Abstract
In a world where accessing mental health and wellbeing services remains challenging due to resource constraints and societal stigmas, the pandemic has intensified the need for innovative access methods. Traditional face-to-face service avenues have become strained, highlighting the importance of digital interventions in mental health care. This research details the journey towards use of app-based voice biomarker assessments for those seeking mental health support in increasingly digital landscapes.
The study, spanning 12 weeks, investigated the effectiveness of using voice biomarkers for mental health assessment among young adults aged 16-24. An app-based platform facilitated weekly self-assessments at the participants' discretion and incorporated machine learning to analyse voice data for signs of mental health struggles.
Participants engaged in weekly sessions, prompted by a question about their day to capture voice samples, with machine learning algorithms providing the analytical backbone. These were assessed alongside established mental health questionnaires to identify markers of depression and anxiety. This blend of technology and traditional assessments aimed to uncover insights into the participants' mental health.
Our analysis reveals that while demographic factors marginally influence model accuracy, the severity of mental health conditions significantly impacts detection capabilities. This nuanced understanding underscores the potential of digital interventions in providing more accessible, personalised mental health support.
In the broader context of the post-pandemic push towards digital mental health solutions, our findings contribute knowledge towards technological innovations aligned with the goal of services that are not only effective, but also accessible. The research shows the potential and challenges associated with voice biomarker technology in revolutionising mental health assessment and enhancing care through digital means.
The study, spanning 12 weeks, investigated the effectiveness of using voice biomarkers for mental health assessment among young adults aged 16-24. An app-based platform facilitated weekly self-assessments at the participants' discretion and incorporated machine learning to analyse voice data for signs of mental health struggles.
Participants engaged in weekly sessions, prompted by a question about their day to capture voice samples, with machine learning algorithms providing the analytical backbone. These were assessed alongside established mental health questionnaires to identify markers of depression and anxiety. This blend of technology and traditional assessments aimed to uncover insights into the participants' mental health.
Our analysis reveals that while demographic factors marginally influence model accuracy, the severity of mental health conditions significantly impacts detection capabilities. This nuanced understanding underscores the potential of digital interventions in providing more accessible, personalised mental health support.
In the broader context of the post-pandemic push towards digital mental health solutions, our findings contribute knowledge towards technological innovations aligned with the goal of services that are not only effective, but also accessible. The research shows the potential and challenges associated with voice biomarker technology in revolutionising mental health assessment and enhancing care through digital means.
| Original language | English |
|---|---|
| Pages | 1 |
| Number of pages | 1 |
| Publication status | Published (in print/issue) - 10 Sept 2024 |
| Event | European Conference on Mental Health - Kraków, Poland Duration: 9 Sept 2024 → 11 Sept 2024 Conference number: 12 https://ecmh.eu/ |
Conference
| Conference | European Conference on Mental Health |
|---|---|
| Abbreviated title | ECMH |
| Country/Territory | Poland |
| City | Kraków |
| Period | 9/09/24 → 11/09/24 |
| Internet address |
