SensaBand

SensaBand — Project Story

Inspiration

SensaBand was inspired by observing how people with Autism, ADHD, and Sensory Processing Disorders are frequently blindsided by sensory overload from noise, bright lights, or crowded spaces. Most tools today are reactive—displaying data after the fact—so I wanted a system that predicts rising stress and warns users early, preserving independence, dignity, and safety.

What I learned

• Multi-sensor fusion is powerful: combining heart rate, HRV, EDA, ambient noise, and light gives a fuller picture of stress than any single signal.
  
• Personalization matters: thresholds vary day-to-day and person-to-person, so adaptive models are essential.
  
• Wearable constraints: low-power sensing, Bluetooth reliability, and battery trade-offs greatly influence design decisions.
  
• Privacy-first design: handling health data requires transparent opt‑in flows and local processing where possible.

How I built the project

• Hardware concept: small wearable with HR/SpO₂, EDA, IMU, light, and microphone sensors; Bluetooth for mobile connectivity and optional LTE for remote alerts.
  
• Mobile app: ingests sensor streams, computes risk levels (green/yellow/red), and provides interventions (breathing guides, screen dimming, grounding exercises, SOS).
  
• Optional clinician dashboard: aggregates anonymized trends and allows clinicians to monitor patterns and adjust thresholds.
  
• Architecture emphasizes local-first processing for privacy, with cloud-based analysis optional and strictly opt-in.

High-level flow:

1. Sensor sampling → 2. Local filtering & feature extraction → 3. Risk scoring & personalization → 4. User alert + intervention suggestions

Challenges faced

• Predicting rather than reacting: overload precursors are subtle and variable; models needed continual personalization and online adaptation.
  
• Privacy vs. utility: designing opt-in data flows, encryption, and minimal data retention while preserving clinical usefulness.
  
• Sensor noise and motion artifacts: required robust filtering and quality checks to avoid false alarms.
  
• Power and usability: balancing sampling rates and battery life while keeping the device comfortable.
  
• User-centered design: alerts had to be discreet, calming, and non-intrusive for neurodivergent users.

Final reflection

SensaBand started with a single question: “What if we could warn someone before sensory overload?” Through iterative design, sensor fusion, and attention to privacy and user experience, it evolved into a practical system that blends wearable tech, adaptive models, and empathetic design. The project reinforced that thoughtful technology can restore confidence and autonomy for neurodivergent individuals.

Built With

• na