Acquisition and Analysis of Relationships Between Longitudinal Emotional Signals Produced by an Artificial Intelligence Algorithm and Self-questionnaires Used in the Psychiatric Follow-up of Patients With Mood and/or Anxiety Disorders: a Real-Environment Study.

Major Depressive Disorder Clinical Trial

— EMOACQ-1  

Official title:

Acquisition and Analysis of Relationships Between Longitudinal Emotional Signals Produced by an Artificial Intelligence Algorithm and Self-questionnaires Used in the Psychiatric Follow-up of Patients With Mood and/or Anxiety Disorders: a Real-Environment Study.

Clinical Trial Details — Status: Not yet recruiting

Administrative data

• NCT number: NCT05988840
• Other study ID #: 2023-A01589-36
• Status: Not yet recruiting
• Start date: October 17, 2023
• Est. completion date: October 17, 2024

Study information

• Verified date: August 2023
• Source: Emobot
• Contact: Tanel Petelot
• Phone: +33 51 44 26 67
• Email: tanel.petelot[@]emobot.fr
• Is FDA regulated: No
• Study type: Observational

Clinical Trial Summary

The worldwide prevalence of anxiety and depression increased massively during the pandemic, with a 25% rise in the number of patients suffering from psychological distress. Psychiatrists, and even more so general practitioners, need measurement tools that enable them to remotely monitor their patients' psychological state of health, and to be automatically alerted in the event of a break in behavior.

In this study, the investigators propose to collect clinical data along with longitudinal measurement of patients' emotions. Emobot proposes to analyze the evolution of mood disorders over time by passively studying people's emotional behavior. The aim of EMOACQ-1 is to acquire knowledge and produce a quantitative link between emotional expression and mood disorders, ultimately facilitating the understanding and management of these disorders.

Through this study, could be developed a technological solution to support healthcare professionals and patients in psychiatry, a field known as the "poor relation of medicine" and lacking in resources. Such a solution would enable better understanding, disorders remote & continuous monitoring and, ultimately, better treatment of these disorders.

The investigators will process the data by carrying out a number of analyses, including descriptive, comparative and correlation studies of the data from the self-questionnaire results and the emotional signals captured by the devices.

Finally, the aim will be to predict questionnaire scores from the emotional signals produced.

Recruitment information / eligibility

• Status: Not yet recruiting
• Enrollment: 50
• Est. completion date: October 17, 2024
• Est. primary completion date: August 17, 2024
• Accepts healthy volunteers: Accepts Healthy Volunteers
• Gender: All
• Age group: 18 Years and older

Eligibility

Inclusion Criteria:

• Persons over the age of 18 who volunteer to take part in research

• Must have access to a computer with an Internet connection,

• Written comprehension of French.

Exclusion Criteria:

• N/A

Related Conditions & MeSH terms

• Anxiety Disorders
• Depressive Disorder
• Depressive Disorder, Major
• Major Depressive Disorder

Intervention

Other:
• Acquisition and analysis of relationships between Longitudinal Emotional Signals produced by a software and Self-questionnaires.
Using the tool developed by Emobot, EMOACQ-1 is a study that passively and non-interventionaly collects data by capturing patients' facial expressions throughout the day, and then measures the correlation between emotional signals and the results of measurement questionnaires used in psychiatry.

Locations

Country	Name	City	State
n/a

Sponsors (1)

Lead Sponsor	Collaborator
Emobot

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	Repeated measurements Correlations between emotional signals and studied disorders standardized tests.	Repeated measurements Correlations between emotional signals and studied disorders standardized tests.	10 months

External Links

•   D. Agarwal et al. From Multimodal to Unimodal Attention in Transformers using Knowledge Distillation. Nov 2021, Virtual, United States.
•   E. Garcia-Ceja et al. Depression: A Motor Activity Database of Depression Episodes in Unipolar and Bipolar Patients
•   J. Gratch et al. The Distress Analysis Interview Corpus of Human and Computer Interviews.
•   Mamadou Dia et al. A Novel Stochastic Transformer-based Approach for Post-Traumatic Stress Disorder Detection using Audio Recording of Clinical Interviews, CBMS, June 2023.
•   Mundt JC, Vogel AP, Feltner DE, Lenderking WR. Vocal acoustic biomarkers of depression severity and treatment response.
•   X. Kong et al. Automatic Identification of Depression Using Facial Images with Deep Convolutional Neural Network.