Impact of Bit Errors in Digitized RF Data on Ultrasound Image Quality
Zhao Chen (TU Delft - Electronic Instrumentation)
Mehdi Soozande (ImPhys/Acoustical Wavefield Imaging )
Rik Vos (Erasmus MC, ImPhys/Acoustical Wavefield Imaging )
Hans Bosch (ImPhys/Acoustical Wavefield Imaging )
Martin Verweij (TU Delft - ImPhys/Medical Imaging, Erasmus MC, ImPhys/Acoustical Wavefield Imaging )
Nico de Jong (Erasmus MC, ImPhys/Acoustical Wavefield Imaging )
Michiel Pertijs (TU Delft - Electronic Instrumentation)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
This article quantitatively analyzes the impact of bit errors in digitized RF data on ultrasound image quality. The quality of B-mode images in both linear array and phased array imaging is evaluated by means of three objective image quality metrics: peak signal-to-noise ratio, structural similarity index, and contrast-to-noise ratio, when bit errors are introduced to the RF data with different bit-error rates (BERs). The effectiveness of coding schemes for forward error detection and correction to improve the image quality is also studied. The results show that ultrasound imaging is inherently resilient to high BER. The image quality suffers unnoticeable degradation for BER lower than 1E-6. Simple 1-bit parity coding with 9% added redundancy helps to retain similar image quality for BER up to 1E-4, and Hamming coding with 33.3% added redundancy allows the BER to increase to 1E-3. These results can serve as a guideline in the datalink design for ultrasound probes with in-probe receive digitization. With much more relaxed BER requirements than in typical datalinks, the design can be optimized by allowing fewer cables with higher data rate per cable or lower power consumption with the same cable count.