Non-profit organization The Bionics Institute of Australia and Deakin University collaborated to develop an objective way of measuring tinnitus. The findings were recently published in the journal PLOS ONE.
Tinnitus occurs when one hears phantom noises such as buzzing, ringing, or a whirring sound. The noise may be persistent, subtle, loud, or happens occasionally. It is also not an illness but can be a symptom of conditions such as hearing damage or an ear infection.
In its chronic form, tinnitus can lead to stress, cognitive dysfunction, and even depression. However, it is often mild and affects about 20% of adults.
Existing Research on Tinnitus
Previous animal and human studies have associated tinnitus with the central nervous system, namely auditory areas in the brain. A study from 2013 suggested that the symptom is the result of "abnormal activity in multiple overlapping brain networks." They also proposed that activity in the auditory cortex also affects how loud the phantom noise can be.
Despite how common tinnitus is, no clinical tests exist to measure and assess tinnitus. Moreover, a recent study from Anglia Ruskin University described how coronavirus has affected people with pre-existing tinnitus with no available treatment or developed the symptom after getting infected.
Objective Measurements
In the new study, researchers used a brain imaging technique called functional near-infrared spectroscopy (fNIRS) where light is used to measure brain oxygen activity. This was accompanied by machine learning algorithms to assess tinnitus-related brain activity.
25 volunteers with chronic tinnitus and 21 controls wore a cap that shines lights into their heads and recorded the reflected light. The lights would change depending on a resting state and how the brain responded to auditory and visual stimuli.
The cap had 16 sources and 16 detectors, forming pairs called channels, covering the frontal, temporal, and occipital cortical regions. The auditory stimuli consisted of short segments of pink noise, a sound resembling a waterfall that masks unwanted noise via earphones. The visual stimulus consisted of circular checkerboard patterns that trigger cortical responses.
Those with tinnitus, wrote the researchers,"temporal-occipital connectivity showed a significant increase with subject ratings of loudness." The imaging also showed which cortical regions are active in tinnitus such as the auditory cortex and frontal cortex. The machine learning algorithms would then classify patients if they had tinnitus or not as well as severity levels of the symptom.
The fNIRS and machine learning combination classified patients with tinnitus with an accuracy of 78.3%, noted the study. The algorithms also had an accuracy of 87.32% in rating individuals with either slight/mild and moderate/severe tinnitus.
One of the limitations in the study, wrote the authors, is that the near-infrared light cannot reach deep cortical regions. There are also several subtypes of tinnitus while the cap can only classify general tinnitus. In conclusion, the researchers said objectively measuring tinnitus can help with the development of new treatments and assess the progress of patients.
Read Also: The Future of NANO Type Hearing Aids: What Research Says
Check out more news and information on Tinnitus on Science Times.
