A Cloud-Centric Application for Elderly Heart Disease Detection with Machine Learning and Confusion Matrix

Elderly individuals often face limited mobility and remote locations, which can hinder their access to specialist healthcare and result in missed diagnoses and poorer health outcomes. This paper addresses this issue by proposing a novel, cloud-centric application for Remote Patient Monitoring (RPM). The goal of this application is to enable elderly patients to track their vital signs from home using Internet of Things (IoT) sensors, leveraging machine learning for analysis. A high-performing Random Forest model is employed to analyze the data, detecting early signs of cardiovascular disease with an accuracy of 82.6%. Doctors can remotely monitor patient health data, which is integrated with electronic health records, facilitating timely follow-up care and personalized treatment recommendations. The method presents a user-centric approach that combines remote self-diagnosis with advanced technology to improve healthcare accessibility for the elderly. The current iteration focuses on heart disease detection, but future developments could expand the application to a broader range of health parameters. It is essential to note that this application serves as a complementary tool to professional medical advice, not a replacement. Clear communication about these limitations within the app is crucial. This research highlights the importance of doctor supervision and professional evaluation in conjunction with self-monitoring through the application.