AI Summary & Key Takeaways
AI InsightComputer simulation results show that the proposed loss function is an attractive candidate for ECG data compression in the presence of Gaussian noise.
📌 3 Key Takeaways
Özet
ElectroCardioGram (ECG) is a graphical representation of the electrical activity that occurred during the heartbeat. It plays a significant role in the diagnosis and analysis of heart diseases. ECG signals must be recorded continuously for the effective detection and diagnosis of heart diseases. However, such records as it produces large amounts of data at a level that makes it difficult storage and transmission can also be impaired due to the ambient noise. Thanks to the reasons mentioned above, an efficient ECG data compression algorithm is required even in a noisy environment. This study proposes ε-insensitive quadratic loss based Support Vector Regression (ε-quadratic SVR) technique for the compression of ECG signals. There is a well-known relationship between loss functions and noise distributions. The proposed ε-insensitive quadratic loss function provides the optimal solution against Gaussian noise. Computer simulation results show that the proposed loss function is an attractive candidate for ECG data compression in the presence of Gaussian noise.