Abstract: During a disease outbreak or some natural or manmade disaster, it has been observed that several people suffer due to mismanagement and miscommunication between hospitals. This happens because most of the hospitals although are equipped with modern healthcare infrastructure, still rely on pen paper based record keeping of the patients. This not only causes wastage of resources but also complicates and increases the patient’s bills. Having paper based records although helps in keeping track of an individual patient, but it gives us no knowledge about the health of a community as a whole. Our project tries to solve the above problems by automating the healthcare systems by providing a one stop close loop solution. Using healthcare automation a patient can easily be registered to a hospital and doctor can keep track of his disease and treat him better. The patient also would be able to carry his reports and prescription on the go without any hustle. Location based emergency response service can help greatly in providing medical assistance to the victims in the nick of time. By creating Electronic Medical Record of every patient we can also determine patterns in the diseases caused in the patients over a period of time and can try to trace its source. Using K-Means clustering algorithm we can also predict what disease a patient would have by taking his blood report and symptoms as input and verifying them against the historical data stored in database. The project will consist of 3 modules, viz, Patient and Hospital Data Accumulation for Automation purpose, Location based emergency services, discovering disease patterns and identifying drug and medicine usage.
Keywords: Disease Prediction, Hadoop, KNN.