The study of spiral waves is very important for obtaining fundamental knowledge about the nature of arrhythmogenic processes. The visualization of vortex phenomena in natural conditions is equally important in this regard. One of the possible ways to get the appearance of a spiral wave is to read the mechanical phenomena of cardiac tissues surfaces, which are able to express the manifestations of vortex phenomena due to mechano-electrical feedback. In particular, this paper considers such an important phenomenon as the break up of a spiral wave into two small ones. Using a high-frequency camera and deep learning methods, presumably cases of spiral wave full and non-full break up were visually obtained in the first seconds of fibrillation recorded with the electrocardiogram. Visualization of these phenomena can contribute to a deeper understanding of arrhythmogenic processes and help in therapeutic purposes because they play an important role in fibrillation development.