Big data analytics has bestowed the gift of personalization on a range of industries. For example, marketers can target consumers with data-fed recommendation engines like that on Amazon.com. Now, it’s positioned medicine on the brink of a revolution in individualized treatment. Precision medicine can enable healthcare professionals to prevent and treat illness with granularity down to a single patient’s genome. But some already quake at the thought of such personal data being stored in servers vulnerable to hackers or mishandlers.
In 2016, the American Heart Association announced a collaboration with Amazon Web Services Inc. to build out its Precision Medicine Platform. So far, the platform has enabled doctors and researchers to collaborate on data analytics initiatives much more briskly and effectively than before, according to Laura Stevens (pictured), AHA data scientist.
Stevens spoke with John Furrier (@furrier), host of theCUBE, SiliconANGLE Media’s mobile livestreaming studio, during the AWS re:Invent conference last November in Las Vegas, Nevada. She explained how data is changing medicine and addressed questions around securing it. (* Disclosure below.)
This week, theCUBE spotlights Laura Stevens in our Women in Tech feature.
The global precision medicine market is expected to reach $88.64 billion by 2022. It is a specialized domain that combines data on a patient’s genes, environment and lifestyle for a clear picture of his or her health. “That then results in prevention and treatment that’s catered to you as an individual rather than a one-size-fits-all approach,” Stevens said.
The AHA’s Institute for Precision Cardiovascular Medicine created the Precision Medicine Platform to advance research and treatment of this type. The platform is a data marketplace that houses and provides analytic tools that enable high-performance computing and sharing of personal data, clinical trial data, pharmaceutical data, hospital data, etc.
Busting compliance bottlenecks
Those are the types of data subject to strict compliance with the Health Insurance Portability and Accountability Act of 1996, known as HIPAA, and other laws; so strict that paranoia over compliance has historically strangled big data analytics in healthcare. The “willfully misunderstood” HIPAA is “used by everyone in healthcare as an excuse to avoid even the most beneficial data integration,” David Sable wrote in Forbes. Sable runs the Special Situations Life Sciences Fund and teaches entrepreneurship in biotechnology at Columbia University.
Various efforts are springing up to secure stored genetic information. Last year, global cybersecurity company Northrop Grumman Corp. released a white paper setting out a framework for securing precision medicine data. The company intends for the framework to aid the White House Precision Medicine Initiative and the National Institute of Standards and Technology frameworks.
“… Northrop Grumman views the security of genomic data as a vital factor in the success of precision medicine going forward,” according to Amy Caro, vice president of the health solutions business unit at Northrop Grumman.
For the AHA, the security built into the AWS cloud passes all crucial compliance tests, according to Stevens. “Even if you have data that you’d like to use, it’s sort of a walled garden behind your data so that it’s not accessible to people that don’t have access to the data, and it’s also HIPPA compliant. It meets the utmost secure standards of healthcare today,” she said.
Data for the greater good
The security concerns must be weighed against the potential benefits of precision medicine. The National Institutes of Health is building a database of genetic information to aid researchers in treating and preventing cancer and other diseases. It is aiming to collect data from 1 million Americans. Genetic information sets from larger demographics result in data applicable to larger, more diverse populations. Genetics-based medical research in the U.S. has had issues with ethnic over-representation, so those not of European heritage may be out of sync with its results.
The AHA’s Myresearchlegacy.org invites anyone to donate their genetic, health and lifestyle data to help researchers treat patients. Right now, researchers are conducting precision medicine studies on treating breast, pancreatic and other cancers. None of this would be possible without the advances in compute power and storage that have powered big data analytics and artificial intelligence in other fields.
“The combination of benefits from process optimization, the ongoing transformation of medical data collection along the analog to digital continuum, and the availability of cheap memory and processing power and coding talent make the evolution of precision medicine inevitable,” Sable wrote.
Of the 13 precision medicine grants the AHA and AWS made last year, one in particular illustrates this point quite clearly. David Kao, M.D., University of Colorado, Denver, is aggregating data from numerous clinical trials with the Precision Medicine Platform. “Trying to determine how those trials compare and what outcomes we can generate and research insights we can generate across multiple data sets is something that’s been challenging,” Stevens said.
The platform’s services, like Amazon Elastic MapReduce, or EMR, clusters with Apache Spark big data framework, speed up aggregation and analytics, Stevens explained. “To be able to implement things like serverless AI and artificial intelligence and machine learning on that data set is time consuming,” she said. “Having the power of an EMR cluster that is scalable makes that so much faster so that we can then answer our research questions faster and identify those insights and get them out in the world.”