Atrial fibrillation (A-fib) is an irregular and often very rapid heart rhythm (arrhythmia) that can lead to blood clots in the heart. A-fib increases the risk of stroke, heart failure and other heart-related complications. During atrial fibrillation, the heart's upper chambers (the atria) beat chaotically and irregularly out of sync with the lower chambers (the ventricles) of the heart. Deep learning has become the most widely used approach for cardiac image segmentation in recent years. Segmentation and detection of human atria is crucial importance for precise diagnosis and treatment of atrial fibrillation, the most common cardiac arrhythmia. Recent developments in deep learning especially deep convolutional neural networks (CNN), improved the performance of medical image classification methods. However, training a deep CNN from scratch with medical images is complicated task as it requires large number of labelled data. In journal paper Epidemiology and natural history of atrial fibrillation, atrial fibrillation (AF) is the most common arrhythmia in the western world with an incidence of about 0.4% in men and 0.6% in women. In this paper, analysis and visualization of computed tomography of human heart in an explorative manner is carried out. UNet Architecture a standard form of CNN architecture which is used for segmentation of images.