| Abstrakt: |
One important category of cardiovascular disorders is arrhythmias, and early identification and diagnosis are essential to averting high-risk incidents like sudden cardiac death. Even though automated arrhythmia identification entrenched on the electrocardiogram (ECG) has garnered interest, the static characteristics utilized in conventional approaches are unable to sufficiently characterize the many mild variations of the ECG, leading to the omission of important but weak infectious information. While characteristics generated from deep learning (DL) hold promise for arrhythmia categorization, their interpretability remains a challenge. In this study, we propose heartbeat dynamics as a unique and effective interpretable characteristic for arrhythmia classification. It simulates geomorphologic swap in the heartbeat, is more responsive to subtle fluctuations in the pulse, and represents underlying dynamical changes at the electrophysiological level throughout the cardiac cycle. An irregular heartbeat, or cardiac arrhythmia, constitutes of the common heart conditions. It happens when your heart beats quickly, slowly, or irregularly. Cardiac arrhythmias are common in most people. There are many levels of cardiac arrhythmias, from moderate to severe. While most arrhythmias are benign and unimportant, some can be extremely serious and even fatal. Many of them, though not very hazardous, might induce symptoms that could be rather irritating in your day-to-day life. [ABSTRACT FROM AUTHOR] |
