Secure Healthcare IoT Monitoring Dataset - Anomaly Detection for Patient Safety

Dataset Overview

This dataset contains real-time physiological and network-level data collected from a Secure Healthcare IoT Monitoring System involving 2000 patients. It combines biometric health readings with network security metadata to enable both health monitoring and cybersecurity threat detection in IoT healthcare systems.

Key Features

• Patient_ID: Unique identifier for 2000 patients
• Timestamp: Real-time data collection timestamp
• Heart_Rate (bpm): Heart rate measurements (60-100 bpm range)
• Body_Temperature (°C): Body temperature readings (36.0-39.5°C)
• Blood_Pressure: Blood pressure measurements
• Device_ID: Unique IoT device identifier (1331 unique devices)
• IP_Address: Network IP address (1972 unique IPs)
• Access_Type: Type of network access (234 types)
• Action_Performed: Action logged by the system
• Target: Binary label (0 = Normal, 1 = Anomalous event)

Target Distribution

• Normal Activity (0): 1,724 records (86.2%)
• Anomalous Events (1): 276 records (13.8%)

Use Cases

• Healthcare IoT anomaly detection and intrusion detection systems
• Cybersecurity threat analysis in medical IoT networks
• Patient health monitoring with security overlay
• Binary classification: Normal vs. Anomalous behavior
• Network traffic analysis in healthcare systems
• Real-time alert systems for patient safety and data security

Machine Learning Applications

• Binary classification: Anomaly detection (Normal/Anomalous)
• Supervised learning: Security threat prediction
• Time-series analysis: Patient vital sign trends
• Feature engineering: Combining health + network features
• Imbalanced dataset handling techniques (86:14 ratio)

Research Applications

• Healthcare cybersecurity research
• IoT security in medical environments
• Patient data privacy protection
• Real-time threat detection in healthcare networks