Experiment Of Kidney Ultrasound Images Processing Using Various Filter Utilizing Streamlit

Ultrasound is still considered a significant noninvasive imaging technique. However, ultrasound images are generally contaminated with noise. It is possible to reduce grain noise in ultrasound images by using various filters. In this paper, each image is then filtered using non-adaptive, adaptive, and morphological filter utilizing Streamlit. Filters will be applied to four images produced by different ultrasound machines. The performance of each filter is analyzed by the Mean Square Error (MSE) and Peak Signal to Noise Ratio (PSNR). The conclusion, the filter is the most significant factor to reduce the speckle noise from the four different ultrasound machines on kidney images. The response table means that the order of factor from the first to the last is filter, image, and filters size, from this experiment, the minimum error is 0 and the maximum PNSR is 89.3 by Kuan Filter. The application is simple and user friendly as it can be designed as needed.

I. Introduction

Ultrasound imaging are known to be reliable imaging modality because of its trait of being real-time, non-invasive, ease of performance, non-toxicity, and low cost[1]–[3]. Due to its many positivity, ultrasound is a commonly used modality for diagnostic method. However, ultrasound images often suffer from reduced contrast and the interference of speckle noise, which complicates the interpretation of these images.