Predicting Heart Attack using Big Data Analytics
In the U.S., it is observed that heart failure has become the prime reason for the cause of death and disability. The present methods to heart treatment are to some degree limited to the clinical assessments of the illness. Yale researchers, in the recent study, have leveraged big data analytics effectively to enhance the prediction of the survival rate of a patient with a heart failure. They were also able to describe the data-driven categories of various patients that were different in their response to commonly used therapies. The researchers stated that this novel approach described in detail in the Journal of the American Heart Association, will pave way for better care in this incurable chronic condition.
Led by two assistant professors in Yale’s Section of Cardiovascular Medicine, Dr. Tariq Ahmad and Dr. Nihar Desai, the research team evaluated health data from a huge registry of more than 40,000 patients. With the use of a statistical machine learning technique, researchers were able to calculate outcomes for the patients even after one year from diagnosis. The researchers also used cluster analysis methods to classify the patients into four clinically recognizable categories that responded differently to medical therapies. Nevertheless, big data methods outperformed the current measures of heart failure to better predict the risk factor.
As the final step, they used their findings to design a predictive online tool which could be integrated into the EHR systems. While applying these advanced analytic strategies this approach yielded valuable results to improve research and offer a personalized care for patients as well as enhanced intelligence to clinicians