Automatic detection of infectious diseases

To prevent spreading of contagious diseases, an early detection is important. At the time an outbreak is diagnosed today, many people may already be infected. Thus, we need a system for detecting an epidemic outbreak at a very early stage. An ideal system should be able to prevent large outbreaks of infectious diseases, such as different strains of Influenza, Cholera, Plague, Ebola, Anthrax, and SARS. An infected person may be contagious before the first symptoms appear. This person can in the period of disease evolution, infect several associated citizen before consultation with a general practitioner. To prevent spreading of infections an early detection of contagion is very important.

Background

In this project, we will examine how we can use different kind of physiological indicators monitored by people with chronic diseases to detect an infection in a very early phase. We will start with people with diabetes who regularly monitor their blood glucose values and physical activity data using a communication enabled blood glucose meter. The basis for our approach is the assumption that people experience an increase in their blood glucose value when they get an infection. Monitoring and control of blood glucose levels is critical in the management of Type 1 and Type 2 diabetes, and the patients measure their blood glucose levels up to several times a day to minimize long-term complications. If values from all diabetes patients in a region were instantly accessible, it would provide a foundation for extended data analysis. Accumulations of high values in specific areas could be spotted, and the national health care service could monitor and act on such signals. In addition, the diabetics may be offered the possibility to compare the quality of their regulation against their peers.

We will also try to take into account measured health data (physiological values) from other groups of people suffering from chronic diseases.

Goals

• Develop personal health models for early detection of contagious diseases based on input from medical and physical sensors.
• Construct a prototype system that integrate the results and detect deviation from normal the condition

Problems

• How can physiological models for different groups of contagious diseases be developed?
• How can we develop a personal health model based on input from medical and physical sensors?
• How can health models be used for early detection of contagious diseases?

Methods

The project will be based on research methods from medicine, medical informatics, and computer science. The project will involve modeling, construction, collection, and examination of clinical data from different patient groups.
The project will be divided into 5 phases:
1. Discussion of
    a. Existing solutions for monitoring blood glucose and insulin values
    b. Monitoring of health data (physiological values) from patients suffering from other groups of chronic diseases
    c. Different models for construction of systems for detection of contagious diseases
    d. Monitoring of epidemiological values
2. Collection of measured data from:
   a. patients with chronic diseases who use different kind of health monitors
   b. patients with contagious diseases at UNN
3. Specification of experiments and prototype system
4. Experimental part in which a prototype is constructed and tested.
5. Clinical trials at UNN.

Project manager

Gunnar Hartvigsen (UiT/NST), Johan Gustav Bellika, (UiT/NST)

Project members from the partners

1 PhD position at UiT financed through TTL, NST

Researchers

Gunnar Hartvigsen (UiT/NST), Johan Gustav Bellika (UiT/NST), medical doctor

Project start/stop

Start 01.01 2007 stop 31/12 2010

Contribution to health care

The contribution to healthcare may come on two different levels; on a personal level and on population level. On the personal level the patients may be warned of emerging infections on an earlier stage of an infection and may initiate preventive actions. On a population level we may be able to give early warnings about disease outbreak, much earlier than what have been possible before.

Contribution to new industry

If the project succeeds in showing that outbreak may detected at an earlier stage, than what is possible today, it will create a demand for new equipment and software tools that support early detection of infections and disease outbreaks.

Contact info

e-mail: Gunnar Hartvigsen, mobile phone: +47 906 57 785
e-mail: Johan Gustav Bellika, mobile phone: +47 957 48 049

Tips en venn

Skriv ut