Diffusion Imaging

Diffusion imaging is a form of imaging based on measuring random Brownian motion of water molecules per voxel.  Diffusion is primarily hindered by cell membrane boundaries and represents the diffusion of a various compartments. 

Current Projects

Advanced B0 Distortion Correction: DWI scans are susceptible to B0 geometric distortions due to imhomogeneities in the main magnetic field.  I am currently developing nonlinear correction techniques to correct the DWI scans when a field map is not present.

 B0 image pre-correction (Top) and Structural T2 Image (Bottom)

B0 image pre-correction (Top) and Structural T2 Image (Bottom)

 Advanced Distortion Correction (Top) and Structural T2 Image (Bottom)   

Advanced Distortion Correction (Top) and Structural T2 Image (Bottom)

 

Advanced Eddy Current Correction:  Optimized eddy current correction in order to better correct for large head movements, linear stretching in the phase encoding direction due to gradient coils heating up and improve resampling scheme to reduce spatial smoothness.

     Advanced Eddy Current Correction.                         Standard Eddy Current Correction

    Advanced Eddy Current Correction.                         Standard Eddy Current Correction

Rician noise removal:  Diffusion metrics are highly susceptible to noise, especially in areas of low SNR.  Denoising helps reduce noise levels while maintaining anatomical details.

 

 Top Row: B=3000 scan with high levels of noise  Bottom Row: B=3000 scan after de-noising for Rician noise profile

Top Row: B=3000 scan with high levels of noise

Bottom Row: B=3000 scan after de-noising for Rician noise profile