Addendum: B-spline interpolation in medical image processing

Area of Science:

• Medical Imaging
• Image Processing
• Computer Vision

Background:

• Image interpolation is crucial for medical imaging analysis.
• Various interpolation schemes exist, necessitating evaluation of their performance.
• Previous surveys highlight the need for optimal interpolation methods in medical contexts.

Purpose of the Study:

• To evaluate B-spline interpolation techniques of degrees 2, 4, and 5.
• To compare these B-spline methods against established criteria for image interpolation.
• To determine the suitability of high-degree B-splines for medical image processing.

Main Methods:

• Analysis of B-spline interpolation algorithms of varying degrees (2, 4, 5).
• Evaluation based on criteria from a 1999 survey on medical image interpolation.
• Assessment of Fourier properties, interpolation error, and computational time.

Main Results:

• High-degree B-spline interpolation demonstrated superior Fourier characteristics.
• These methods yielded the smallest interpolation errors among those tested.
• B-spline interpolation of higher degrees offered reasonable computing times.

Conclusions:

• High-degree B-splines are preferable interpolators for medical image processing.
• These techniques are particularly advantageous when high precision is a requirement.
• The findings are robust across different geometric transformations and image content, provided aliasing is avoided.