A Cloud Edge Based Heart Disease Detection Using DenseNet Convoluted Radial Basis Neural Network for Diabetic Patients

Heart disease (HD) is a complex medical condition that has the potential to affect a vast number of people globally.The quick and precise identification of HD is crucial in healthcare, particularly in the cardiology field.In the pre-processing stage of the data mining process, a high-dimensional database is employed to classify HD.This unprocessed dataset contains redundant and inconsistent data, which expands the search space as well as data storage.Using deep-learning methods, the suggested research tries to recognize significant cardiac complaint prediction properties.This research proposed novel heart disease detection techniques by feature extraction and classification through the DL (Deep Learning) architectures.Data collection of 1 Lakh samples has carried out from Cleveland, UCI open-source repository, which has 74 features with a balanced instance rate.Here the input heart disease data has been pre-processed and segmented for filtering and edge normalization.The input image has been processed based on contrast-based histogram equalization (CHE) and segmented based on a threshold of the image.Then the segmented image was extracted to obtain the in-depth features and classifying the features using DenseNet with a Convoluted radial basis neural network.Several clinical measures are used to measure the risk contour in patients, which aid in early diagnosis.In the proposed model, various regularization methods are applied to avoid overfitting.On the dataset, the suggested model attains 72% sensitivity, 90% recall, 88% F-measure, 94% throughput, 92% training accuracy, and 95% testing accuracy.This is compared to other deep learning (DL) methods using a variety of performance metrics, demonstrating the effectiveness of the proposed approach.