Amazon And Google Are Finally Tackling One Of Healthcare’s Biggest Problems: Unstructured Health Data
Healthcare is a broad concept: clinical encounters with patients, population health data and trends, community health outcomes, pharmaceutical trials, etc., are just a few of the different siloes that entail the concept of “healthcare” as a whole. Even if just looking at an individual level— medical records, history of medications, previous lab testing and results, prior procedures—there are so many different siloes that entail an individual’s healthcare profile.
Given the value of each of these components individually, it’s intuitive to think that these different data sets should speak with each other and be viewed holistically when approaching patient care or population health decisions for a community.
However, this is not the reality. In actual practice, it is extremely challenging to reconcile these various silos to create a composite database in order to make better informed decisions. This is because, often, these different slices of information are in their own individual formats, running on their own proprietary software or system, and perhaps structured in different ways that cannot work together.
This is the exact problem that technology companies are now trying resolve, and Amazon and Google are leading the charge.
Recently, Amazon revealed more details about Amazon HealthLake, a product that can “securely store, transform, query, and analyze health data in minutes.” The product provides end-to-end services: “Extract meaning from unstructured data with natural language processing (NLP) for easy search and querying […] Make predictions with health data using Amazon SageMaker machine learning (ML) models and Amazon QuickSight analytics […] Create a complete and chronological view of patient health data including prescriptions, procedures, and diagnoses […] Support interoperable standards such as the Fast Healthcare Interoperability Resources (FHIR) format.”
Specifically, Amazon HealthLake is built to: convert structured and unstructured data (the site provides examples of lab reports, medical records, insurance claims, doctor’s notes, etc.) into FHIR format; help develop insights and useable meaning from that data in a searchable and indexed format; and promote even deeper predictions and data-driven decision making in conjunction with advanced tools such as Amazon SageMaker and Amazon QuickSight.
Julien Simon, Global AI & Machine Learning Evangelist for Amazon, explains in the AWS news blog: “The ability to store, transform, and analyze health data quickly and at any scale is critical in driving high-quality health decisions. In their daily practice, doctors need a complete chronological view of patient history to identify the best course of action. During an emergency, giving medical teams the right information at the right time can dramatically improve patient outcomes. Likewise, healthcare and life sciences researchers need high-quality, normalized data that they can analyze and build models with, to identify population health trends or drug trial recipients. Traditionally, most health data has been locked in unstructured text such as clinical notes, and stored in IT silos. Heterogeneous applications, infrastructure, and data formats have made it difficult for practitioners to access patient data, and extract insights from it. We built Amazon HealthLake to solve that problem.”
Google is also becoming a formidable player in this space, with its recent release of the Google Cloud Healthcare Data Engine. The initiative’s mantra is simple: “Empower healthcare and life sciences leaders to make decisions from disjointed healthcare data.”
As the company describes, “Healthcare Data Engine builds on and extends the core capabilities of the Google Cloud Healthcare API to make healthcare data more immediately useful by enabling an interoperable, longitudinal record of patient data.” This can be leveraged to “Make better real-time decisions around population health,” and for “resource utilization, optimizing clinical trials and accelerating research, identifying high-risk patients, and other critical needs with health insights.”
Most importantly, given Google’s vast experience and success in enterprise services, its tool is built for scalability: “Healthcare Data Engine is backed by Google Cloud’s highly scalable and secure HIPAA-compliant managed services and leverages Google’s Cloud Healthcare API and BigQuery for robust processing. Healthcare Data Engine brings the power of Google BigQuery’s analytics and AI to the healthcare industry, enabling healthcare organizations to process petabytes of their patient data. Scale quickly to meet the fluctuating needs across your systems and facilities to meet complex needs, like managing population health.”
Indeed, whether it’s patient care, population health initiatives, or really any other modality, healthcare entails the generation of billions of data points, especially in the modern setting where information technology and healthcare are so intertwined. If these products by Amazon, Google, and other potentially emerging companies can provide a better way to leverage these data points in an safe, secure, and effective manner, it may allow for more informed and data-based decision making for generations to come.