Telemedicine: Building an application to monitor SPO2, index, heart rate, body temperature, supporting remote medical examination for medical centers

Abstract

In recent times, the utilization of technology has brought about a revolutionary shift in the delivery and accessibility of healthcare services. The advent of telemedicine allows medical services to be provided and accessed remotely, eliminating the need for in-person interactions between doctors and patients. Through the use of mobile devices, computers, and other technological platforms, healthcare professionals can conduct seminars, diagnose patients, provide advice, and offer care without the necessity of face-to-face meetings. Consequently, the predictive assessment of patients has become crucial in today's advanced society. In this research, I propose the utilization of Machine Learning techniques to design a telemedicine application model. The aim is to evaluate and analyze whether patients have contracted Covid-19 or not. Specifically, the ensemble learning algorithms of Extreme Gradient Boosting (XGBoost) are employed. Additionally, the Logistic Regression algorithm is incorporated to ensure the utilization of the most accurate data available. While continuously exploring and assessing different algorithms, other methods such as Random Forest, Decision Trees and Naive Bayes are also considered. Machine learning algorithms have the ability to analyze extensive patient data, including medical records, diagnostic images, and real-time physiological signals, to extract valuable insights. This assists healthcare professionals in making informed decisions. The thesis proposes the development of an application that visualizes research findings and supports users in accessing telemedicine applications more efficiently.