The purpose of this blog is to provide an overview of healthcare interoperability and examine how various cloud offerings address industry needs as each relates to the access, exchange, and integration of data across the care continuum.
As the healthcare industry continues to evolve, technology will either be the driving force or at the forefront of industry innovations. With this in mind, there has been and still remains a strong emphasis on the stability of healthcare technology.
As with many other industries, major components of healthcare technology are centered on the reporting and exchange of data. In years past when healthcare institutions were sioloed and the sharing of medical related data was not a priority, the focus was less on integration and more on the development of solutions.
As healthcare is shifting to a more value based approach, healthcare systems will have to more effectively exchange information with internal and external partners to enhance patient care. Although solutions currently exist to support these integration points, the setup and hosting of such systems is a larger effort as it generally involves greater business and legal requirements.
In order to support easier transmission of data and a more ad hoc approach to data access, standardization and an API strategy is vital to address these industry needs and challenges. Either it is promoting interoperability between healthcare systems or advocating for enhanced machine learning technologies, it is evident that the industry would benefit from cloud offerings that follow industry standards and practices.
HL7 - FHIR:
Although there are many published standards in the industry, HL7’s FHIR standard is a leader in healthcare interoperability. FHIR, which stands for Fast Healthcare Interoperability Resources, is HL7’s latest standard and is focused on formatting various data points together to model meaningful objects.
This concept of standardizing data supports interoperability because it makes for more efficient data exchange by taking advantage of patterns like Representational State Transfer (REST) software architecture. As with any other environment, the development of these APIs is not the challenge, rather scalability and proper security measures is the point of concern. Vendors like CareCloud, an electronic medical record distributor, are turning to cloud providers to be able to more efficiently support the sharing of data and start leveraging the cloud’s FHIR data hosting services.
As CareCloud’s chief technology officer said,
“Through our technology partnerships with Google Cloud, we’re able to leverage data transfer technology and make it readily available.”
This demonstrates that these industry standard driven capabilities are seen as impactful and worth the investment for both the solution vendors and cloud platforms.
Another pronounced standard within healthcare technology is DICOM, which is an abbreviation for Digital Imaging and Communications in Medicine. This standard promotes proper storing, transmitting, and printing of medical images.
One of the core technologies in medical imaging informatics that rely on this standard is the PACS solution, picture archiving and communication system. Traditionally, these solutions would handle much of the workflow as it relates to radiologist reading and reporting on imaging studies, including the management of the underlying data.
A major trend in the industry is taking the ‘A’ out of PACS, which refers to transferring the responsibilities of managing the archiving and storage of images from the PACS vendors back to the consumers, being the hospitals or healthcare systems. Although this introduces the need for other technologies to manage the storage of these images, it opens the door for digital innovators to create meaningful solutions which point to this centralized data store.
One example of this is the fruition of zero-footprint viewers which allow for easier access of medical images from a browser or mobile device, something that the traditional PACS could not support. This type of innovation paired with the emergence of machine learning techniques in the cloud will make APIs that support the sharing of DICOM images invaluable to both vendors and healthcare systems.
The APIs will prove that the industry will quickly become dependent on scalable cloud technologies to support this interoperability at enterprise scale.
As Ilia Tulchinsky and Joe Corkery MD, leaders in Google Cloud’s Healthcare and Life Sciences division, wrote in a recent blog post…
From the beginning, our primary goal with Cloud Healthcare API has been to advance data interoperability by breaking down the data silos that exist within care systems. The API enables healthcare organizations to ingest and manage key data—and better understand that data through the application of analytics and machine learning in real time, at scaleIlia Tulchinsky and Joe Corkery Google Cloud Healthcare and Life Sciences division
This does not only highlight Google’s endeavours to advocate for proper healthcare interoperability but also the ability to perform machine learning in a scalable and productionzed fashion.
It is evident that Google Cloud has exciting cloud technologies and a goal to provide the healthcare industry easier access to data but also to promote the use of their advanced machine learning capabilities. Their wide range of cloud offerings support this claim and positions them in a more competitive part of healthcare’s cloud computing market.
A core offering of Google Cloud Healthcare API is around hosting data stores and supporting access with a RESTful API. The three types of data stores are; HL7v2, FHIR, and DICOM.
As with any other effort on Google Cloud, it all starts with the creation of a project and from there it is dependent on the business needs. In relation to a data store in the Healthcare API, it will reside within a dataset. A dataset that can be called from various points of clinical care efficiently by making requests to the API.
One dataset is not bound to a specific data store, rather you can have three varieties of stores in a single dataset and all with a different number of instances. This makes it very easy to organize data and manage access to each data store. For example, an academic medical center may want to define a dataset with separate access permissions which contains de-identified data for a research study. This would allow researchers to have access to clinical data in a secure but accessible manner.
In an effort to provide patients with better access to their clinical data, the United States Department of Health and Human Services is promoting interoperability by implementing new rules. Although the common theme is to encourage secure exchange of healthcare data, the most recent regulations focus on the development of a Patient API. Requiring hospitals to offer patients access to their personal health data through an API will increase transparency and drive digital innovative solutions aimed at enhancing the patient experience outside of the doctor's office. Another important ruling that came from this recent news is the fact that payers will have to make their provider directories open to the public. This aligns with the patient centric API approach, because it will allow patients to determine which providers are within their network and consider providers that may be the best fit for their healthcare needs. Given the emerging demand, a variety of software development opportunities will materialize and rely upon industry standards, like HL7 FHIR.
Either it is supporting the creation of APIs to support data exchange between care systems or helping to decompose PACS to promote more accessibility to medical images, the standard driven cloud technologies will have a large impact on the industry.
This paired with the centralization of data in a scalable cloud environment will complement machine learning innovations and enhance delivery of care across the clinical domain because of the vast compute power cloud vendors support.