Parkinson's disease is a degenerative disorder of the central nervous system. Currently there is no cure for this condition with no advanced methods for diagnosis available. As a measure, Mathematician Max Little, who was recently made a TED Fellow, has devised a cheap and quick test to identify the disease. This test involves voice algorithms which analyse data to detect signs of Parkinson's.
Max will be commencing the TEDGlobal conference in Edinburgh, calling for volunteers to contribute to the voice database. By implementing the voice algorithms, the system was able to spot those with Parkinson's disease with an accuracy of 86%.
The system is fitted with voice data from various patients. With every test, the system "learns" to detect differences in voice patterns by tracking the motion of vocal chords. This is achieved by machine learning. A large amount of database is constructed where it is known if one has the disease or not, then the database is trained to filter true symptoms of the disease from other factors.
Mr. Little believes the system is smart enough to differentiate voice patterns caused by common cold from the ones effected by Parkinson's. He aims to keep adding to the database to maintain a vast voice bank in order to facilitate the database to learn better.
Source & Image Credit: BBC