Fast scan conversion algorithms for displaying ultrasound sector images

Area of Science:

• Medical imaging
• Signal processing
• Computational geometry

Background:

• Ultrasonic sector-scans are crucial for medical diagnostics.
• Efficient and accurate interpolation is vital for image quality and processing speed.
• Existing interpolation methods have limitations in speed or accuracy.

Purpose of the Study:

• To develop and evaluate two novel, fast algorithms for ultrasonic sector-scan interpolation.
• To compare the performance of these new algorithms against conventional methods.

Main Methods:

• Developed two interpolation algorithms based on line-drawing techniques.
• Eliminated multiplications in the innermost loops for enhanced speed.
• Compared new algorithms with 2-D windowed sinc, bicubic spline, 4x4 bicubic spline, bilinear, and nearest neighbor interpolators.

Main Results:

• The new algorithms are significantly faster, with the most accurate one being approximately eight times faster than nearest neighbor interpolation.
• Quantitative errors of the new algorithms are comparable to nearest neighbor interpolation.
• Subjective image quality falls between nearest neighbor and bilinear interpolation.

Conclusions:

• The developed algorithms offer a substantial speed improvement for ultrasonic sector-scan interpolation.
• They provide a good balance between speed and accuracy, suitable for real-time applications.
• These algorithms represent an advancement in efficient medical image processing.