Deconvolution-Based CT and MR Brain Perfusion Measurement: Theoretical Model Revisited and Practical Implementation...

Area of Science:

• Medical Imaging
• Biophysics
• Computational Neuroscience

Background:

• Deconvolution analysis of CT and MR brain perfusion data is a key clinical tool.
• Ongoing research seeks to refine these analytical methods for improved accuracy.

Purpose of the Study:

• To provide a comprehensive explanation of the physiological model for intravascular tracer systems.
• To detail practical implementation aspects of deconvolution algorithms for brain perfusion analysis.
• To discuss regularization techniques for achieving physiologically plausible results.

Main Methods:

• Derivation and explanation of the underlying physiological model.
• Focus on algebraic deconvolution methods using singular value decomposition (SVD).
• Discussion of necessary preprocessing steps and regularization strategies.

Main Results:

• A detailed theoretical and practical framework for deconvolution-based perfusion analysis.
• Emphasis on necessary simplifications of the physiological model for current CT and MR scanner data.
• Insights into obtaining physiologically reasonable results through regularization.

Conclusions:

• The paper offers a thorough guide to CT and MR brain perfusion analysis.
• It highlights the importance of understanding the physiological model and practical implementation details.
• Provides a foundation for further research and clinical application of deconvolution techniques.