Neuroradiologic MR applications with multiparametric color composite display

Area of Science:

• Medical Imaging
• Neuroscience
• Computer Science

Background:

• Magnetic Resonance (MR) imaging is crucial for diagnosing neuropathology.
• Displaying complex neuroanatomic and neuropathologic details can be challenging with standard MR images.

Purpose of the Study:

• To demonstrate a PC-based multiparameter full-color composite display technique for MR images.
• To assess the technique's ability to display clinically important neuroanatomic and neuropathologic tissues.

Main Methods:

• Applied image-processing software to spatially aligned multiparameter MR image sets (proton density, T1-weighted, T2-weighted) from 14 patients.
• Generated intensity-based color composites using custom and commercial software on a PC with 24-bit graphics.
• Utilized quantitative color channel applications to differentiate tissue types into unique color regions.

Main Results:

• The technique successfully generated full-color composites from multiparameter MR image sets.
• Quantitative analysis confirmed the ability to differentiate anatomically and pathologically confirmed tissues into unique color regions.
• Pathologic correlation supported the diagnostic relevance of the displayed tissue contrast information.

Conclusions:

• PC-based full-color composite generation is a viable technique for displaying diagnostically relevant MR imaging contrast information.
• This method allows for simultaneous display of uniquely color-coded neuroanatomic and neuropathologic tissue information.
• The technique produces composites that maintain partially natural-appearing images while highlighting diagnostic details.