DITESMU: Early Diagnosis of Depressive Symptoms Using Multimodal Statistical Machine Learning

At LabLENI, we are leading a pioneering research project: DITESMU (Early Diagnosis of Depressive Symptoms Using Multimodal Statistical Machine Learning). This project is funded by the Generalitat Valenciana (CIGE/2023/160) for emerging research groups.
Objectives
The main goal of DITESMU is to develop biomarkers using advanced machine learning techniques applied to physiological signals. These signals are collected during simulated conversations with realistic virtual humans in a laboratory setting.
Why is DITESMU unique?
This project is the first of its kind to model depressive symptom patterns at the physiological level during open social interactions. By utilizing fully automated virtual humans, we aim to understand how depressive symptoms influence social behavior and develop new methods for early detection.
Preliminary Results
Initial analyses of signals, including electrocardiograms, electroencephalograms (EEG), electrodermal activity, and eye tracking, suggest that EEG is a key signal for detecting depressive symptoms in social contexts.
Our team is preparing to publish these exciting findings in an upcoming scientific article—stay tuned for updates!
Visual Insights
Explore the images below to see one of the virtual humans used in our research and gain a glimpse into our innovative approach.
This project reinforces our commitment to advancing mental health research using cutting-edge technology.
For more information, feel free to contact us or follow our updates on social media.
#MentalHealth #LabLENI #HumanTechUPV #MachineLearning #VirtualHumans #ResearchInnovation