Preliminary study of image processing techniques for the detection of varicocele based on 2D ultrasound images

Abstract

Varicocele is quite common among male population but can only be confirmed and treated through surgery which may affect the fertility. While ultrasound (color Doppler technique) can be used to diagnose the varicocele, the need to perform the Valsalva maneuver technique when performing the color Doppler ultrasound scanning might make patient become uncomfortable. Therefore, this preliminary study has been done to see whether any image processing techniques can be used to enhance the 2D ultrasound image to help the diagnosis. The purpose of this study is to enhance the ultrasound image of 2D varicocele testicle by using image enhancement and edge detection. The images used are the retrospective images of the normal testicle, and testicle with varicocele and severe varicocele (dilated vein). The image enhancement was done using median filter while the edges were detected using Sobel operator. Based on the analysis, sobel technique does an excellent job to show the varicocele region especially the dilated vein. To summarise, further image processing methods could be applied using different image processing techniques as the analysis could be helpful in giving the urologist the second opinion during diagnosis of varicocele before the surgery.