 58 controls. Severity of tinnitus was rated using the Tinnitus Handicap Inventory and subjective ratings of tinnitus loudness and annoyance were measured on a visual rating scale. Differences in both evoked responses and measures of connectivity (derived from resting state data) were compared between groups. These signal features are now being used with machine learning data analytics to classify tinnitus patients from controls and tinnitus patients at different severity levels with high accuracy.

Results
Resting state measures of connectivity between temporal regions and frontal and occipital regions showed significant difference when compared to controls. In the tinnitus group both visual and auditory evoked responses were significantly reduced in the visual and auditory regions of interest. Our results to date show accuracy of over 80% in differentiating both tinnitus from controls using fNIRS signal features as well as mild versus severe tinnitus as rated subjectively.

Conclusion
Our findings validate, in a larger sample, our initial findings which showed feasibility of using fNIRS to develop an objective measure of tinnitus. An objective measure of tinnitus presence and severity would greatly benefit clinicians, researchers and patients by providing a tool to assess new treatments and monitor patients’ treatment progress. Such a tool is an important step in developing effective treatments for tinnitus.