Multimodal sensing and machine learning to detect mental and social states
Novel Multimodal Neural, Physiological, and Behavioral Sensing and Machine Learning for Mental States
This project will test wearable skin sensors, a conversational virtual human, audiovisual emotion recognition, and machine-learning models to detect mental states in healthy adults and in adults with drug‑resistant epilepsy who already have intracranial EEG implants.
Quick facts
| Phase | Not applicable |
|---|---|
| Study type | Interventional |
| Enrollment | 90 (estimated) |
| Ages | 18 Years and up |
| Sex | All |
| Sponsor | University of Southern California Academic / other |
| Locations | 2 sites (Downey, California and 1 other locations) |
| Trial ID | NCT07110688 on ClinicalTrials.gov |
What this trial studies
The program develops integrated hardware and software: a skin‑like wearable that measures physiological and biochemical signals, a conversational virtual human platform to evoke natural social responses, audiovisual affect recognition software, synchronization tools, and machine‑learning methods to model the combined data. Researchers will record responses during virtual social interactions and controlled challenges such as the cold pressor test in healthy volunteers and in patients who already have clinically implanted intracranial EEG (iEEG) electrodes. The iEEG recordings are from electrodes placed for clinical seizure localization and are unrelated to clinical care changes. The goal is to demonstrate that multimodal signals can be synchronized and modeled to reveal moment‑to‑moment mental and social states.
Who should consider this trial
Good fit: Ideal candidates are adults (≥18) who can give informed consent and either are healthy volunteers or have drug‑resistant epilepsy with already‑implanted intracranial EEG electrodes and can follow study instructions.
Not a fit: People under 18, patients without implanted iEEG, or individuals unable to consent or cooperate are unlikely to be eligible or to receive direct benefit from this protocol.
Why it matters
Potential benefit: If successful, these tools could enable more precise, real‑time detection of emotional and cognitive states and support development of wearable monitoring and brain‑based diagnostics.
How similar studies have performed: Components like affect recognition, wearable physiological sensing, and iEEG state decoding have shown promise in prior work, but combining these multimodal measures with virtual conversational platforms and biochemical skin sensors is relatively novel.
Eligibility criteria
Show full inclusion / exclusion criteria
For the healthy population, all subjects over the age of 18 who are able and willing to give informed consent will be eligible. For the epilepsy population, inclusion criteria are: * Patients who suffer from drug-resistant epilepsy and already have intracranial EEG (iEEG) electrodes implanted based on clinical criteria for their standard seizure localization (unrelated to our study) will be eligible. Most patients are healthy adults outside of their epilepsy. * Subjects \>= 18 are only included in this study. * All patients with the above conditions and with already-implanted electrodes who are willing to participate and able to cooperate and follow research instructions will be recruited.
Where this trial is running
Downey, California and 1 other locations
- Rancho Los Amigos National Rehabilitation Center — Downey, California, United States (Recruiting)
- University of Southern California — Los Angeles, California, United States (Recruiting)
Study contacts
- Principal investigator: Maryam Shanechi, PhD — University of Southern California
How to participate
- Review the eligibility criteria above with your treating physician.
- Visit the official trial page on ClinicalTrials.gov for the most current contact information and recruitment status.
- Contact the listed study coordinator or principal investigator to request pre-screening. Pre-screening is free and never obligates you to enroll.