Frequency-difference MIT imaging of cerebral haemorrhage with a hemispherical coil array: numerical modelling

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Electromagnetics

Background:

• Cerebral hemorrhages pose significant diagnostic challenges.
• Current imaging techniques have limitations in sensitivity and specificity.
• Novel coil array designs offer potential for improved detection.

Purpose of the Study:

• To investigate the feasibility of detecting cerebral hemorrhages using a hemispherical MIT coil array.
• To evaluate the impact of noise and boundary errors on image quality and detection accuracy.
• To determine optimal operating parameters for hemorrhage detection.

Main Methods:

• Simulations using a finite difference method and an anatomically realistic head model with 12 tissue types.
• Reconstruction of frequency-difference images from modelled data.
• Introduction of varying levels of phase noise and boundary errors (displacement and scaling).

Main Results:

• A noise level of 3 m degrees (standard deviation) was sufficient for visualizing peripheral strokes (approx. 49 ml).
• Head displacement errors must be within 3-4 mm to avoid significant image artifacts.
• Size scaling errors need to be within 3-4% to prevent unacceptable image artifacts.

Conclusions:

• The hemispherical MIT coil array shows promise for detecting cerebral hemorrhages.
• Image quality and diagnostic accuracy are sensitive to noise and boundary errors.
• Precise control over imaging parameters is crucial for reliable hemorrhage detection.