Data Mining Reveals Heart Attack Risk Factors

A new study uses data mining to determine the most important risk factors in increasing the chances of a person suffering a myocardial infarction (MI).

Subhagata Chattopadhyay, MD, PhD, of the Camellia Institute of Engineering (Kolkata, India) conducted a study to determine the important predisposing risk factors of MI from a sample of 300 real-world cases with various levels of cardiac risk - mild, moderate, and severe. The patients had 12 known predisposing factors: age, gender, alcohol abuse, cholesterol level, smoking (active and passive), physical inactivity, obesity, diabetes, family history, and prior cardiac event. Dr. Chattopadhyay then built a risk model that revealed specific risk factors associated with heart attack risk.

The study revealed that according to the risk level, high blood cholesterol (HBC), intake of alcohol (IA), and Passive Smoking (PS) play the most crucial role on severe, moderate, and mild cardiac risks, respectively. The study also observed that male patients in the age group of 48-60 years (mean age 53.45 years) are more prone to suffer severe and moderate heart attack risk, while females over 50 years (mean 53.23 years) are affected mostly with mild risk. The study was published in the August 2013 issue of the International Journal of Biomedical Engineering and Technology.

“The essence of this work 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,” said Dr. Chattopadhyay, who is a MD with a PhD in Information Technology. “The reliability of the data used should be checked, and this has been done in this work to increase its authenticity. I reviewed several papers on epidemiological research, where I'm yet to see these methodologies used.”

The use of computational data mining techniques allows researchers to extract meaningful information from real-life clinical data, removing at least some aspect of the subjectivity of clinical prognosis and allowing the epidemiology to work at the patient level more precisely. But data mining approaches often have inherent problems in that the classification of the data for information retrieval is based on decision making learnt from examples set by doctors, and so they incorporate the very subjectivity that the mining wishes to avoid.

Related Links:
Camellia Institute of Engineering