Toolbox for vessel X-ray angiography images simulation

Abstract

In recent years, automatic computer techniques have been proven to be a great tool for the rapid detection and disease diagnosis. The core of those diagnostic systems are usually artificial intelligent algorithms like convolutional neural networks, in which thousands of images are needed for training. However, the available datasets of biomedical images, specially for X-ray angiography, are scarce. Therefore, we propose a toolbox for X-ray angiography images simulation to increase the number of available images as an alternative to data augmentation method for training artificial intelligence algorithms. The toolbox was developed with a set of functions to simulate complex vessel structures, as well as stenosis and aneurysms, in X-ray angiography images.