Ameliorating slice gaps in multislice magnetic resonance images: an interpolation scheme

Area of Science:

• Medical Imaging
• Image Processing
• Magnetic Resonance Imaging

Background:

• Standard 2D MRI protocols create slice gaps (SG) in orthogonal plane images.
• This data loss reduces image resolution and limits 3D reconstruction.
• Existing interpolation methods may not fully utilize orthogonal plane information.

Purpose of the Study:

• Introduce a novel interpolation method for 2D MRI datasets.
• Enhance image resolution using a priori scanning knowledge.
• Ameliorate data loss from slice gaps.
• Reconstruct 3D datasets from 2D MRI images.

Main Methods:

• Developed and simulated a novel interpolation algorithm in Matlab.
• Validated the algorithm using Shepp-Logan and Gaussian phantoms (2D/3D, varying matrix sizes).
• Assessed performance on human brain MRI and ACR accreditation phantom datasets.

Main Results:

• The novel interpolation scheme demonstrated higher accuracy than common methods.
• Quantitative analysis using squared error and mean squared error confirmed superior performance.
• Qualitative assessment via mean structure similarity matrix and MRI scans supported the method's effectiveness.

Conclusions:

• An efficient interpolation approach effectively resolves slice gaps in 2D MRI.
• The method successfully fills missing data points by leveraging orthogonal plane information.
• This technique has significant applications in clinical MRI, fMRI, DTI, and MRA for generating 3D volumes from 2D acquisitions.