Article Impact Level: HIGH Data Quality: STRONG Summary of Intelligent Systems, Blockchain, and Communication Technologies (pp. 544–553). Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-82377-0_ Kasem Khalil et al.
Points
- This study introduces a real-time heart attack detection method using Fast Fourier Transform and Artificial Neural Networks to convert ECG signals into the frequency domain for improved feature classification.
- The system is implemented on an Xilinx Virtex-7 FPGA and offers a low-power, high-efficiency solution suitable for embedded and portable medical devices. Its detection accuracy is 92.41 percent.
- The method detects heart attacks up to twice as fast as traditional techniques, significantly improving early intervention and patient outcomes in critical cardiovascular emergencies.
- The technology is designed for integration into wearable devices and enables continuous real-time monitoring, addressing the urgent need for rapid and accessible heart attack diagnosis.
- The approach may be extended to detect other health conditions, such as seizures or dementia, highlighting its broad potential impact in enhancing future medical diagnostics.
Summary
This study presents a novel real-time heart attack detection method using Fast Fourier Transform (FFT) and Artificial Neural Networks (ANN). The approach focuses on transforming electrocardiogram (ECG) signals from the time domain into the frequency domain through FFT for effective feature extraction. The extracted features are then classified using a well-trained ANN. Implemented on an Xilinx Virtex-7 FPGA using VHDL, the system provides an efficient, low-power solution suitable for portable and embedded medical devices. Experimental results demonstrate that the proposed method achieves an accuracy of 92.41%, outperforming traditional machine learning algorithms such as Logistic Regression, k-Nearest Neighbors, Random Forest, and Support Vector Machine.
The system is designed to detect heart attacks faster and more accurately than conventional methods. Compared to traditional detection techniques, the new technology is reported to be up to twice as fast, without sacrificing classification accuracy. This enhancement is particularly critical given that heart disease is the leading cause of death in the United States, with someone dying from a heart attack every 40 seconds. Faster detection enables earlier intervention, significantly improving patient outcomes and reducing mortality. The technology is designed to be embedded in wearable devices, making it accessible for real-time monitoring, which is critical for timely heart attack treatment.
The study highlights the potential of combining FFT and ANN for effective heart attack detection. By improving the speed and accuracy of heart attack diagnosis, this method could be a game-changer for health monitoring systems, especially in settings where time is critical. Future applications of this technology may extend to other health conditions, such as seizures or dementia, offering a broader impact on healthcare through faster, more efficient diagnosis.
Link to the article: https://link.springer.com/chapter/10.1007/978-3-031-82377-0_44
References Khalil, K., Khan, Md. R., Mohaidat, T., & Bayoumi, M. (2025). Enhanced heart attack detection with neural networks. In A. Abdelgawad, A. Jamil, & A. A. Hameed (Eds.), Intelligent Systems, Blockchain, and Communication Technologies (pp. 544–553). Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-82377-0_44
