Predictive Medical Diagnostic Tool for Electronic Health Records (EHR)

  • Robert A. Sowah University of Ghana
Keywords: patient health records, machine learning, historical data, central platform, remote access


Healthcare management cannot be undertaken efficiently unless the necessary required data is available. It is essential to ensure that the most appropriate and accurate information for the management of patient health is continuously collected, processed, and provided in a required format and timely. Many avoidable shortcomings in the health sector that result in poor quality are due to the inaccessibility of data and mining necessary knowledge from it. The development and establishment of a robust health information system are essential for the sustainable management of patient healthcare records. This research work was carried out to develop a centralized database for the management of patient health records. This paper seeks to eliminate the traditional paper-based approach of keeping records in various hospitals of the health sector and also to create a central platform where patient health records can be accessed across multiple hospitals that run this system. The web application developed allows for the registration of patients, vital signs entry, doctor diagnoses, laboratory results upload, drug administration as well as service payment. The incorporation of seven (7) different machine learning algorithms in the web application assists the doctor in analyzing historical data on patients as well as their laboratory test results accurately. The mobile app also operates like the web application with an additional chat functionality that makes it possible for the patient to have remote access to medical help by initiating a chat session with a doctor without being present at the hospital.

Author Biography

Robert A. Sowah, University of Ghana

Computer Engineering