Tinnitus Detector Circuit

Not sure if this has already been posted but I didn't find anything on a cursory search:

http://ieeexplore.ieee.org/document/6107720/?reload=true

@Cityjohn and any other guys who have access to an electronics lab to tinker around with... check it out and let me know!

I've attached the full article as well.

 

All I can say now is Wow... I'll have to read more after next week.

 

Ugh. This is smoke and mirrors. ECoG is an approach to measure electrical potentials from the surface of the brain. It is used routinely to measure neural processing from the brain of people with intractable epilepsy or tumors. Measuring ECoG requires a neurosurgeon to remove a portion of your skull. EEG is a distant echo of the ECoG that can be measured from the scalp. The technology described in the link above describes some fairly routine signal processing for breaking the ECoG or EEG signal down into different frequency bands with a low-power device using the ratio of those bands as a signature of tinnitus. There may be a slight, but statistically significant difference in the EEG in people with vs without tinnitus but it's not something that would have diagnostic value at the level of an individual person. The EEG and ECoG are too noisy and there are too many mitigating factors. I see no real innovation or deep thinking here.

 

@HomeoHebbian, you think with some more signal processing it would be possible to filter out the noise? Also, any suggestions on ways to objectively detect tinnitus? Personally I see that as the only way to understand it. I think it first needs to be detected, whatever the etiology, before any kind of intervention is tested. This will be a surefire way to see what works and what doesn't without relying on self-reporting.

 

When it comes to an objective biomarker of tinnitus, we don't know what is the signal and what is the noise. So, your question is right on the money, but it's hard for me to give you a straight up/down answer. These paper -

Lateralized tinnitus studied with functional magnetic resonance imaging: abnormal inferior colliculus activation.
Melcher JR, Sigalovsky IS, Guinan JJ Jr, Levine RA. J Neurophysiol. 2000 Feb;83(2):1058-72.

Tinnitus, diminished sound-level tolerance, and elevated auditory activity in humans with clinically normal hearing sensitivity.
Gu JW, Halpin CF, Nam EC, Levine RA, Melcher JR. J Neurophysiol. 2010 Dec;104(6):3361-70. doi: 10.1152/jn.00226.2010.

Brainstem auditory evoked potentials suggest a role for the ventral cochlear nucleus in tinnitus.
Gu JW, Herrmann BS, Levine RA, Melcher JR. J Assoc Res Otolaryngol. 2012 Dec;13(6):819-33.

got pretty close to an objective biomarker. They used fMRI and ERPs to show hyper-excitability in the inferior colliculus (a midbrain nucleus) and/or auditory cortex of subjects with tinnitus as compared to control subjects matched both for age and hearing loss. Even then, the difference emerged at the population level - i.e., the differences were not pronounced enough to determine whether an individual subject had tinnitus or not.

The 'real world' problem here is to plug someone in (without removing their skull, preferably) and use a combination of measurements from their ear and central nervous system to conclude whether they have tinnitus, in the same way that epileptiform events or seizure provide an objective biomarker for epilepsy. I think it is possible and this is something my lab is actively working on. I'm skeptical that an approach like the one described in the above invention disclosure will reach this high bar because factors like age, hearing loss, fatigue, attentional state, IQ etc. etc. can also affect measures of power in different frequency bands of the EEG/ECoG. What's needed for an objective measure of tinnitus is something more closely linked to an objective readout of the phantom sound, the ghost in the machine, so to speak. There is nothing in their approach that is specific to capturing the phantom sound. It seems to me that they are engineers who have developed an approach for a low-impedance sensor with low power consumption. That's great, I just bristle at the packaging as a "tinnitus detector".

I'm with you 100%. We need an objective biomarker. We cannot make meaningful progress on beating this thing with only subjective handicap indices etc. as proof of therapy (bcs of placebo effects etc.). Prof. Jennifer Melcher made some important strides in this direction. She has retired from tinnitus research but I aim to pick up where she left off so that we can get started - without further delay - with rigorous tests of potential therapies.