Kira Radinsky – Hypertension
A joint study between Kira Radinsky, the chief Scientist of Ebay in Israel, and the MK&M Big Data Science Institute , seeks to identify drugs that lower blood pressure
Hypertension affects roughly 20% of the world population. It is a leading cause of mortality and morbidity, including stroke, heart failure, coronary artery disease, and chronic kidney disease. Treating hypertension is extremely complex, as the numerous parameters involved have resulted in a vast array of possible drug combinations, especially considering that hypertension drugs fall into different drug classes, not to mention varieties. Even the experts cannot agree on specific treatment guidelines, optimal dosing or drug combination strategy, leading to additional confusion.
Enter the Kira Radinsky project, which attempts to use machine learning algorithms to achieve two goals. The first is to identify effective treatment choices for hypertension based on big data analysis of a large cohort of hypertensive patients. The second is to identify concomitant drugs not taken for hypertension, which may help to lower blood pressure.
Machine learning algorithms are a relatively new area of research in computer sciences and statistics, aiming to identify novel and valid patterns in data. Machine learning encompasses different modeling tools, which utilize computers to uncover “hidden insights” through learning from trends in large sets of data.
In this case, the data used by researchers was patient records from Maccabi Health Services. They took a sampling of records and identified patients whose blood pressure had been successfully lowered after receiving their first-ever drug treatment for hypertension, versus those whose treatment had been unsuccessful. They then used machine learning algorithms to see if they could find the patterns that would enable them to predict from another sampling of patients, whose treatment would have been successful in lowering their blood pressure.
Analysis of the results revealed that the majority of patients indeed needed drug combinations, rather than single drug treatment. They also found that when a beta blocker was given alone or in combination, it exhibits significantly higher success rates as compared to other drug classes. Most importantly, they learned that data science methodology using machine learning may be an effective means for repurposing medications already on the market, for new indications.