Optimal number of averaged frames for noise reduction of ultrasound images

Abstract

Data averaging is a common noise reduction practice in several data processing procedures. The authors have used this approach in order to minimize the noise of the ultrasound signals employed in the backscatter analysis of the heart. They assess the optimum number of frames to be used in the averaging routine in order to obtain, in the search for maximum noise reduction, a good compromise between the two opposite criteria: indicating on the one hand the use of the greatest number of frames and the other hand the smallest. Results obtained show that 10 frames averaging allows 50% noise reduction which is equivalent to 90% of the noise reduction obtained by averaging a 4 times greater number of frames.<>