Enhancement of Coronary Blood Vessels based on Frangi’s Vesselness Filter and Morphological Operations
Sukanya A1, Rajeswari R2
1Sukanya A, Department of Computer Applications, Bharathiar University, Coimbatore, India.
2Rajeswari R, Department of Computer Applications, Bharathiar University, Coimbatore, India.
Manuscript received on 04 July 2019 | Revised Manuscript received on 08 July 2019 | Manuscript published on 30 August 2019 | PP: 274-281 | Volume-8 Issue-10, August 2019 | Retrieval Number: I8168078919/2019©BEIESP | DOI: 10.35940/ijitee.I8168.0881019
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Cardiovascular diseases (CVDs) are the global cause of deaths and therefore research in modern medical image processing aims to develop a medical tools to assist the clinicians in vessel extraction, artery detection and 3D reconstruction. Vessel extraction is an important and trivial step which depends extremely on enhancement method. Extraction of coronary artery blood vessels from 3 Dimension (3D) Coronary Computed Tomography Angiography (CCTA) images is a demanding research objective to strengthen the diagnosis and therapy of coronary artery illness. This paper presents a vessel enhancement method of coronary artery blood vessels using Frangi’s vesselness measure and morphological operators. In the first stage of the proposed work, Preprocessing is performed to consider only the heart region. Next Frangi’s vesselness measure is calculated for the 3D CCTA images. While calculating the Frangi’s vesselness measure, four different types of gradient operators are used for calculating the Hessian matrix viz., Sobel, Prewitt, central difference and intermediate difference operators. In the second stage, the vessels are enhanced by morphological operations based on top hat and bottom hat operations. These morphological operations help in further enhancing the blood vessels. The proposed methodology was applied on 12 3D CCTA dataset and evaluated using quality measures such as MSE, PSNR, SSIM and FSIM. The results obtained based on the four gradient operators are compared. The statistical test viz., one way ANOVA was carried out on the results. The proposed method using Prewitt operator is able to extract even small vessels and the results seem to be promising.
Keywords: Coronary artery, Coronary computed tomography angiography (CCTA), Hessian filter, Morphological operations, Vessel enhancement.
Scope of the Article: Image Processing and Pattern Recognition