Колаборативен модел на информациски систем за здравствена заштита

Abstract

Recent trends in health care support systems are focused on developing patient-centric pervasive environments and the use of mobile devices and technologies in medical monitoring and health care systems. This is of particular importance for patients with chronic diseases who require 24-hour medical care. Collaborative health care system model, presented in this dissertation, allows monitoring of users’ health parameters and his physical activities. The proposed model uses a social network that allows communication between users with same or similar condition and exchange of their experiences. This model creates an opportunity for increasing users health care within their homes, 24-hour medical monitoring, on the one hand, and increased capacity of health institutions on the other hand, resulting with reduction of overall costs for consumers and healthcare institutions. The proposed model helps its users to actively participate in their health care and prevention, thereby providing an active life in accordance with their daily responsibilities at work, familly and friends. Recommendation algorithm, which is part of the social network of the proposed model, gives to the users recommendations for performing a specific activity that will improve their health. These recommendations are based on the users’ health condition, prior knowledge derived from users’ health history, and the knowledge derived from the medical histories of users with similar characteristics. The proposed recommendation algorithm is based on the dependence between the values of the health parameters (e.g. heart rate, blood pressure, arrhythmias) and the users’ physical activities (e.g. walking, running, biking). The basic idea is to find out which physical activities affect change (improvement) of the value of health parameters. This dependence continues to be used by the algorithm to recognize the same or similar health situations found in another user with similar characteristics. If there is information in the users’ history that after performing some physical activity their health condition has improved, the algorithm accepts this knowledge and proposes the activity to other users with similar health problems. Thus, this system recommendation affects on improving the health condition, thereby obtaining better life quality of the user. The aim of this doctoral thesis, what we want to achieve with the proposed model, is not only monitoring the health condition of the user. To a great extent, the benefits of the proposed model are building complex structures of health data that would be used for diagnosis, treatment and prevention of various diseases. Last but not least, to encourage users to lead an active life and thus to participate directly in their own health care.