Telemonitoring and AI for Mental Health Assessment

Introduction

Mental health disorders such as anxiety and depression are increasingly recognized as significant public health challenges. Leveraging advanced technologies, including artificial intelligence (AI) and telemonitoring devices, offers a promising path toward improving diagnosis and treatment. This validation study compares the performance of two innovative tools: the FACS system from DYAGNOSYS and the CardioWatch 287 bracelet from CORSANO. Both aim to enhance mental health assessments by providing accurate, reliable, and non-invasive diagnostics.

Objectives

General Objective

To determine the accuracy and reliability of the FACS system and CardioWatch 287 bracelet in diagnosing anxiety and depression, using validated questionnaires as a benchmark.

Specific Objectives

Accuracy and Reliability: Compare the diagnostic accuracy of the FACS system and CardioWatch with validated screening tools.
Clinical Integration: Assess the feasibility of incorporating these devices into clinical practice, considering factors such as cost, ease of use, and required training.
Participant Acceptance: Evaluate the comfort and usability of the devices for participants.

Research Methodology

1. Study Design

The study involves 240 medical students recruited from the UNIVÉRTIX University Center, representing a population of 636 students. The sample size was determined using the OpenEpi® statistical tool, with a 95% confidence interval and a 5% margin of error.

2. Recruitment and Criteria

Inclusion Criteria

Students aged 18 or older
Regularly enrolled in the university

Exclusion Criteria

Students younger than 18
Students experiencing significant stress within the week prior to assessment

Participants were invited via email and required to provide formal consent through a signed Informed Consent Form (ICF).

3. Screening Questionnaires

Anxiety Assessment Tools

GAD-7 (Generalized Anxiety Disorder Scale): Evaluates generalized anxiety with a scoring range of 0-21.
BAI (Beck Anxiety Inventory): Measures physical and cognitive symptoms of anxiety, with scores ranging from 0-63.

Depression Assessment Tools

PHQ-9 (Patient Health Questionnaire-9): Assesses symptoms of major depression on a 0-27 scale.
BDI (Beck Depression Inventory): Evaluates depression severity across a 0-63 scoring range.
HAM-D (Hamilton Depression Rating Scale): Requires trained professionals for evaluating depression severity.

Combined Tool

DASS-21: Measures depression, anxiety, and stress using a self-reported 21-item questionnaire.

4. FACS System: AI-Driven Facial Expression Analysis

The FACS system uses the Facial Action Coding System to analyze emotional states through video-based facial expressions. It provides standardized scores for stress, anxiety, and depression by quantifying specific Action Units (AUs).

5. CardioWatch 287: Continuous Monitoring

The CardioWatch 287 bracelet tracks various physiological parameters:

Heart Rate
Oxygen Saturation
Respiratory Rate
Body Temperature
Activity and Sleep Patterns

These measurements are captured at variable frequencies, enabling high-resolution data collection for mental health assessments.

6. Additional Assessments: Communication Analysis

The study evaluates non-verbal communication parameters to assess emotional and mental states:

Posture
Gestures
Eye contact
Breathing patterns
Vocal characteristics

7. Data Analysis

Participants are divided into two groups:

Group 1: Diagnosed with anxiety or depression based on validated questionnaires.
Group 2: No diagnosis of anxiety or depression.

The diagnostic outcomes from the FACS system and CardioWatch 287 are cross-referenced with these groups to evaluate accuracy and reliability.

Anticipated Outcomes

The study aims to establish a strong correlation between the diagnostic data from the FACS system and CardioWatch 287 with results from validated questionnaires. Successful results will demonstrate the devices’ potential for integration into clinical practice, improving accessibility and reliability in mental health diagnostics.

Full article

Telemonitoring and ai for Mental health assessment