Amazon Web Services (AWS) announced the general availability of Amazon HealthLake, a HIPAA-eligible service for healthcare and life sciences organizations to ingest, store, query, and analyse their health data at scale.
Amazon HealthLake uses machine learning to understand and extract meaningful medical information from unstructured data, and then organises, indexes, and stores that information in chronological order. The result provides a holistic view of patient health.
The service leverages the Fast Healthcare Interoperability Resources (FHIR) industry standard format to further enable interoperability by facilitating the exchange of information across healthcare systems, pharmaceutical companies, clinical researchers, health insurers, patients, and more.
Amazon HealthLake is a new service that is part of AWS for Health, a comprehensive offering of AWS services and AWS Partner Network solutions used by thousands of healthcare and life sciences customers globally. AWS for Health provides proven and easily accessible capabilities that help organisations increase the pace of innovation, unlock the potential of health data, and develop more personalized approaches to therapeutic development and care.
As part of AWS for Health, Amazon HealthLake further facilitates customers’ application of analytics and machine learning on top of their newly normalised and structured data. Doing so enables customers to examine trends like disease progression at the individual or population health level over time, spot opportunities for early intervention, and deliver personalised medicine.
The healthcare industry is being transformed through the cloud and the utilisation of data, helping organisations uncover new insights and deliver improved patient care. Healthcare organisations are creating huge volumes of patient information every day, and the majority of this data is unstructured and contained in clinical notes, laboratory reports, insurance claims, medical images, recorded conversations, and graphs that are in different formats and spread across disparate systems. Before customers can derive a single insight (e.g. flag high-risk diabetic patients predicted to develop further complications), they have to aggregate, structure, and normalise this data. Then it must be tagged, indexed, and put in chronological order. This is a time-consuming and error-prone process.
Some healthcare organisations use optical character recognition and build rule-based tools to automate the process of transforming unstructured data and extracting clinical information (e.g. diagnoses, medications, and procedures). However, these options are often inaccurate. Even after organisations are able to aggregate and structure their data, they still need to build their own analytics and machine learning applications to reveal relationships in the data, discover trends, and make precise predictions. The cost and operational complexity of this work is prohibitive to most organizations. As a result, the vast majority of organisations cannot realise the full potential of their data to help improve the health of patients and communities.
Amazon HealthLake removes this heavy lifting by using highly accurate machine learning to automate the extraction and transformation of unstructured health data so organisations can apply advanced analytics and customised machine learning models to their information.
Using Amazon HealthLake, organisations can easily move their FHIR-formatted health data from on-premises systems to a secure data lake in the cloud. Amazon HealthLake uses specially tuned machine learning models that understand medical terminology to identify and tag each piece of clinical information. The service then enriches data with standardised labels (e.g. medications, conditions, diagnoses, etc.) so the data can be easily searched and analyzed. Amazon HealthLake also indexes events like patient visits into a timeline, giving medical professionals a holistic, chronological view of each patient’s medical history.
Also Read: Top 7 Data Visualisation Tools
Once this heavy lifting is completed, customers can apply analytics and machine learning on top of this newly normalised and structured data. For example, customers can apply analytics using Amazon QuickSight to understand patient and population-level trends, as well as build powerful machine learning models with Amazon SageMaker to help make accurate predictions about the progression of disease, the efficacy of clinical trials, the eligibility of insurance claims, and more.
Amazon HealthLake also stores data in the FHIR format to facilitate the exchange of information so that it is easy for organizations, researchers, and practitioners to collaborate and accelerate breakthroughs in treatments, deliver vaccines to market faster, and discover health trends in patient populations. Customers who do not already have data in the FHIR format can work with AWS Connector Partners, such as Diameter Health, InterSystems, Redox, and HealthLX, who have built validated Amazon HealthLake connectors to transform existing healthcare data into FHIR format and move it to Amazon HealthLake.
“More and more of our customers in the healthcare and life sciences space are looking to organize and make sense of their reams of data, but are finding this process challenging and cumbersome,” said Swami Sivasubramanian, Vice President of Amazon Machine Learning for AWS. “We built Amazon HealthLake to remove this heavy lifting for healthcare organizations so they can transform health data in the cloud in minutes and begin analysing that information securely at scale. Alongside AWS for Health, we’re excited about how Amazon HealthLake can help medical providers, health insurers, and pharmaceutical companies provide patients and populations with data-driven, personalised, and predictive care.”