Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{180425,
        author = {Kapil Pal and Anita Yadav},
        title = {ECG Signal based Cardiac Arrhythmia Prediction using Hybrid Model Gated Recurrent Unit with Deep Learning based Transformer},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {1},
        pages = {1786-1796},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=180425},
        abstract = {Cardiac Arrhythmia is one of the critical diseases and it is very difficult to diagnose without any specialized cardiologists. ECG signals play an important role in detecting the abnormality in rhythm of heart beats; it means it also helps to identify the normal and abnormal condition of heartbeat. ECG machines and Holter Monitor are dedicated devices used to record and represent heart activity in the form of electrical signals during the treatment of arrhythmia patients. Different conditions of arrhythmia are Bradycardia, Tachycardia and Normal condition of heart, In Bradycardia State heart beats generally slower than Normal condition less than 60 beats per minute but in normal and healthy condition heart beats 60 to 100 times in 60 seconds and in tachycardia condition it just opposite of bradycardia heart beats more than 100 times in 60 seconds. These three different condition of arrhythmias is analyzed through the waves present in cardiac cycle and these waves are represented as P, Q, R, S, T and U wave each wave have its own specific property, count and duration so in this work we have QRS complex which leads the role to identify the number of beats in one minute but  mainly our hybrid model works on basis of count of R wave and these waves are extracted from raw ECG signal which recorded from real heart patients. In this research work to predict these different state of arrhythmia in heart patient we have developed a hybrid model a combination of Gated Recurrent Unit and Deep Learning based transformers and we have used four popular datasets and all datasets are freely available on physionet.org and in resultant of our work we found that our model is very lightweight and efficient to process large amount of ECG data and generate the results with better accuracy to predict different state of arrhythmia.},
        keywords = {Cardiac Arrhythmia Prediction, Deep Learning based Transformers, ECG Signals, Gated Recurrent Unit.},
        month = {June},
        }