HMT Newsletter Sign Up

 

 

 

 

 

 

 


 

 

 

 

Click to See More Information about RSNA 2014

Health Management Technology White Paper Library

 

 Viewpoint

Utilizing voice recognition as diagnostician

An impressive example of how existing technology heretofore used primarily to smooth data entry into patient records has been updated and recreated into a powerful diagnostic tool.

Email this article to a friend
  

   By Phil Colpas, Managing Editor, September 2012

The Parkinson’s Voice Initiative is lending its ear, and it’s looking for a few good voices. Make that several thousand good voices.

In what was clearly some inspired outside-the-box thinking, PVI Project Director Max Little, Ph.D, and his team discovered it was possible to determine if someone has Parkinson’s disease by simply analyzing voice recordings through the telephone, using voice-recognition technology to formulate an automated test that takes as little as 30 seconds.

Why is this important to HMT readers? Because it’s an impressive example of how existing technology – heretofore used primarily to smooth data entry into patient records – has been updated and recreated into a powerful diagnostic tool.

In order to increase its data pool and make its already accurate analyzer even more precise, PVI aims to collect 10,000 recordings – via telephone – from people across the globe.

An estimated 6.3 million people worldwide suffer with Parkinson’s, a degenerative disorder of the central nervous system. The motor symptoms result from the death of dopamine-generating cells in the substantia nigra, a region of the midbrain. The cause is unknown and there is no cure; but the earlier the diagnosis, the more effective the treatment.

Parkinson’s affects the limbs with incurable weakness, tremor and rigidity – and it affects the vocal cords in much the same way. Because the disease has no discernible biomarkers, no simple blood test can determine its presence. Current tests for Parkinson’s are expensive, time consuming and logistically difficult, often resulting in delayed diagnoses – so PVI developed the technology to test for symptoms using voice recordings alone.

Little says PVI’s high-speed, low-cost, voice-based tests are as accurate as clinical tests, but are administered remotely – and patients can do the tests themselves.

The initiative is asking people to call and lend their voices to the study (phone numbers for nine countries are listed at www.ParkinsonsVoice.org). Calls take an average of three minutes to complete, and all adults can participate (both healthy people and Parkinson’s patients are needed).

According to Little, the goals of PVI are fourfold:
•    Reduce logistical difficulties in routine practice – no need to visit the clinic for checkups.
•    High-frequency monitoring for individualized treatment decisions.
•    Cost-effective mass recruitment for treatment trials. Recruiting very large numbers into trials for new treatments will speed up the search for a cure.
•    Population-scale screening programs could find signs of the disease before the damage is irreparable.

PVI’s commercial partners include Aculab Cloud, a cloud telephony platform; and PatientsLikeMe, a data-driven social networking health site.

Little announced the launch of PVI in June at a TED Talks speech he delivered in Edinburgh, Scotland. During the event, Little revealed PVI had already reached 6 percent of its 10,000-voice target – in just eight hours.


Tags:  Viewpoint