Over the past year, Amazon Web Services has made significant efforts to infiltrate the healthcare space, including partnering with Cerner and developing artificial intelligence for clinicians.
Today, AWS continues to leverage its technology and other resources to help hospitals and health systems improve operations and patient care. Shez Partovi, MD, worldwide head of healthcare, life sciences and genomics at AWS, and Patrick Combes, worldwide technical leader of healthcare and life sciences at AWS, have both been instrumental in the company's efforts to enter the healthcare industry.
Below, they detail what AWS is doing to promote interoperability among health systems as well as how its partnership with Cerner is unfolding.
Editor's note: Responses have been lightly edited for clarity and length.
Question: How can hospitals and other healthcare organizations leverage machine learning and analytics to help predict patient health events?
Dr. Shez Partovi: As the U.S. healthcare industry moves towards value-based care, artificial intelligence and machine learning, paired with data interoperability, will improve patient outcomes while driving operational efficiency to lower the overall cost of care. By enabling data liquidity securely and supporting healthcare providers with predictive machine learning models, clinicians will be able to seamlessly forecast clinical events, like strokes, cancer, or heart attacks, and intervene early with proactive personalized care and a superior patient experience.
A great example of the inroads being made in predicting patient health events is a machine learning model developed by Cerner and AWS that predicts Congestive Heart Failure up to 15 months before its clinical manifestation. Integrating this predictive tool into individual health records can support provider decision making in real-time. Future projects will look to improve pre-procedural decision and interventions for chest pain using these types of cardiovascular prediction models.
The INOVA Translational Medicine Institute (ITMI) is another great example. AWS architecture facilitates secure data storage and management, which enables ITMI researchers to develop predictive care and personalized treatments for newborns suffering from congenital disorders, as well as for patients of all ages with cancer-causing genetic mutations.
Q: What needs to be done to promote interoperability in healthcare?
Patrick Combes: Most EHRs do not travel with a patient outside the hospital walls. As a result, only a portion of a patient’s healthcare data is available to providers at any given patient visit, leading to a fragmented view of a patient’s health history. The industry is making significant strides toward reducing barriers to healthcare data interoperability but there is more work to be done.
One huge barrier is that most health and patient data is stored as unstructured medical text. Identifying and retrieving information in this format is a time-consuming manual process. Today, AI-enabled recognition and translation applications on dedicated smart devices transform data in multiple formats such as voice, image, scan or PDF into a common text format. Data in this universal, liquid form can be shared seamlessly between EHRs to paint a more comprehensive health picture for every patient. Supplying advanced analytics and machine learning tools with more robust and complete patient datasets enhances medical and scientific insights tied to patient outcomes in an accurate, scalable, secure, and timely manner.
Another challenge is breaking down information silos. Provider organizations are overcoming this barrier by gathering records and making them more easily transferable between all organizations. Change Healthcare is a perfect example of how using AWS to build clinical interoperability services for data exchange and record retrieval can improve the patient experience across multiple points of care. The technology delivers faster, more accurate and complete patient data to customers of Change Healthcare; helping them deliver continuous, quality care, while increasing efficiency and lowering costs.
Eliminating barriers and providing greater visibility into a patient’s medical history, interoperability not only provides a better patient experience at the doctor’s office, but it also empowers patients to take greater control over their entire healthcare journey.
For example, last year, Amazon Web Services (AWS) worked with Fred Hutchinson Research Center to create FHIR-enabled storage and APIs, facilitating better care coordination between oncologists and primary care providers. Fred Hutch used these APIs to create an application to provide patients with tools to support their care regimens, including appointment follow-up and engagements with multiple providers, and providing visibility into disease management. This digital-therapeutic approach helped improve patients’ quality of life, health outcomes, and overall experience.
Q: What plans do AWS and Cerner have coming up for 2020?
SP: Overall, Cerner’s work with AWS in 2020 will enhance patient and clinician experience and drive a new era of health system interoperability. The net effect will be to reduce workflow friction at the point-of-care for both patients and providers and deliver meaningful insights to advance quality while reducing cost of care.
By adding machine learning services like Amazon Lex, Amazon Polly, Amazon Transcribe, and Amazon Translate, developers will be able to create chatbots that give patients access to their personal health records, with the ability to ask questions about their medication, diagnoses, and medical conditions. These new predictive capabilities will help healthcare providers improve point-of-care documentation, patient safety, and operational efficiency. Cerner is also looking at how to address clinician burnout by leveraging voice processing and other AWS ML services to enable doctors and providers to spend less time filling out forms and more quality time with their patients.
Moreover, AWS is helping Cerner leverage machine learning and analytics to gain new clinical and business insights that can significantly improve patient care. Utilizing AWS’ ML, analytics, and Internet of Things services, Cerner will help advance patient care experience, improve health at the population level, and reduce per capita costs.
Cerner is taking advantage of AWS’ global cloud-based products to offer providers and healthcare organizations the ability to access and leverage clinical data in new ways to drive faster innovation, improve quality of care, and provide patients with more influence and control over their healthcare.
To achieve these goals, Cerner is standardizing its entire AI and ML environment on AWS to create innovative new predictive technology. For example, the Cerner Machine Learning Ecosystem, a new platform that was built using AWS tools, will enable data scientists to build, deploy, monitor, and manage ML models at scale. These models will allow Cerner to uncover new and more accurate predictive and digital diagnostic insights and help drive earlier health interventions.
Q: Working with Cerner, how do you ensure that patient data is safe?
PC: AWS provides tools and capabilities to enable healthcare organizations to improve their security and compliance posture and gain access to an environment built for the most security-sensitive organizations.
Healthcare providers, payers and IT professionals use AWS services to process, store, and transmit Protected Health Information and personally identifiable information. AWS enables covered entities and their business associates to leverage the secure AWS environment subject to the European Union’s General Data Protection Regulation’s requirements, U.S. Health Insurance Portability and Accountability Act, and HITRUST.
Understanding how to build healthcare applications on AWS means understanding the shared responsibility model. In the AWS Cloud, security is shared between AWS and the customer, meaning that certain elements of security — such as physical security of the underlying infrastructure — are the responsibility of AWS. Customers are still responsible for other aspects of security, such as the security measures used to protect applications, which is no different applications running in a traditional data center. AWS customers retain ownership and control of their content, along with the ability to encrypt it, protect it, move it and delete it in alignment with their organization’s security policies.