Prediction of Cardiac Disease Based on Patient's Symptoms

This research work which initiated at an early detection of all the probable symptoms and signs which might further lead to detection of cardiac diseases using data collected from previous patients as well as data input received from the user at that particular time. Current scenario of health-care data used for surveillance are no longer simply a time building series of aggregate daily counts. Instead, a wealth of proposed spatial as well as temporal demographic, and symptom information is available at the data presented during the time of execution. Our proposed method incorporates all such information that is being used as a classification approach that compares recent healthcare data against data from that particular baseline distribution and hence classifies subgroups of the given data. In addition, the data sample data used is first tested against many types of classifiers and various other proposed test scores have been evaluated. Test best is further chosen to make predictions. This follows a prototype implementation using a python based data mining tool, Orange (version: 0.17.1). The database can be stored in a cloud to centralize it and make access easier.