Research Objectives

• Develop an IoT-based emergency fall detection and response module using wearable sensors (ESP32 and MPU6050) to monitor patient movements, detect falls in real time, capture GPS location, and automatically notify nearby hospitals and ambulance services via a mobile app.
  
• Implement an AI-powered woun classification and triage recommendation system that uses a Convolutional Neural Network (CNN) trained on a diverse wound image dataset to classify injury types, estimate severity, and suggest appropriate treatment pathways integrated into a mobile-accessible interface for emergency medical personnel.
• Design and deploy a predictive resource management system utilizing Long Short-Term Memory (LSTM) models trained on historical hospital data to forecast ICU bed availability and medicine stock needs, enabling proactive resource allocation through a user-friendly dashboard connected to hospital systems.
• Develop a smart blood inventory management and emergency coordination platform combining hybrid ML models (ARIMA and LSTM) for blood demand forecasting, GPS-based donor tracking, and automated donor outreach via SMS/email to optimize blood supply, reduce wastage, and improve emergency response logistics.
• Integrate all modules into a unified, interoperable SMARTCARE system with real-time data synchronization, secure cloud infrastructure, intuitive user interfaces, and compliance with healthcare data privacy standards to ensure seamless coordination among patients, caregivers, hospitals, and emergency responders.