Predictive Analytics for Improving Physician Insights in Health Informatics

In this modern world everything is expected to be handy without taking risks and spending much time. The thought brought the invention of the patient monitoring system. The on-body sensors such as temperature sensor, pressure sensor, glucose monitoring sensor, ECG sensor, and SPO2 sensor are used to observe the physical conditions of the patient. The data collected through the sensors will be directed to the doctor's side. The ability of delivering this kind of sensitive information to the doctor's database has to be accurate enough to make the decision clearly by the doctor. Currently the research work on the body area network is mostly based on machine learning techniques for identifying the diseases. The proposed work uses a deep Convolutional Neural Network algorithm for health informatics to classify the data and predict the disease with high accuracy of 86.6%. By adding the layers in CNN, the accuracy is quite improved when compared to the SVM, NB, DT, RF algorithms for health informatics.