Project in Singapore to build voice AI for detecting depression in young elderly people
A multi-sector partnership involving National Healthcare Group and Nanyang Technological University is building voice AI and a community intervention program to detect early onset depression among the elderly.
The three-year project also involves the NHG Polyclinics and Institute of Mental Health (IMH) and two NTU Singapore schools: Lee Kong Chian School of Medicine (LKCMedicine) and College of Computing and Data Science (CCDS).
WHAT IT’S ABOUT
The project – called SoundKeepers – will involve more than 600 seniors aged 55 and over associated with the Hougang and Woodlands outpatient clinics.
Voice recordings of the participants from casual conversation or reading passages are collected with permission and identified as a representative sample for the AI model. The voice samples are anonymized and stored in a central storage terminal.
The AI, to be Jointly developed by LKCMedicine and CCDS of NTU Singapore, it will then analyze these sample recordings for their acoustic properties, including pitch, volume, timbre, rhythm, gloss, jitter and harmonics-to-noise ratio.
“When we use our voice, we activate and coordinate more than a hundred different muscles and neurobiological processes. A change in speech acoustic characteristics can reveal abnormalities in these neurobiological processes,” explains Dr. Lee Eng Sing, co-principal investigator of the project, said. He is also an assistant professor and clinical director of the Primary Care and Family Medicine research program at NTU LKCMedicine.
Meanwhile, a second component of the project is a 24-week community intervention program led by the IMH. It includes module-based psychoeducation, community activities (such as exercise and recreational activities) and friendship building. The program is implemented with the help of social service providers, Fei Yue Community Services and Club HEAL.
In addition, a randomized controlled trial – led by the NHG outpatient clinics – will recruit 300 participants, including those observed to have early onset depression, to compare their levels of loneliness, anxiety, well-being and depression literacy before and after the intervention program . These will also be compared with a control group.
This cluster-wide project has received S$5.6 million ($4.2 million) in funding from the Lien Foundation.
WHY IT’S IMPORTANT
The NHG-based project focuses on the condition called subsyndromal depression (SSD), an emerging form of depression seen among seniors. The project partners consider the condition to be a ‘largely untreated health risk’.
In a statement, they claim that SSD is five times more likely to become depression within a year; People with this condition are reportedly 12 times more likely to develop dementia.
There are five times more people with SSD than those who have depression in a year, they noted.
Based on self-reported assessments, the condition is also said to affect approximately 13% of seniors living in community homes – which may be an underestimate.
Furthermore, it was observed that seniors with early depression incur more health care costs and make more use of outpatient services than seniors without this care.
“Currently, SSD is not actively diagnosed or treated,” admits Dr. Mythily Subramaniamad, also co-principal investigator of the SoundKeepers project. She is also assistant chair of the Research Medical Board at IMH.
Their project, Dr. Subramaniamad added, will “facilitate the early detection and diagnosis of SSD with a tool that can be easily used in the community.”
“(A)s while previous studies have shown the feasibility of detecting depression through vocal analysis, little work has been done on the more subtle SSD,” said NTU CDDS professor Guan Cuntai, another co-principal investigator on the project .
Once SoundKeepers has proven effective, it is hoped to expand this to more outpatient clinics, GP clinics and participant cohorts. Project partners also see the Voice AI-driven project potentially becoming part of Singapore’s national mental health ecosystem through existing government-funded community programs such as CREST and COMIT.
THE BIG TREND
Another social service provider in Singapore, Lion Befrienders, is also using AI and mobile technologies to monitor and improve the mental conditions of seniors living at home.
In 2021 it started a two-year period pilot with software company Opsis to monitor the mental state of more than 4,000 seniors by recording and analyzing their facial expressions.
Last year the Singapore University of Technology and Design collaborated with Lion Befrienders launch a mobile application to prevent dementia in seniors through multilingual instructions. The app, called Ami, is installed on more than a thousand digital tablets used in one of the service’s active aging centers.
ON THE RECORD
“We need new ways to listen to our seniors. Although they may not express their concerns in words, we can now try to hear it through their voices,” Lee Poh Wah, CEO of the Lien Foundation, said in a statement.