Pittsburgh Health Data Alliance teams with AWS for new machine learning use cases
The Pittsburgh Health Data Alliance, which comprises researchers from UPMC, University of Pittsburgh and Carnegie Mellon University, is working with Amazon Web Services to advance innovation in areas such as oncology, precision medicine, voice-enabled AI and imaging.
WHY IT MATTERS
AWS, through a machine learning-focused research sponsorship is helping scientists at member institutions to speed their research and commercialization efforts across eight projects, PHDA officials say. These include:
- Technology to create an individual risk scores for individual cancer patients, helping physicians predict the course of their diseases and responses to treatment;
- An application that reads patients' verbal and visual cues to diagnose and treat mental health symptoms;
- Tools to reduce medical diagnostic errors by mining secure, anonymized data from a patient’s medical record.
Pitt researcher David Vorp, for instance, is using AWS technology to help improve the diagnosis and treatment of abdominal aortic aneurysms.
Clinicians currently can use only the simple measurements of an aneurysm’s diameter and growth rate to predict the risk of a rupture. But Vorp and his team are "developing an algorithm that will provide clinicians with an objective, predictive tool to guide surgical interventions before symptoms appear, improving patient outcomes," he said.
THE LARGER TREND
When Pittsburgh Health Data Alliance was first launched in 2015, the it was billed as a "one-of-a-kind" new partnership between academic heavyweights, seeking to capitalize on the troves data proliferating in electronic health records, diagnostic images, prescriptions, genomic profiles, insurance claims and consumer devices.
The goal, says UPMC CEO Jeffrey Romoff at the time was to build an "innovation ecosystem" for health data in western Pennsylvania and beyond.
Since then, PHDA has been focusing on an array of new computing and machine learning advances, such as Amazon's SageMaker and EC2, to more rapidly translate lab insights into treatments and services to improve health and wellness.
Other recent innovations from PHDA members include AI for the pharma supply-chain and a new telemedicine spinoff from UPMC, and robotic trauma care technology developed for the U.S. Department of Defense by Carnegie Mellon and Pitt.
ON THE RECORD
"We believe that machine learning can significantly accelerate the progress of medical research and help translate those advances into treatments and improved experiences for patients," said Swami Sivasubramanian, vice president of machine learning for AWS. "We are excited to bring our machine learning services and cloud computing resources to support the high-impact work being done at the PHDA."