Traceable Dielectric-Property Dataset for Femoral-Fracture Monitoring (1-4 GHz): Cross-Database Validation and ML Proof of Concept

This dataset provides traceable dielectric properties for biological tissues relevant to microwave-based femoral fracture monitoring in the 1-4 GHz band. It serves as a baseline for forward electromagnetic simulations and algorithmic benchmarking. Contents: Anchor Tables: Dielectric properties for 23 tissue types (healthy tissues, healing-stage constituents, bone variants) at 1–4 GHz. Dispersive Models: Single-term Cole-Cole parameters optimised for time-domain FDTD simulation. Fracture Scenarios: Effective dielectric properties for 5 fracture types × 5 healing stages. Uncertainty Bounds: Calculated Maxwell-Garnett and Bruggeman bounds (EMT). Simulation Config: Representative 2-D FDTD configuration (JSON). ML Code: Python script and results for the synthetic fracture classification experiment (Proof of Concept). Key Features: Full traceability to IT'IS v5.0 database and Gabriel et al. (1996). Cross-database validation methodology. Uncertainty bounds derived from physiological variability. Machine-readable formats (CSV, JSON, Python) for direct computational use. Related Publication: Aldelemy A, Al-Dulaimi A, Ali NT, Abd-Alhameed RA. "Development of a Traceable Dielectric-Property Dataset for Femoral-Fracture Monitoring in the 1-4 GHz Band with Cross-Database Validation and Machine-Learning Proof of Concept." IEEE Access, 2025.