Clinical evidence

Where’s the proof?
Here it is.

11 clinical studies across France, the US, Canada, and Germany. Pooled validity data presented at ASCP 2026 as a late-breaking poster, co-authored with NRx Pharmaceuticals (Nasdaq: NRXP), CentraleSupélec and Johns Hopkins. Academic partners on three continents. This is what clinical-grade looks like.

r=0.895

MADRS rmCorr

Clinical gold standard

ρ=0.834

PHQ-9 Δ sensitivity

Sensitivity to change

Clinical studies

Across 4 countries

Academic institutions

Yale → Charité

Pooled validity · 3 studies · n = 45 · ASCP 2026

Emobot fills the gap between clinical assessments.

In outpatient care, PHQ-9 is administered every 2 weeks. Emobot’s continuous signal shows the day-to-day variation between those two points: the nuance a biweekly questionnaire can’t see.

Emobot: continuous signal (passive, daily)PHQ-9: discrete self-report (every 2 weeks)

MADRS rmCorr

r = 0.895

n = 45 · pooled across 3 studies · p = .016

HAM-D17 Spearman

ρ = 0.833

Concordance with clinician rating

PHQ-9 Δ sensitivity

ρ = 0.834

Sensitivity to change, Spearman

Late-breaking poster: Perzo A., Sapko M.T., Petelot T., Séguier R., Javitt J.C. — Continuous Multimodal Passive Monitoring of Depressive Symptoms via Smartphone: Clinical Validation of EMOCARE. ASCP 2026 · American Society of Clinical Psychopharmacology (Emobot × NRx Pharmaceuticals × CentraleSupélec × Johns Hopkins). Read the poster →

What these numbers mean

Near-clinician accuracy.
Continuous. Patient-owned.

The MADRS (Montgomery–Åsberg Depression Rating Scale) is the gold-standard clinician-administered scale for depression severity. A repeated-measures correlation of r = 0.895 between Emobot’s daily multimodal score and MADRS means the platform tracks clinical depression severity with near-clinician accuracy, continuously, passively, without a single survey.

On sensitivity to change (arguably the more useful clinical property), the Emobot score correlates at ρ = 0.834 with PHQ-9 deltas. And on HAM-D17, the classic depression rating scale, concordance reaches ρ = 0.833.

Because the platform is patient-facing, this isn’t a passive surveillance score pushed to a clinic. It’s a daily signal the patient sees in their own app, and when it dips, gently invites them to rebook with you.

MADRS rmCorr (repeated measures)

0.895

HAM-D17 Spearman

0.833

PHQ-9 Δ sensitivity (Spearman)

0.834

PHQ-9 rmCorr (N=19)

0.723

BDI-II Spearman

0.660

Industry context

Most digital health biomarkers land at r = 0.5–0.7 against clinician scales. Emobot’s multimodal fusion (facial, vocal, actigraphy, digital behavior) is what lifts the correlation to near-clinician territory.

Research portfolio · April 2026

11 clinical studies. Real identifiers.

France, USA, Canada, Germany. CE Mark pathway, US FDA clearance pathway, and large-scale real-world deployments. Pooled validation presented as a late-breaking poster at ASCP 2026 with NRx Pharmaceuticals and Johns Hopkins. Findings published across medical journals (JMIR Mental Health) and AI / biomedical-engineering venues.

EMC1Multi-site clinical validation

Multimodal severity-score validation (France)

ID: NCT06601140

PI: Pr Raymund Schwan

Site: CHU Nantes & CHU Rennes · France

Status: 80% complete

Multi-site French clinical validation of the Emobot multimodal score against clinician-rated scales. Results have fed peer-reviewed outputs in both medical journals (JMIR Mental Health) and AI / biomedical-engineering venues.

EMC2FRClinical validation

Multimodal monitoring in depression (France)

ID: NCT06860165

PI: Dr Clara Roux-Pertus

Site: CHU Nîmes · France

Status: 50% complete

Validation of Emobot's multimodal depression score in a non-interventional psychiatric cohort.

EMC2-BDClinical validation

Multimodal severity tracking (France)

ID: Sponsor: Emobot

PI: Dr Michael Sikorav

Site: France

Status: 50% complete

French clinical validation cohort extending Emobot's multimodal score across mood-state transitions. Shares infrastructure with the EMC1 protocol.

MOODELINGObservational · Sponsor GHU Paris

Large-scale passive phenotyping in mood disorders

ID: ID-RCB 2023-A02264-41

PI: Pr Fabien Vinckier

Site: GHU Paris Psychiatrie & Neurosciences · France

Status: Ongoing · N=600

Flagship 600-patient observational cohort combining Emobot with clinician-rated scales across MDD and bipolar populations.

COGPSYProspective · Cognition

Cognitive markers × multimodal passive monitoring

ID: NCT07295652

PI: Pr Fabien Vinckier

Site: GHU Paris Psychiatrie & Neurosciences · France

Status: Recruiting

Interaction between cognitive markers and Emobot's passive multimodal signals in depression. Shares GHU-Paris infrastructure with the EMOKET esketamine cohort.

EMC1-MDD-USRegulatory · FDA pathway

US FDA clearance study

ID: Sponsor: Emobot

PI: To be announced

Site: USA

Status: Q2 2026 → Q1 2027 · N=92

Clinical study supporting the US FDA clearance pathway for the Emobot multimodal score.

REMOODRCT · Reimbursement

French reimbursement RCT

ID: Sponsor: Emobot

PI: To be announced

Site: France

Status: Q2 2026 → Q1 2028 · N=504

Randomized controlled trial underpinning the French reimbursement dossier (ETT / remboursement).

DeeP-DDObservational

Digital phenotyping in treatment-resistant depression (Canada)

ID: Douglas Mental Health Institute / McGill

PI: Dr David Benrimoh

Site: Montréal · Canada

Status: Active

Canadian validation cohort led from McGill / Douglas, extending Emobot to Canadian TRD populations.

Must-DCommunity health · Large cohort

Multimodal severity tracking in underserved populations

ID: CIAPM · UCSD / FHCSD

PI: Dr Suzi Hong

Site: San Diego · USA

Status: Ongoing · ~N=800

Largest US cohort. Deployment across Family Health Centers of San Diego with UCSD leadership.

US rTMS pilotPragmatic · Multi-site

Multi-site rTMS pragmatic pilot (USA)

ID: Details on request

PI: Clinical leadership · partner network

Site: US interventional-psychiatry clinics

Status: Jan–Jul 2026 · N=45

Pragmatic US deployment of Emobot alongside real-world rTMS care; outcomes captured under clinic conditions.

US Esketamine pilotPragmatic

Esketamine (Spravato) outcomes with passive monitoring

ID: Details on request

PI: Clinical leadership · partner network

Site: US interventional-psychiatry clinics

Status: Jan–Aug 2026 · N=45

Companion pragmatic pilot for Spravato programs: multimodal monitoring between dosing sessions.

Trial integration · announced April 2026

Emobot × EMOKET integration: first passive monitoring trial in Esketamine

ID: NCT06630065

PI: Pr Chantal Henry (+ Pr Fabien Vinckier for COGPSY)

Site: GHU Paris Psychiatrie & Neurosciences · France

Status: Announced April 2026

First trial integrating passive multimodal monitoring into a running Esketamine protocol, combining Pr Henry's EMOKET cohort with Pr Vinckier's COGPSY infrastructure.

Academic partner institutions

New Haven, CT · USA

Boston, MA · USA

Baltimore, MD · USA

San Diego, CA · USA

Montréal, QC · Canada

Berlin · Germany

Paris · France

Scientific publications

Method and clinical validation, in the open literature.

Every headline number above traces back to a peer-reviewed or preprint publication. Below are the most-cited papers behind Emobot’s multimodal score. Full list on the articles page.

Conference proceedingsMay 27, 2026

Continuous Multimodal Passive Monitoring of Depressive Symptoms via Smartphone — Clinical Validation of EMOCARE

Authors: Antony Perzo · Michael Todd Sapko, M.D., Ph.D. · Tanel Petelot · Renaud Séguier · Jonathan C. Javitt, M.D., M.P.H.

Venue: ASCP 2026 · Late-Breaking Poster · Annual Meeting of the American Society of Clinical Psychopharmacology · Emobot × NRx Pharmaceuticals × CentraleSupélec × Johns Hopkins

First public pooled clinical validation of EMOCARE, Emobot's passive multimodal Depression Thermometer. Interim analysis across three prospective observational studies (EMC1, EMC2-FR, EMC2-BD) in n = 45 adults with Major Depressive Disorder or Bipolar Disorder…

Read →

Clinical validationMarch 8, 2026

Clinical Validation of the EMOCARE-Derived Depressive Symptom Severity Score using Established Clinician- and Self-reported Scales: Preliminary Evidence Across 3 Prospective Studies

Authors: Emobot clinical team et al.

Venue: medRxiv preprint · JMIR Mental Health (in review)

Pooled analysis of three prospective studies in adults with mood disorders, comparing Emobot's passively-derived multimodal depression score against the clinician-rated MADRS and patient-reported PHQ-9. Reports moderate-to-strong convergent validity and sensit…

DOI: 10.64898/2026.03.08.26347894v1Read →

Speech emotion AIAugust 19, 2025

EmoSLLM: Parameter-Efficient Adaptation of LLMs for Speech Emotion Recognition

Authors: Hugo Thimonier · Antony Perzo · Renaud Seguier

Venue: arXiv preprint (2508.14130) · CentraleSupélec × Emobot

Proposes a parameter-efficient way to turn a general Large Language Model into a speech emotion recognizer. Audio features are extracted, mapped into the LLM's representation space via a learnable interface, then fused with the transcript and a task prompt. Lo…

arXiv: 2508.14130Read →

See all 7 publications →

Regulatory status

Usable today.
Because it’s a patient-facing app.

Emobot is a patient-facing wellness app: it lives inside the patient’s phone, and the patient optionally shares with their clinician. It’s not a remote monitoring device, which is why a US clinician can deploy it today, today’s active-patient population included. In parallel, we’re running the regulatory pathway in both the EU and the US.

Usable today· April 2026

Patient-facing wellness app. Deployed across US interventional-psychiatry clinics. No device-class gate to start with a patient today.

CE Mark pathway (EU)· Ongoing · EMC1 + EMC2

EU regulatory pathway in progress: 80% complete on EMC1 (Pr Schwan, CHU Nantes/Rennes) and 50% on the EMC2 studies.

US FDA clearance· EMC1-MDD-US · Q2 2026 → Q1 2027

Dedicated US regulatory study (N=92) planned to support the FDA clearance pathway for the multimodal score.

French reimbursement· REMOOD · Q2 2026 → Q1 2028

Randomized controlled trial (N=504) backing the French reimbursement dossier.

4 patents· Filed / granted

Passive multimodal monitoring, on-device neural networks, vocal biomarker extraction, multimodal fusion architecture.

Peer-reviewed· ASCP 2026 · medical + AI venues

Pooled validation of EMOCARE presented as a late-breaking poster at ASCP 2026 (Emobot × NRx Pharmaceuticals × CentraleSupélec × Johns Hopkins). Method & model papers in AI / biomedical-engineering venues.

Convinced by the data?

Let us show you how Emobot performs in a real clinical workflow. 30-minute demo.

Book a Demo Read the FAQ →

Where’s the proof?Here it is.

Emobot fills the gap between clinical assessments.

Near-clinician accuracy.Continuous. Patient-owned.

11 clinical studies. Real identifiers.

Multimodal severity-score validation (France)

Multimodal monitoring in depression (France)

Multimodal severity tracking (France)

Large-scale passive phenotyping in mood disorders

Cognitive markers × multimodal passive monitoring

US FDA clearance study

French reimbursement RCT

Digital phenotyping in treatment-resistant depression (Canada)

Multimodal severity tracking in underserved populations

Multi-site rTMS pragmatic pilot (USA)

Esketamine (Spravato) outcomes with passive monitoring

Emobot × EMOKET integration: first passive monitoring trial in Esketamine

Method and clinical validation, in the open literature.

Continuous Multimodal Passive Monitoring of Depressive Symptoms via Smartphone — Clinical Validation of EMOCARE

Clinical Validation of the EMOCARE-Derived Depressive Symptom Severity Score using Established Clinician- and Self-reported Scales: Preliminary Evidence Across 3 Prospective Studies

EmoSLLM: Parameter-Efficient Adaptation of LLMs for Speech Emotion Recognition

Usable today.Because it’s a patient-facing app.

Convinced by the data?

Where’s the proof?
Here it is.

Near-clinician accuracy.
Continuous. Patient-owned.

Usable today.
Because it’s a patient-facing app.