Voice-Based Technology to Detect Mental, Physical Disorders Receives US, Australian Patents

Marta Figueiredo, PhD avatar

by Marta Figueiredo, PhD |

Share this article:

Share article via email
vocal biomarker technology

Sonde Health’s vocal biomarker technology being developed to potentially diagnose and monitor mental and physical disorders, including Parkinson’s disease, has recently been issued patents in the U.S. and Australia.

Patent No. 9,936,914, issued by the U.S. Patent and Trademark Office, provides coverage until 2032, and patent No. 2014374349, issued by Intellectual Property Australia, covers the technology until 2034. Both patents are focused on vocal biomarkers that can be used in the screening and monitoring of physical and psychological disorders.

“These newly issued patents represent, to the best of our knowledge, the only coverage for unique classes of vocal biomarkers that demonstrate enhanced vocal analysis performance and accurate assessment of a broad range of physical and psychological conditions,” Jim Harper, PhD, Sonde’s chief operating officer, said in a press release.

The technology works by detecting and analyzing subtle vocal changes to create a range of brain, muscle, and respiratory health measurements that provide a more complete picture of health in just seconds. It is being designed to work through any voice-enabled device, including smartphones.

Boston-based Sonde Health — a digital medicine company recently launched by PureTech Health — licensed the health-focused voice scanning technology from the Massachusetts Institute of Technology (MIT) Lincoln Laboratory in June 2016.

Pilot studies demonstrated the technology’s potential to detect and objectively measure symptoms in a range of conditions including Parkinson’s disease, depression, cognitive impairment, mild traumatic brain injury, and concussion.

To date, it has been tested in more than 3,000 individuals for the detection of neurophysiological disorders as part of the ongoing validation of its platform.

Speech is one of the most complex human motor skills, requiring incredible coordination of multiple brain circuits, muscles, and respiratory function.

It is highly sensitive to disturbances in the basal ganglia — an area of the brain involved in the planning and execution of motor tasks — which is one of the most affected brain regions in Parkinson’s disease.

As a result, speech changes are among the most robust motor abnormalities in Parkinson’s disease. Between 60 and 80 percent of patients may experience abnormal speaking rates, reduced vocal loudness and pitch, monotone speech, and harsh or breathy vocal quality.

Doctors have been using changes in speech to help diagnose depression as early as the 1950s. Increasing evidence suggests speech alterations are also associated with several other mental and physical disorders.

A recent study showed that a speech analysis technique was able to identify Parkinson’s patients — with more than 80 percent accuracy — through speech and vocal patterns even at early stages of the disease.

Identifying vocal biomarkers of diseases has the advantage of being an inexpensive, noninvasive, and simple-to-administer method, since it can be performed remotely through a person’s smartphone.

In Parkinson’s, vocal biomarker technology could provide valuable information not only related to diagnosis and treatment, but also in the setting of clinical trials and disease-modifying therapies before the onset of motor symptoms.

This type of technology can potentially change the way mental and physical health is monitored, managed, and diagnosed.