Microwave Scattering Simulations for Brain Stroke Detection

Finite-element method (FEM) simulations of microwave scattering from a three-dimensional human head model segmented into 21 tissue types. The data was generated to analyze the detection and classification of hemorrhagic and ischemic brain strokes by comparing electric field values and dielectric properties of normal and abnormal tissues. The simulations were conducted by Awais Munawar Qureshi and published in 2020.

Use Cases

• Analyze the contrast in electric field values between normal and stroke-affected brain tissues to identify stroke locations.
• Train a classifier to differentiate between hemorrhagic and ischemic stroke types using simulated backscattered microwave signals.
• Validate microwave imaging algorithms by using the simulated scattering data from the 21-segment head model as ground truth.
• Study the specific absorption rate (SAR) to assess the ionization effects of microwave signals on the 3D head model.

Strengths

• The head model is anatomically detailed, segmented into 21 different tissue types.
• Simulations include analysis of two distinct brain stroke types (hemorrhagic and ischemic) at various locations.
• The FEM method was validated with mesh convergence and iterative solver comparisons for error-free results.

Limitations

• The dataset is based on a single, simulated 3D head model, limiting generalizability to anatomical variations.
• Data is derived from computational simulations, not physical measurements, which may not capture all real-world complexities.
• The temporal scope is a snapshot from 2020, and the underlying MRI data source and simulation parameters are not fully detailed.

Provenance

Source

Awais Munawar Qureshi via Dryad.

Collection Method

Generated via finite-element method (FEM) simulations of microwave scattering on a 3D head model developed from an MRI database.

Freshness

Last updated 2020-06-30.