Scientists are using a mind-reading cap to detect an invisible condition

Up to 15 percent of adults around the world are plagued by a constant roaring or ringing in the ears that only they can hear. And with no way to test or treat it, this subjective experience, called tinnitus, can go undiagnosed for years.

But in a new study, researchers announce they can objectively test for both the presence and severity of someone's tinnitus for the first time, using a brain-reading cap and machine learning. In their trials, this technique was able to determine the severity of a person's tinnitus with close to 90 percent accuracy.

Having a way to objectively measure tinnitus will not only ease the burden felt by millions of people who struggle to convince people of their condition, but will also help develop clinical treatments for this condition.

The findings were published Wednesday in the journal PLOS One.

Cutting through the noise — What causes tinnitus appears to be as varied and subjective as its resultant ringing. But it is often the result of hearing damage, caused by explosions on a battlefield, a noisy construction site, or even the staccato rumble of drums. It's theorized that the noise damage affects the sound-sensitive cells of the cochlea, an organ in the inner-ear, as well as auditory pathways in the brain.

The noise tinnitus creates isn't the same for everyone. For some, it may be so soft that it's barely noticeable, but for those with chronic tinnitus, it can drastically affect their daily lives and lead to feelings of depression and stress.

The stakes are high — as many as 20 percent of people have tinnitus, and a fifth of them may experience severe mental health issues as a result.

"Chronic tinnitus affects around 6–20 [percent] of adults with approximately 20 [percent] of these experiencing it in a severe form with symptoms such as depression, cognitive dysfunction and stress," the authors write in the paper. "Despite its wide prevalence, there is currently no clinically used objective test that can measure changes in brain activity related to tinnitus. There is no objective way to determine the presence or severity of tinnitus, or assess whether treatments are effective."

But it doesn't stop at changing one's outlook on life. Tinnitus appears to change the brain's activity and perhaps its circuitry, too. Previous studies that use brain scanning techniques like EEG, MEG, and PET suggest tinnitus can cause spontaneous neural activity, as well as changes in neural synchrony and reorganization of the brain's sound map.

To investigate these tell-tale neural signals better, researchers turned to a different kind of brain-scan technology called functional near-infrared spectroscopy (fNIRS.) fNIRS scans are done using a cap worn on a person's head like a winter beanie. It is portable, cheap, and quieter than other technologies, the researchers say.

Suiting up — For their study, researchers recruited 25 people diagnosed with tinnitus and 21 people without. The participants were matched for both age and hearing loss.

The tinnitus group rated the severity of their condition using the Tinnitus Handicap Inventory. Subjective measurements of tinnitus loudness and annoyance were also recorded.

All the participants then suited up in their fNIRS caps and completed three different trials. In the first, researchers measured their neural activity at rest while the participant sat in an empty, quiet room. For the second two trials, researchers measured the participants' neural activity when exposed to auditory stimuli in the form of "pink" noise, and visual stimuli in the form of abstract shapes.

The researchers then used statistical modeling and machine learning to find patterns in the resulting neural signals.

What they found — Analysis of the fNIRS scans revealed a significant statistical difference in the neural connectivity between cortical areas of the brains of people with and without tinnitus. Interestingly, the researchers also found that people with tinnitus displayed damped responses to both visual and auditory stimuli. This suggests tinnitus may not just interfere with how a person processes sound signals, but other sensory information, too. Essentially, tinnitus may alter a person's entire experience of the world.

Armed with these scans, a machine learning algorithm was able to differentiate between tinnitus and non-tinnitus brains with 78 percent accuracy. It could also tell the severity of a person's tinnitus with 87 percent accuracy from their brain activity alone.

What's next — The test developed here is a promising first step, but it is not quite ready to roll out to clinics around the world. But ultimately this could lead to the development of effective tests and treatments for this common and debilitating condition, the authors say.

"Much like the sensation itself, how severe an individual’s tinnitus is has previously only been known to the person experiencing the condition," the scientists said in a statement accompanying the research.

"Our ability to track the complex changes that tinnitus triggers in a sufferer’s brain is critical for the development of new treatments.”

Abstract: Chronic tinnitus is a debilitating condition which affects 10–20% of adults and can severely impact their quality of life. 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. We investigated the sensitivity of functional near-infrared spectroscopy (fNIRS) to differentiate individuals with and without tinnitus and to identify fNIRS features associated with subjective ratings of tinnitus severity. We recorded fNIRS signals in the resting state and in response to auditory or visual stimuli from 25 individuals with chronic tinnitus and 21 controls matched for age and hearing loss. Severity of tinnitus was rated using the Tinnitus Handicap Inventory and subjective ratings of tinnitus loudness and annoyance were measured on a visual analogue scale. Following statistical group comparisons, machine learning methods including feature extraction and classification were applied to the fNIRS features to classify patients with tinnitus and controls and differentiate tinnitus at different severity levels. Resting state measures of connectivity between temporal regions and frontal and occipital regions were significantly higher in patients with tinnitus compared to controls. In the tinnitus group, temporal-occipital connectivity showed a significant increase with subject ratings of loudness. Also in this group, both visual and auditory evoked responses were significantly reduced in the visual and auditory regions of interest respectively. Naïve Bayes classifiers were able to classify patients with tinnitus from controls with an accuracy of 78.3%. An accuracy of 87.32% was achieved using Neural Networks to differentiate patients with slight/ mild versus moderate/ severe tinnitus. Our findings show the feasibility of using fNIRS and machine learning to develop an objective measure of tinnitus. Such a measure would greatly benefit clinicians and patients by providing a tool to objectively assess new treatments and patients’ treatment progress.