In a step towards predictive healthcare, Max Healthcare Pvt. Ltd and Deakin University of Australia on Monday signed a partnership to use Big Data and predictive analytics to analyse its patient history records.

The data that will be analysed will include admissions, diagnoses and outcomes, images database and computerized registries.

The two stakeholders signed an agreement to collaborate for a project that will focus on data analytics for healthcare. Max Healthcare runs 10 hospital branches in India, out of which eight are in the National Capital Region. Big Data refers to massive amounts of digitized data that could be present on the Internet, government records and private digitized records of companies. Analysing this data requires machines with high processing capabilities and is important for a variety of reasons ranging from making sense of seemingly meaningless data to predictive analysis.

“The primary objective of the project will be to search through existing data sets for hidden patterns of both the predictable and preventable events in managing the healthcare of individuals," said Svetha Venkatesh, director, pattern recognition and data analytics, Deakin University. “This paradigm is novel, since it is capable of both hypothesis generation and testing, while being agnostic and unbiased to prior assumptions."

The pilot project from this pact will focus on heart disease, specifically on patients admitted with symptoms of acute myocardial infarction (stroke). The university chose heart disease for the first project as it is the biggest killer among non-communicable diseases not just in India, but around the world.

“There is a huge potential in India for Big Data analytics in healthcare. This is the future. The challenges in India are the same as the ones around the world such as clinical efficiencies," Venkatesh added.

The predictive model, which has been used in one hospital in Australia, will be based on building statistics from the hospital patient records from various hospital sources. A joint team will analyse the data, extract relevant features and build a validated machine learning model for specific prediction task.

“This association will be a landmark project and take healthcare management to a different level," said Sandeep Budhiraja, clinical director and director of the Institute of Internal Medicine, Max Healthcare.

Deakin and Max will share the final model as a prototype programme and the model, if successful, will be implemented at Max Healthcare on cases for over a year and the predictive accuracy would be monitored.

“This area has a tremendous potential, especially in finding the relationship between diseases and drug discovery. Some of the biggest challenges is to analyse the text written by doctors that most hospitals do not have, data for outpatient records and maintaining electronic records of data," said Anurag Agrawal, principal scientist at CSIR Institute of Genomics and Integrative Biology.

My Reads Logout