Output details

Accurate Telemonitoring of Parkinson's Disease Progression by Noninvasive Speech Tests

For the 6.3m people with Parkinson’s disease (PD), telemedicine offers the possibility of remote monitoring with substantial opportunities for lowering the inconvenience and cost of physical visits. We developed an innovative telemedicine system based on speech recordings, using new robust measures of dysphonia.  Using the largest PD database in existence (~6000 recordings), we verified the feasibility of machine learning techniques for remote replication of clinical scores.  This inspired an initiative to crowdsource data using 10,000 telephone calls (http://www.parkinsonsvoice.org), a TED talk and substantial media coverage (http://www.bbc.co.uk/news/technology-18427851). This research resulted from an EPSRC Case Studentship sponsored by Intel.