Artificial intelligence to detect heart attacks?

Greek researchers have used online analytical processing (OLAP), a technique usually associated with financial and marketing analysis, to build the foundations for a heart attack calculator. Their model integrates 'lifestyle factors including depression, education, smoking, diet, and obesity, [which all] play a part in the risk of cardiovascular disease.' The research team said that 'their approach works much more quickly than conventional statistical analysis.' According to the leading scientist, 'Due to the ease of use of the methodology, a physician has the advantage of easily identifying high-risk patients by simply entering their personal data in the model.' A question remains: will this model be freely available? ...

Greek researchers have used online analytical processing (OLAP), a technique usually associated with financial and marketing analysis, to build the foundations for a heart attack calculator. Their model integrates 'lifestyle factors including depression, education, smoking, diet, and obesity, [which all] play a part in the risk of cardiovascular disease.' The research team said that 'their approach works much more quickly than conventional statistical analysis.' According to the leading scientist, 'due to the ease of use of the methodology, a physician has the advantage of easily identifying high-risk patients by simply entering their personal data in the model.' A question remains: will this model be freely available? ...

The research team has been led by Hara Kostakis of the Technological Educational Institute (TEI) Piraeus Research Centre (site in Greek). He worked withBasilis Boutsinas from the University of Patras and Demosthenes B. Panagiotakos of the Harokopio University in Athens, Greece. The fourth member of the team, Leo D. Kounis, doesn't come from a university. Kounis works at PartsInterlink, Spare Part Provider for American Cars located at Rhodopolis, Attica, Greece.

How did this team get its initial database to work with? "They obtained data for almost 1000 patients enrolled in the CARDIO 2000 study who had been hospitalised with the first symptoms of ACS, acute coronary syndrome. They recorded details of body mass index, family history, physical activity, high blood pressure, high cholesterol, and diabetes were recorded. They then matched the data against healthy individuals as a scientific control."

Instead of using statistical methods, the researchers chose to online analytical processing (OLAP), a technique developed in the early 1990s and which "was exploited primarily in industrial and commercial applications, for financial and marketing analysis. Fundamentally, OLAP provides a multidimensional view of data that allows patterns to be discerned in even the largest datasets that remain invisible even to the most expert user of spreadsheets. In a standard model, sales, purchases, pricing, customer base, and other economic measurements are used, Kostakis' colleagues at the University of Patras have adapted this system instead to accommodate the risk factors of heart disease."

This interesting research work has been published in a new scientific journal from the Inderscience group, the International Journal of Knowledge Engineering and Soft Data Paradigms (IJKESDP) under the name "A computational algorithm for the risk assessment of developing acute coronary syndromes, using online analytical process methodology" (Volume 1, Issue 1, Pages 85-99, 2009).

Here is the beginning of the abstract. "This paper investigates patterns in cardiovascular risk factors from a large population sample of cardiac patients and their matched controls. Various factors were taken into consideration and were used as inputs to effectively demonstrate online analytical process, OLAP methodology. OLAP is a new method that is used to explore the role of several risk factors in cardiovascular disease risk assessment. It equally serves as a means to extract knowledge from the investigated factors' levels."

Curiously, the full paper is not accessible from the abstract, but it is available from the Table of Contents of the first isue of this new journal. I can't give you a direct link to the the full article (PDF format, 15 pages, 152 KB), because it changes all the time.

Anyway, here is an excerpt from the conclusions of this paper. "This research work has introduced a computational algorithm for effectively addressing and optimising computational time and risk assessment. Moreover, the advantage of the particular OLAP model is the correct treatment of the dependencies between the time dimension and the other dimensions that are included in a model. The outcome of this study will benefit the health sector, as it will contribute to better understanding and consequently better prevention of cardiovascular disease."

Sources: Inderscience Publishers, December 18, 2008; and various websites

You'll find related stories by following the links below.

Newsletters

You have been successfully signed up. To sign up for more newsletters or to manage your account, visit the Newsletter Subscription Center.
See All
See All