Noise power spectrum of ultrasound uniformity images

Abstract

Background: The Noise Power Spectrum (NPS) is an effective tool for providing information on how the total noise variance of uniformity images is distributed in frequency space. The NPS is a tool which can widen the understanding of noise in ultrasound as noise in ultrasound includes both electronic stochastic noise and static speckle. However, surprisingly very little research has been done and unlike projection XRI and CT, NPS is not included in QC protocols for ultrasound. Objectives: (1) To establish a scan protocol to obtain the NPS for ultrasound scanners and linear transducers (2) To specify an IQWorks analysis tree for the calculation of All Noise, Stochastic and Static NPS for ROIs at near field, focal region and far field (3) To investigate the variation of NPS with output power over the three regions (4) To investigate the short term reliability of the methodology. Research Design: The protocol was developed on a Siemens ACUSON Sequoia™ and two phantoms (Gammex RMI404GS LE, CIRS Model 551). ImageJ and IQWorks were used to calculate the NPS at the near field, focal region and far field. The reliability of the methodology was checked by calculating the coefficient of variation (CoV) of the mean pixel value (MPV), image variance and NPS over repeat acquisitions. Results: The NPS of ultrasound systems includes a high low-frequency component arising from the speckle. The far field region experienced the highest stochastic and static NPS values as the output power decreased. There was a sudden increase in the stochastic NPS at the focal region for both phantoms when the phantom decreased from -4dB to -7dB. CoVs for both phantoms were generally less than 8%. Conclusion: Noise in ultrasound is mainly of low-frequency component. There are interesting changes of the NPS at different depths and as the output power is reduced. Recommendations: The NPS tool can be used as part of an ultrasound QC programme using the NPS analysis tree specified in this project.