Lumen Coronary Artery Border Detection Using Texture and Chi-square Classification

In this paper, we present the lumen coronary artery border detection using intravascular ultrasound (IVUS) images. The approach make used of texture analysis based on Binary Robust Independent Elementary Features (BRIEF) and Chi-square classification. This proposed method can detect the boundary and calculate the area within the lumen coronary artery border. This method was tested on thirty samples of IVUS images which were obtained from Computer Vision Centre, Bellaterra, Dept. Matematica Aplicada i Analisi, Universitat de Barcelona, Barcelona. The Bland Altman plot is used to show the variation between the proposed automatic segmentation method and ground truth when three different threshold were used. The segmentation performance of the proposed method is measured using Jaccard Index (JI), Hausdorff Distance (HD), Area Overlap Error (AOE), Percentage Area Difference (PAD) and Dice Similarity Index (DI). In this study, the results show that the border detection is better when threshold TH5 is used.