SMART HEALTH – IoT Based Healthcare Simulator

Step 1: MAX30102 sensor measures patient HR=72 bpm, SpO₂=98%.

Step 2: ESP32 reads sensor via I2C (SDA=GPIO21, SCL=GPIO22).

Step 3: ESP32 connects to Wi-Fi and sends JSON: {"hr":72,"spo2":98,"temp":36.5,"ts":1234567890}

Step 4: Cloud server stores data and updates dashboard every 5 seconds.

Step 5: Doctor receives real-time alert if HR < 50 or > 120 bpm.

Wearable Health Device Architecture

• Sensors: Optical PPG (HR/SpO₂), thermistor (temp), accelerometer (steps/activity)
• MCU: Low-power ARM Cortex-M4 processes sensor data locally
• Bluetooth LE: Transmits data to smartphone app (BLE 5.0, range ~10m)
• Cloud Sync: Smartphone app uploads data to health cloud platform
• Battery: 200–400 mAh Li-Po, 1–7 days depending on usage

• Free-fall detection: |acceleration| < 0.4g for >100ms → free-fall phase
• Impact detection: |acceleration| > 2.5g → impact phase
• Post-fall: Orientation change + no movement for >5s → fall confirmed
• Alert: ESP32 sends SMS/push notification to caregiver within 2 seconds

Machine Learning in Healthcare IoT

• Anomaly Detection: Identifies unusual patterns using Z-score, Isolation Forest algorithms
• Classification: Categorizes health states (Normal/Warning/Critical) using decision trees
• Predictive Analytics: Forecasts health deterioration 2–6 hours in advance using LSTM
• Pattern Recognition: Detects arrhythmia, sleep apnea, fall patterns from sensor data
• Note: All results in this simulator are educational simulations, not clinical diagnoses