Int J Performability Eng ›› 2021, Vol. 17 ›› Issue (11): 955-965.doi: 10.23940/ijpe.21.11.p6.955965

Previous Articles     Next Articles

Design and Analysis of Dual Transition Function Cellular Automata-based Filter of Brachial Plexus Ultrasound Images

Ankur Bhardwaja,*, Sanmukh Kaurb, A.P. Shuklac, and Manoj Shuklad   

  1. a,b,dAmity School of Engineering and Technology, Amity University, Noida, India;
    cKIET Group of Institutions, Ghaziabad, India
  • Submitted on ; Revised on ; Accepted on
  • Contact: *E-mail address:

Abstract: The sensory system of human upper limbs is controlled by the brachial plexus nerves. Upper limb surgical anesthesia is based on accurately segmenting this nerve system using ultrasound scans. The incorporation of images can include many types of noise. Speckle noise is one of the common noises introduced during the capturing of ultrasound images, and denoising of this is one of the major challenges as it causes a lot of trouble for clinicians during the diagnosis process. Cellular Automata is a computational model that uses simple rules to represent a complex system. It has been used extensively in image processing operations. Because of the ease with which a digital image may be mapped to a cellular automaton and the ability to perform different image processing procedures in real time, it appears to be a natural tool for image processing. In this paper a novel cellular automaton based despeckling filter for ultrasound images of brachial plexus nerves is suggested and its effect is compared with the various existing noise filters such as Kuan, Lee, Lee diffusion, frost diffusion and wavelet filter. Results are compared and analyzed both qualitatively and quantitatively. Various parameters such as mean square error, signal to noise ratio, square root mean square error, peak signal to noise ratio, structured similarity index are used for quantitative comparison. The effect of all filters under consideration have been analyzed at low, medium, and high levels of noise in ultrasound images. It has been observed that the proposed filter performs well both in terms of noise filtering as well as retaining the structure of the original image in most cases. For qualitative analysis, expert opinion of scientific officer grade D at ACTREC, Mumbai has been considered.

Key words: speckle noise, MSE, PSNR, SRMSE, SSIM, cellular automata