Heart attack detection is crucial for timely medical intervention and can significantly reduce mortality rates. Early and accurate detection enables prompt treatment, which is vital for improving patient outcomes and saving lives. This paper presents a novel method for heart attack detection, leveraging Fast Fourier Transform (FFT) for feature extraction and Artificial Neural Networks (ANN) for classification. The proposed method aims to provide accurate and real-time detection of heart attack events by transforming time-domain ECG signals into the frequency domain and utilizing a well-trained ANN to classify the patterns indicative of a heart attack. Implemented on a Xilinx Virtex-7 FPGA using VHDL, the system demonstrates efficient resource utilization and low power consumption, making it suitable for portable and embedded medical devices. Experimental results show that the proposed method achieves an accuracy of 92.41%, outperforming traditional methods such as Logistic Regression, k-Nearest Neighbors, Random Forest, and Support Vector Machine. This work underscores the potential of combining FFT and ANN for effective heart attack detection, offering a promising solution for real-time health monitoring systems.