Title: Boundary detection of renal using contour segmentation

Authors: R. Gomalavalli; S. Muttan; P.M. Venkata Sai

Addresses: Department of ECE, College of Engineering, Anna University, Guindy, Chennai 600025, Tamil Nadu, India ' Department of ECE, College of Engineering, Anna University, Guindy, Chennai 600025, Tamil Nadu, India ' SRMC & RI, Sri Ramachandra University, Porur, Chennai 600116, Tamil Nadu, India

Abstract: In this paper, an active method for computer-aided renal segmentation on abdominal Computed Tomography (CT) images with anatomic building attention is presented. The proposed segmentation system is considered to assist the medical practitioners in clinics and learning methodology. The planned method is a common sound segmentation of renal. The major elements of the proposed system are (i) detect and check the objects; (ii) region of interest of renal location is detected; (iii) determination of Dice similarity coefficient (≤1) degree of accuracy is acquired; (iii) novel work is to verify whether the neighbouring organs are alarmed or not (DSC <0.8501); and (iv) statistical quality measurement of boundary is done. This is a compatible method to avoid clinical needle biopsy. To overcome the limitations present in most of the cases, Active Contour Detection method is applied. The qualitative examination on 52 patients indicates the average comparison of 76.2% between the segmentation methods.

Keywords: active contour detection; segmentation; neighbouring organs; Sorensen dice similarity coefficient; biopsy; structural similarity index measure; peak signal to noise ratio; Jaccard coefficient; Jaccard distance; mean squared error.

DOI: 10.1504/IJBET.2018.094428

International Journal of Biomedical Engineering and Technology, 2018 Vol.28 No.1, pp.53 - 66

Received: 20 Jun 2016
Accepted: 27 Sep 2016
Published online: 03 Sep 2018 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article