⬡ Hacklythics 2026 — Mental Health Track

Depression
screening that
never sleeps.

MoodLens passively monitors biometric and linguistic signals from your wearable and phone to detect early signs of depression — before crisis strikes.

970M
People with mental disorders worldwide
43%
Of those affected who receive treatment
$6T
Projected annual economic cost by 2030
0
Active effort required from the user
// 01 — The Problem

Mental healthcare is reactive.
We're making it proactive.

Depression leaves digital fingerprints weeks before a crisis. Today, those signals go undetected.

23.4%

Adults affected yearly

Over 60 million Americans experience a mental illness annually, yet most won't see a mental health professional — held back by stigma, cost, and access. (CDC, 2024)

40%

Adolescents with no care

Four in ten adolescents who experience a major depressive episode receive zero treatment, leaving them vulnerable to worsening outcomes. (Annie E. Casey Foundation, 2024)

💡

The detection gap

Psychiatric care happens in scheduled appointments. Crises don't. There's almost no infrastructure to catch someone in the weeks before they reach a breaking point.

📡

Data we already have

Wearables and smartphones continuously collect sleep patterns, heart rate variability, activity levels, and communication behavior — all known early indicators of depressive episodes.


Two streams. One score.
Calibrated action.

MoodLens fuses biometric and linguistic signals into a PHQ-9-aligned severity score, then responds proportionally.

STEP 01

Wearable Data

Sleep, HRV, steps, SpO2 via Google Health Connect

STEP 02
🧠

Biometric Model

Logistic regression on Databricks → PHQ Score 1

STEP 05
⚖️

Score Fusion

Weighted combination → Final PHQ-9 Score

STEP 06
🎯

Intervention

Tiered response from affirmations to human responders

STEP 03
💬

Message Analysis

Call log metadata & anonymized text patterns

STEP 04
🤖

RoBERTa Model

Fine-tuned sentiment analysis → PHQ Score 2

PHQ 0 – 9
Mild

Positive Affirmations

No active intervention is needed. The app delivers personalized affirmations on demand, proactively supporting resilience and emotional well-being.

PHQ 10 – 19
Moderate

AI Conversation

The user is connected to an empathetic AI voice conversation powered by a large language model and ElevenLabs text-to-speech — a stigma-free, always-available bridge to support.

PHQ 20 – 27
Severe

Human-in-the-Loop

Emergency contacts, volunteer mental health professionals, or medical responders are alerted and provided with PHQ trend data and flagged linguistic signals for rapid triage.

// 03 — Technology

Built on rigorous
research foundations.

Every model and methodology is grounded in peer-reviewed clinical literature.

Biometric Intelligence

Synthetic Data + Spearman Correlation

Because ethically labeled biometric-PHQ datasets are scarce, we generated synthetic training data using Spearman correlation coefficients from:

Rykov et al. "Digital Biomarkers for Depression Screening With Wearable Devices." JMIR Mhealth Uhealth, 2021. PMC8576601 ↗
Linguistic Intelligence

Fine-tuned RoBERTa

State-of-the-art transformer model fine-tuned on depression-related linguistic patterns. RoBERTa achieves 94–96% accuracy on sentiment benchmarks through dynamic masking, 160 GB training data, and no NSP objective.

Liu et al. "RoBERTa: A Robustly Optimized BERT Pretraining Approach." arXiv:1907.11692, 2019. arXiv ↗
ML Infrastructure

Databricks Inference

Both models are hosted as inference endpoints on Databricks, providing scalable, low-latency predictions from the mobile app with enterprise-grade reliability.

Health Data

Google Health Connect

Integrates with Google Health Connect to pull passive biometric data from Pixel Watch, Fitbit, and other supported wearables — no manual logging required.

AI Conversation

LLM + ElevenLabs Voice

The moderate-tier intervention uses a large language model paired with ElevenLabs text-to-speech synthesis to deliver empathetic, naturalistic voice-based support conversations.

Frontend

React + Vite

Mobile-first React application built with Vite for fast builds and HMR. Privacy-first architecture: message content is analyzed locally or anonymized, never stored in raw form.


The people behind
MoodLens.

A multidisciplinary team united by a belief that technology can close the gap between depression and care.

A

Ankit Bansal

Biometrics and Health Connect

Architected the biometric model pipeline, synthetic data generation workflow.

T

Shruti Murarka

NLP / RoBERTa

Databricks model deployement. Agentic AI development.

T

Swebert Correa

App Development

Built the React mobile frontend and Health Connect integration for passive biometric data collection from wearables.

// 05 — Contact

Interested in
MoodLens?

Whether you're a researcher, clinician, investor, or potential collaborator — we'd love to hear from you.

Let's build better mental healthcare together.

MoodLens is an open-source Hacklythics project exploring the frontier of passive mental health monitoring. We're looking for research partners, clinical advisors, and anyone passionate about making mental healthcare more proactive.

Send us a message

Message sent!

Thanks for reaching out. We'll get back to you within 48 hours.