PUBLISHED PAPERS #08.05
| Sholpan Zhumagulova, Orken Mamyrbayev, Dina Oralbekova, Kymbat Momynzhanova. ECG Classification Improvement Through Preprocessing and Adaptive Learning |
|---|
| Abstract. Electrocardiography (ECG) in 12 leads is a standard method for diagnosing cardiovascular diseases. In recent years, there has been significant progress in the use of deep learning methods for ECG analysis, which has improved the accuracy of automatic pathology classification. This paper examines methods for enhancing ECG classification quality that do not require changes to the model architecture. The main focus is on approaches such as signal preprocessing, self-adaptive learning, and the inclusion of patient metadata. Experimental results show that the proposed methods contribute to improved classification accuracy, especially when analyzing long recordings and combining data from different leads. The findings have practical significance for the development of intelligent diagnostic systems for cardiovascular diseases. |
| Keywords: ECG, deep learning, signal preprocessing, data preprocessing, self-adaptive learning |
 Download PDF |
| DOI: https://doi.org/10.30546/MaCoSEP2025.1113 |