Coronary Artery Blockage Detection by Automated Segmentation of Vessels in X-Ray Angiograms

Coronary heart disease leading up to stenosis, the partial or total blocking of coronary arteries, is the leading cause of death worldwide. Multi-vessel coronary artery disease affecting two or more coronary requires interpretive expertise on the assessment, the process of interpreting is complex and a time-consuming. Auto-mated identification and classification of angiograms with blockage detection from minimally invasive procedures would be of great clinical value. This study aims at OpenCV method with the help of adaptive thresholding, and brightness corrections for the segmentation of the blood vessels and used the same masked images for the detection of blockage in x-ray angiograms using deep neural nets. The proposed model have achieved 97% accuracy on the task of classifying the x-ray angiogram to ‘Blockage Detected’ and ‘Normal’, with a F1-Score of 0.9532. These results open the way to a fully automated method for the identification of Blockage from X-ray angiograms.