Study uses Sonde voice tech to monitor patients at home
- March 21, 2023
- Steve Rogerson

Massachusetts-based Sonde Health and Massachusetts General Hospital (MGH) have been selected for a pilot study to examine the use of voice in remote detection and monitoring of mild cognitive impairment.
The project aims to leverage vocal biomarkers for cognitive assessment of patients aged 55 and above in a home environment.
Sonde Health is a health technology company that aims to bring accessible health monitoring to everyone. It can turn smartphones and other devices into health monitoring tools.
Working with the MGH frontotemporal disorders (FTD) unit, Sonde has been selected by the Massachusetts Artificial Intelligence & Technology Center for Connected Care in Aging & Alzheimer’s Disease (MassAITC) to lead a pilot study focused on leveraging vocal biomarkers for remote detection and monitoring of mild cognitive impairment in the home environment.
Funded by MassAITC and the National Institute on Aging (NIA), a division of the US National Institutes of Health (NIH), the project is part of a $1.7m grant to explore the use of artificial intelligence and other technologies for in-home care. Specifically, it will evaluate the feasibility of obtaining voice recordings of older individuals in the home environment that can be used to monitor speech and memory functions longitudinally.
“Nearly 90% of older adults wish to stay in their homes for as long as possible,” said David Liu, CEO of Sonde Health. “Digital technologies, and digital biomarkers in particular, have great potential to support this shift. By monitoring cognitive health from afar through vocal biomarkers, Sonde’s technology could help facilitate this desire to age in place, offering these patients the ability to remain in a familiar environment without sacrificing quality care.”
The project will enrol 50 adults from the FTD unit. Participants will be aged 55 and above and represent a range of cognitive function from normal cognition through subjective cognitive decline, mild cognitive impairment (MCI) and mild dementia.
The longitudinal study will examine the potential for Sonde’s vocal biomarker platform to monitor and detect changes in cognitive function for individuals at home. Participants will record 10 to 15 different voice samples and responses to cognitive assessments both in the laboratory setting and from their personal smartphones at home.
Voice recordings will be analysed for acoustic features that correlate with mental status, which are expected to provide insights into how vocal biomarkers can be used as an effective tool for tracking cognitive health over time.
The pilot is part of MassAITC’s larger mission to realise the goal of aging at home through interdisciplinary research. It will serve as an important reference data set to test and validate vocal biomarkers relevant to the detection or monitoring of cognitive impairment. This study builds on several current Sonde projects focused on testing cognitive biomarkers, including a multi-year strategic partnership with GN Group to research and develop commercial vocal biomarkers for MCI.
“As the population of older adults continues to grow, there is a critical need for remote monitoring technologies that can detect cognitive impairment early and accurately,” said Deepak Ganesan, director of the MassAITC. “By supporting a study that combines Sonde Health’s vocal biomarker technology and Massachusetts General Hospital’s expertise in clinical dementia research, we believe we’re one step closer to developing the tools and resources needed to elevate aging-in-place care models.”
Brad Dickerson, director of the MGH frontotemporal disorders unit, added: “With this pilot study, we are excited to partner with Sonde Health to contribute new insights into how voice-based remote monitoring of cognitive function for older adults will potentially augment specialised clinical evaluations.”
Leveraging a voice data set with over 1.2 million samples from 85,000 individuals on four continents, Sonde uses audio signal processing, speech science, AI and machine learning to sense and analyse subtle vocal changes due to changes in a person’s physiology to provide insights into health and well-being.