August 22, 2013
Data Mining Study Explores Risk Factors Associated With Cardiovascular Disease
redOrbit Staff & Wire Reports - Your Universe Online
In the absence of other contributing factors, men between the ages of 48 and 60 are at moderate to severe risk of suffering a heart attack, according to new research appearing in the International Journal of Biomedical Engineering and Technology.By comparison, in the absence of other risk factors (such as alcohol consumption, smoking and high blood cholesterol), women over the age of 50 only face a mild cardiac risk, Subhagata Chattopadhyay of the Camellia Institute of Engineering in Kolkata and his colleagues wrote in their paper.
Cholesterol levels, alcohol intake and passive smoking continue to be the most important risk factors when it comes to mild, moderate and severe heart disease risk, the study authors discovered. Those conclusions are the result of a data mining exercise used to construct a risk model for heart attacks.
As part of their research, Chattopadhyay’s team used 300 real-world sample patient cases with varying levels of cardiac risk, culling the results based on twelve predisposing factors: age, gender, alcohol abuse, cholesterol level, smoking (active and passive), physical inactivity, obesity, diabetes, family history, and prior cardiac event.
Determining a person’s risk of experiencing an adverse health event such as a heart attack is difficult, as clinical history, symptoms and warning signs typically do not follow a set path. The risk level tends to vary from patient to patient, and the interpretations of the diagnoses rarely conform to the rules of epidemiology.
Using computational data mining techniques makes it possible for experts to extract useful information from real-world clinical data, the researchers explained. It could also help eliminate the subjectivity of clinical prognosis to some degree, allowing the epidemiology to work more precisely at the patient level.
Chattopadhyay’s study is not the first time these types of data mining techniques have been used. Previous studies have reportedly had issues in that data classification was based on decisions made by doctors. As a result, the records were exposed to the subjectivity the researchers were hoping to avoid in this study.
“The essence of this work essentially lies in the introduction of clustering techniques instead of purely statistical modeling, where the latter has its own limitations in 'data-model fitting' compared to the former that is more flexible,” Chattopadhyay explained.
“The reliability of the data used, should be checked, and this has been done in this work to increase its authenticity,” he added. “I reviewed several papers on epidemiological research, where I'm yet to see these methodologies used.”