Optimized Deep Learning CNN for Efficient Classification of ECG Signals

IoT- based biomedical systems are essential to healthcare because they enable efficient data transfer for processing and the use of state-of-the-art sensors to identify physiological factors. It is well known that tools and sensors have been created to assess physiological factors like heart rate, ,body temperature, blood pressure, SPO2 levels, and blood sugar. These systems' ability to handle obtained data effectively is made possible by deep learning techniques, which is essential to their overall efficacy. Numerous neural network-based techniques are used to classify and analyse physical characteristics. In order to optimise the CNN for ECG classification, this section suggests a novel design that employs Particle Swarm Optimisation (PSO). A comparison is made between the proposed PSO-optimized CNN strategy and alternative methods, including Wavelet feature extraction and classifiers such as Convolution Neural Networks (CNNs), Artificial Neural Networks which is also represented as ANN, and Support Vector Machines which is also represented as SVM. This work presents a paradigm for the classification of ECG signals, which is the foundation for these comparisons. The proposed strategy is designed to improve healthcare outcomes by enhancing the efficiency and accuracy of biomedical data interpretation.