Scientific research on tinnitus is still in its infancy and is limited by the fact that there is no objective or measurable method of determining tinnitus: there are no blood or urine tests, no brain scans, no genetic tests that can confirm whether someone has tinnitus or not.
We want to change this by funding a potentially groundbreaking research into objective measurement of tinnitus: a collaboration between Prof. dr. De Ridder, Prof. dr. Vanneste and Dr. Marco Congedo from the universities of Otago (New Zealand), Texas (USA), Dublin (Ireland) and Grenoble (France). They published in the leading journal “Nature Communications" that they can show with 88% certainty whether someone has ringing in the ears or not, based on an EEG, by means of artificial intelligence using two-dimensional machine learning.
An E.E.G. measures brain activity directly through the skull by picking up the electrical waves in the brain. Special software can then be used to check where the activity comes from in the brain. This is early beginning in the search for an objective measure of this enigmatic symptom.
The study analyzed brain activity based on E.E.G.’s from 153 tinnitus patients and compared that with 264 E.E.G.’s from people without tinnitus. This allows us to say with some certainty whether someone has tinnitus or not.
But with such a small group of patients, we cannot say how loud the tinnitus is and how disturbing it is, which is essential to really objectify the tinnitus. Moreover, the artificial intelligence that we used for this is simple: a support vector machine, which works two-dimensionally, can only give yes-no answers and only looks at brain activity, not at brain connections.
THE FOLLOW-UP RESEARCH
To come to a true objective measure of tinnitus, we need to apply more advanced artificial intelligence (A.I.), such as multidimensional (Riemannian) A.I., deep learning (multi-layered) A.I., genetic algorithms E.D.M.
We will have to compare 5 different forms of A.I. to find the best classifier: the most accurate, the most specific and the most sensitive, and which besides a yes-no answer (+tinnitus or-tinnitus) also quantifies the subjective perceived loudness and tinnitus-related suffering.
This requires powerful computers and time from an A.I. specialist. The research will take between 2 and 3 years.
The TinnitusFree Foundation wants to raise an amount of €300,000 via sponsorships and crowdfunding to finance this research.
A successful outcome of this research will make the tinnitus measurable and therefore negotiable.
Most of all, this will facilitate an explosion of research into treatments, help healthcare and encourage industry to invest in tinnitus.