December 11, 2020 - Amazon Web Services (AWS) recently announced a new HIPAA-eligible healthcare data lake service that will store, transform, and analyze data in the cloud. 

The product, dubbed Amazon HealthLake, aggregates an organization’s data across various silos into a centralized AWS data lake and automatically normalizes the information using machine learning. 

Specifically, the healthcare data lake identifies different pieces of critical information, tags them, and indexes events in a timeline with standardized labels. The data is then structured into Fast Healthcare Interoperability Resources (FHIR) industry standard format.

As a result, information can be easily and securely shared between health systems and with third-party applications. This allows providers to collaborate more effectively and ensures patients access to their personal medical information.

“With Amazon HealthLake, healthcare organizations can reduce the time it takes to transform health data in the cloud from weeks to minutes so that it can be analyzed securely, even at petabyte scale,” Swami Sivasubramanian, vice president of Amazon Machine Learning for AWS, said in the announcement,

“This completely reinvents what’s possible with healthcare and brings us that much closer to everyone’s goal of providing patients with more personalized and predictive treatment for individuals and across entire populations,” he continued. 

Healthcare organizations are creating large volumes of patient information every day to gain a full view of a patient’s health and applying analytics and machine learning to improve care, analyze population health trends, and improve operational efficiency.

But clinical data is complex and may be incomplete, incompatible, and unstructured. Therefore, data must be tagged, indexed, and structured in chronological order to make all of the data comprehensible, Amazon noted. 

Some healthcare organizations build tools to combat these challenges, but oftentimes these solutions fail because the data needs to be normalized across different systems and the tools don’t have the ability to do so. 

But with Amazon HealthLake, organizations such as healthcare systems, pharmaceutical companies, and clinical researchers can spot trends and abnormalities in health data so they can make more precise predictions about progression of a specific disease, the efficacy of clinical trials, and the health of patients and populations, Amazon emphasized. 

Organizations can also copy health data from on-premises systems to a secure data lake in the cloud and normalize every patient record across disparate formats automatically.

And customers can use other AWS analytics and machine learning services with Amazon HealthLake, including Amazon QuickSight, for interactive dashboards and Amazon SageMaker. 

This should streamline the building, training, and deployment of custom machine learning models, Amazon stated.

Currently, top companies like 3M, AstraZeneca, Bristol Myers Squibb, Cerner, GE Healthcare, Pfizer, and Philips have tapped AWS for cloud and machine learning services. 

Specifically, Cerner is focused on using AWS to help solve issues, enhance clinical and operational outcomes, resolve clinician burnout, and improve health equity. 

“Working alongside AWS, we are in a position to accelerate innovation in healthcare. That starts with data,” said Ryan Hamilton, SVP of population health at Cerner.

“We are excited about the launch of Amazon HealthLake and its potential to quickly ingest patient data from various diverse sources and transform the data to perform advanced analytics to unlock new insights and serve many of our initiatives across population health,” Stated Hamilton.

As machine learning becomes more mainstream, companies across different businesses are trying to apply it to their data to deliver meaningful business value. Healthcare is applying machine learning to improve operations and patient care.

At the beginning of October, AWS and Pittsburgh Health Data Services partnered to use machine learning in clinical care, including cancer and depression screenings.

One of the projects focused on machine learning techniques to help experts study breast cancer risks and understand what drives tumor growth. But AWS noted in the October announcement that this project is just the start when it comes to research collaboration to improve patient care.