Artifactual microhemorrhage generated by susceptibility weighted image processing.

Area of Science:

• Medical Imaging
• Neuroimaging
• Radiology

Background:

• Susceptibility Weighted Imaging (SWI) is crucial for detecting microhemorrhages.
• Artifacts in SWI can lead to misinterpretation of pathology, particularly in traumatic brain injury (TBI).
• The 2D homodyne filtering method is a common technique for generating SWI phase data.

Purpose of the Study:

• To investigate the introduction of artifactual microhemorrhages by the 2D homodyne filtering method in SWI.
• To evaluate the impact of open-ended fringelines (OEFs) in phase data on SWI artifact generation.
• To compare the efficacy of 2D homodyne filtering versus 3D phase unwrapping methods in minimizing artifacts.

Main Methods:

• SWI data from 28 TBI patients were acquired using 3T Siemens scanners with GRE and sEPI sequences.
• Image processing involved both 2D homodyne filtering and 3D Fourier-based phase unwrapping followed by high-pass filtering.
• Neuroradiologist reviewed magnitude and phase images for OEFs and microhemorrhages on a PACS workstation.

Main Results:

• Type III OEFs were present in 68% of GRAPPA-accelerated GRE datasets.
• Artifactual microhemorrhages were identified in 89% of SWI cases processed with 2D homodyne filtering when OEFs were present.
• 3D phase unwrapping minimized phase singularities and did not generate microhemorrhage-like artifacts.

Conclusions:

• The 2D homodyne filtering method can create artifactual microhemorrhages in SWI when type III OEFs are present, potentially leading to overestimation of TBI pathology.
• 3D phase unwrapping methods effectively minimize these artifacts.
• Further research is needed to develop methods for properly combining phase data across coils to eliminate these artifacts entirely.