AE-Obi-LSTM: An Efficient IOT Wearable Devices for Cardiovascular Disease Prediction

In the modern world, cardiovascular disease has spread widely. The need of detecting this deadly disease is become an integral part of everyone’s life. The factors that are affecting the CVD are high blood pressure, family history, age, stress, gender, body mass index and unhealthy lifestyle. There are several approaches are done by various researchers to predict the disease, but the former approach required certain improvement to predict the CVD with more accurate and easily. In the proposed method deep learning based cardiovascular disease prediction with high accuracy and low cost is implemented. The framework deals with the collection of data from the iot wearable devices and the pre-processing includes removing missing values and data normalization for eliminate the noise in the data. The Auto Encoder with Optimal bidirectional Long-Short Term Memory (AE-Obi-LSTM) is the most modern proposed method used for the extraction and prediction of cardiovascular disease with higher accuracy. The implementation of AE-Obi-LSTM is carried out in the python tool. The honey badger algorithm is used to optimize the weight of the Bi-LSTM. There are three datasets are used such as; Heart failure UCI dataset, Heart statlog Cleveland dataset and Heart statlog datahub dataset attained the accuracies 98.37%, 99% and 98.25%. From the resultant analysis, it clearly proves that the proposed system gives the best accuracy for predicting the cardiovascular disease.