AI-Based ECG and Heart Disease Prediction System

AI-Based ECG & Heart Disease Prediction System Agentic IoT using ESP32 + AI + n8n Automation + Telegram Voice Alerts + Google Sheets + ThingSpeak Cloud Dashboard

AI-Based ECG & Heart Disease Prediction System Agentic IoT using ESP32 + AI + n8n Automation + Telegram Voice Alerts + Google Sheets + ThingSpeak Cloud Dashboard 1. Project Overview This project is an advanced AI-powered IoT healthcare monitoring system that continuously monitors ECG signals using an ESP32 microcontroller and predicts possible heart abnormalities using AI logic. The system integrates: ESP32 for sensor data acquisition ECG sensor module (AD8232) AI-based heart disease prediction n8n automation workflows Telegram voice alerts Google Sheets cloud logging ThingSpeak real-time dashboard Agentic IoT automation Web dashboard visualization The solution can be used for: Remote patient monitoring Smart healthcare systems Elderly monitoring Preventive cardiac diagnosis Wearable healthcare projects AI-assisted hospital systems 2. System Architecture ECG Sensor (AD8232) │ ▼ ESP32 Board │ ┌───────────┼───────────┐ ▼ ▼ ▼ ThingSpeak n8n Workflow AI Engine Dashboard │ │ ▼ ▼ Telegram Alerts Heart Disease Voice Msg Prediction │ ▼ Google Sheets 3. Features Core Features ✅ Real-time ECG Monitoring ✅ Heart Disease Prediction using AI ✅ Cloud Dashboard Visualization ✅ Telegram Notification Alerts ✅ Voice Notification Automation ✅ Google Sheets Data Logging ✅ ESP32 WiFi Connectivity ✅ ThingSpeak IoT Dashboard ✅ Agentic AI Decision Making ✅ Abnormal Heartbeat Detection ✅ BPM Calculation ✅ ECG Waveform Monitoring 4. Components List Component Quantity Purpose ESP32 Dev Board 1 Main controller AD8232 ECG Sensor 1 ECG signal acquisition Jumper Wires Several Connections Breadboard 1 Prototyping USB Cable 1 ESP32 programming Power Supply 1 System power WiFi Network 1 Cloud communication Smartphone 1 Telegram alerts Laptop/PC 1 n8n & monitoring 5. Working Principle ECG sensor captures heartbeat signals. ESP32 reads analog ECG waveform. BPM is calculated. AI logic evaluates ECG abnormalities. Data uploaded to ThingSpeak cloud. n8n receives webhook data. Telegram sends alert notifications. Voice alerts generated automatically. Google Sheets stores historical records. 6. Circuit Schematic Diagram AD8232 to ESP32 Connections AD8232 Pin ESP32 Pin OUTPUT GPIO34 3.3V 3.3V GND GND LO+ GPIO26 LO- GPIO27 7. Circuit Diagram (Text Representation) +-------------------+ | ESP32 | | | ECG OUT --> GPIO34 | LO+ --> GPIO26 | LO- --> GPIO27 | 3.3V --> 3.3V | GND --> GND | +-------------------+ 8. Flowchart START │ ▼ Initialize ESP32 WiFi │ ▼ Read ECG Sensor Data │ ▼ Calculate BPM │ ▼ AI Prediction Logic │ ┌──────┴───────┐ ▼ ▼ Normal Abnormal │ │ ▼ ▼ Upload Data Send Alert │ │ ▼ ▼ ThingSpeak Telegram Voice │ │ └──────┬───────┘ ▼ Google Sheets │ ▼ LOOP 9. ESP32 Source Code (Arduino IDE) #include #include const char* ssid = "YOUR_WIFI_NAME"; const char* password = "YOUR_WIFI_PASSWORD"; String apiKey = "THINGSPEAK_API_KEY"; const int ecgPin = 34; int threshold = 550; void setup() { Serial.begin(115200); WiFi.begin(ssid, password); while (WiFi.status() != WL_CONNECTED) { delay(1000); Serial.println("Connecting..."); } Serial.println("WiFi Connected"); } void loop() { int ecgValue = analogRead(ecgPin); Serial.println(ecgValue); String condition = "Normal"; if(ecgValue > threshold) { condition = "Abnormal"; } if(WiFi.status() == WL_CONNECTED) { HTTPClient http; String url = "http://api.thingspeak.com/update?api_key=" + apiKey + "&field1=" + String(ecgValue); http.begin(url); int httpCode = http.GET(); Serial.println(httpCode); http.end(); } delay(2000); } 10. AI Heart Disease Prediction Logic Basic AI Logic The AI engine analyzes: ECG amplitude BPM variations Signal irregularities Threshold crossings Heart rhythm pattern Prediction Categories ECG Condition Prediction Normal waveform Healthy Irregular spikes Arrhythmia Risk High BPM Tachycardia Low BPM Bradycardia Noise patterns Sensor Error 11. Advanced AI Enhancement You can improve prediction using: TensorFlow Lite TinyML on ESP32 Edge AI inference Deep learning ECG classification Possible datasets: MIT-BIH Arrhythmia Dataset PhysioNet ECG Database 12. ThingSpeak Cloud Dashboard Setup Using ThingSpeak Steps Create account Create new channel Add fields: ECG Value BPM Prediction Copy Write API Key Insert API key into ESP32 code View real-time graphs 13. Google Sheets Integration Using: Google Apps Script n8n webhook automation Stored Parameters Timestamp ECG BPM Prediction Time ECG Value Heart Rate AI Result 14. Telegram Bot Setup Using Telegram BotFather Steps Open Telegram Search: BotFather Create new bot Copy Bot Token Obtain Chat ID Integrate into n8n workflow 15. Voice Notification Automation Example Voice Alerts Warning! Abnormal heart activity detected. Please check patient condition immediately. Voice Generation Methods Google Text-to-Speech Telegram Voice API ElevenLabs TTS gTTS Python library 16. n8n Automation Workflow Using n8n Automation Platform Workflow Process Webhook Trigger │ ▼ Receive ECG Data │ ▼ AI Analysis │ ▼ Condition Check │ ┌────┴─────┐ ▼ ▼ Normal Abnormal │ │ ▼ ▼ Log Data Telegram Alert │ │ ▼ ▼ Google Sheets Voice Message 17. Example n8n Workflow JSON { "nodes": [ { "parameters": {}, "name": "Webhook", "type": "n8n-nodes-base.webhook", "typeVersion": 1, "position": [250, 300] }, { "parameters": {}, "name": "Telegram", "type": "n8n-nodes-base.telegram", "typeVersion": 1, "position": [600, 300] } ], "connections": {} } 18. Web Dashboard Features Dashboard Includes Real-time ECG graph BPM display AI prediction result Alert status Patient history Device connectivity status 19. Agentic AI Features The system behaves like an autonomous AI agent: ✅ Detects anomalies ✅ Makes decisions ✅ Sends alerts automatically ✅ Stores data autonomously ✅ Predicts heart abnormalities ✅ Triggers emergency notifications 20. Power Consumption Prediction Logic AI Power Optimization The ESP32 predicts usage patterns: State Power Mode Idle Deep Sleep Monitoring Active Alert Mode High Performance Optimization Techniques Deep sleep mode Sensor polling intervals Adaptive WiFi transmission Edge AI processing 21. Security Enhancements Recommended Security HTTPS APIs Secure MQTT Token authentication Encrypted cloud communication User authentication 22. Future Enhancements Future Scope AI Enhancements Deep learning ECG analysis CNN-based arrhythmia detection Cloud AI diagnosis IoT Enhancements MQTT communication Firebase integration AWS IoT Core Edge AI Healthcare Enhancements Multi-patient monitoring Doctor dashboard Emergency ambulance alerts GPS tracking Hardware Enhancements OLED display Battery backup Wearable ECG device Mobile app integration 23. Deployment Guide Hardware Deployment Assemble ECG circuit Upload ESP32 firmware Connect WiFi Verify sensor readings Configure ThingSpeak Configure n8n Setup Telegram bot Test alerts 24. Testing Procedure Test Expected Result ECG Reading Real-time waveform BPM Calculation Accurate BPM Cloud Upload Data visible Telegram Alert Alert message received Voice Notification Audio alert plays Google Sheets Data logged 25. Applications Healthcare Applications Smart hospitals Remote healthcare Elderly monitoring ICU monitoring Fitness tracking Home healthcare systems 26. Advantages ✅ Low-cost healthcare solution ✅ Real-time monitoring ✅ AI-assisted diagnosis ✅ Remote accessibility ✅ Cloud integration ✅ Automation support ✅ Scalable architecture 27. Limitations ⚠ Not a certified medical device ⚠ Requires proper ECG electrode placement ⚠ AI predictions are indicative only ⚠ Internet required for cloud features 28. Conclusion The AI-Based ECG and Heart Disease Prediction System combines: Embedded systems Artificial intelligence IoT cloud monitoring Automation workflows Agentic healthcare intelligence This project demonstrates how ESP32, AI, n8n automation, and cloud technologies can create an intelligent remote healthcare monitoring ecosystem capable of real-time prediction, autonomous alerts, and scalable deployment. 29. Recommended Software & Platforms Arduino IDE ESP32 Board Package ThingSpeak Cloud n8n Workflow Automation Google Sheets Telegram API Documentation TensorFlow Lite for Microcontrollers