Research finds way to detect Parkinson's using Siri-like software

The early diagnosis of the symptoms of the Parkinson's disease, a degenerative disorder of the central nervous system has continued to be difficult to date. However, there may be a solution now. According to a report by the Gizmodo, Max Little, from the University of Oxford has been developing a software, which can gauge the differences in voice patterns. This, he believes will help in effectively detecting symptoms associated to Parkinson's disease. A post in Parkinson's Voice Initiative aptly points out that an interesting "breakthrough" that has been achieved is that it has been known now that 'voice is affected as much by Parkinson's as limb movements.' Hence having a software that can gauge the differences in voice patterns is the next obvious step in this direction. The report quoted Little, as explaining to BBC, "This is machine learning. We are collecting a large amount of data when we know if someone has the disease or not and we train the database to learn how to separate out the true symptoms of the disease from other factors."

First steps towards diagnosis of Parkinson's

In his research, Little used data from 50 patients suffering from the disease. These patients had their voices recorded once a week for six months. Using this data, Little managed to 'develop an algorithm to detect changes in voice purely associated with Parkinson's.' In tests carried out recently, it was found that the software managed to accurately tell a Parkison's patient from a random population with an accuracy measuring 86 percent. 

Little, however has bigger plans. At the TEDGlobal, he announced an extension to the project and also invited the general public to come forth, phone in and leave their voice recordings, in a bid to help him improve the software. He aims to collect some 10,000 voices and has encouraged people from around the world to contribute. Interestingly, if successful, Little may make this technology available to doctors in two years, believing that the software will effectively aid doctors in the diagnosis of the disease. Little further in his explanation to BBC stated, "We're not intending this to be a replacement for clinical experts, rather, it can very cheaply help identify people who might be at high risk of having the disease and for those with the disease, it can augment treatment decisions by providing data about how symptoms are changing in-between check-ups with the neurologist."