Developing Algorithms for Severity Prediction in Mental Health Management Applications

Developing Algorithms for Severity Prediction in Mental Health Management Applications - Developing Algorithms for Severity Prediction in Mental Health Management Applications

Status

Recruiting

Phases

Unknown

Study type

Observational

Source

JPRN

Registry ID

JPRN-UMIN000059442

Enrollment

200

Registered

2025-10-17

Start date

2025-10-01

Completion date

Unknown

Last updated

2026-06-29

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

depression

Interventions

None listed

Eligibility

Sex/Gender

All

Inclusion criteria

Inclusion criteria: 1. Individuals aged between 18 and 65 years at the start of data collection 2. No restriction on sex

Exclusion criteria

Exclusion criteria: 1. Difficulty in responding to questionnaires in Japanese 2. Not owning a smartphone capable of running the applications required for data collection 3. History of arrhythmia 4. Currently undergoing treatment with a pacemaker or antiarrhythmic medication

Design outcomes

Primary

Measure	Time frame
Mental health problems in the workplace, from mild to severe, negatively affect individuals and society. Yet, tools that allow daily monitoring, early detection, and timely intervention remain insufficient. Because the invisible state of mental health lacks clear criteria for visualization, there is a pressing need to establish practical and sustainable indicators and to develop severity prediction algorithms that make use of them.	—

Measure

Time frame

Mental health problems in the workplace, from mild to severe, negatively affect individuals and society. Yet, tools that allow daily monitoring, early detection, and timely intervention remain insufficient. Because the invisible state of mental health lacks clear criteria for visualization, there is a pressing need to establish practical and sustainable indicators and to develop severity prediction algorithms that make use of them.

—

Secondary

Measure	Time frame
We will examine whether the severity of mental health states, classified as mild, moderate, or severe based on psychological measures related to current and future sleep, anxiety, and depression, can be predicted using machine learning algorithms applied to other simple indicators. We will evaluate implementation and adherence rates of each application and questionnaire to identify tools that are feasible for daily use. We will also assess which background factors strongly influence the primary algorithmic outcomes, thereby evaluating which approaches are most effective. Furthermore, we will investigate whether there are correlations between daily questionnaires, heart rate variability, voice measurements, and weekly follow-up results, as well as their temporal fluctuations. In addition, we will explore whether combining other indicators can predict both the "current state" and "subsequent changes" in mental health.	—

Measure

Time frame

We will examine whether the severity of mental health states, classified as mild, moderate, or severe based on psychological measures related to current and future sleep, anxiety, and depression, can be predicted using machine learning algorithms applied to other simple indicators. We will evaluate implementation and adherence rates of each application and questionnaire to identify tools that are feasible for daily use. We will also assess which background factors strongly influence the primary algorithmic outcomes, thereby evaluating which approaches are most effective. Furthermore, we will investigate whether there are correlations between daily questionnaires, heart rate variability, voice measurements, and weekly follow-up results, as well as their temporal fluctuations. In addition, we will explore whether combining other indicators can predict both the "current state" and "subsequent changes" in mental health.

—

Countries

Japan

Contacts

Public ContactMizuki Ohashi

Yokohama City University Research and Industry-Academia Collaboration Division

mizukion@belle.shiga-med.ac.jp05035757535

Outcome results