There is an emerging solution for standardizing healthcare data. FHIR utilizes a set of modular components, known as ‘Resources,’ which can be assembled into working systems that will facilitate data sharing within the EHR and mobile-based apps as well as cloud-based communications. Looking to the future, FHIR framework will be critical for implementation of AI-based technologies in the healthcare sector that utilize electronic data, just as DICOM and PACS have become critical for exchange of digital medical images.
Figure: FHIR-based EHR framework data preparations for machine learning
By focusing on interoperability of information and systems today, we can ensure that we end up in a better place in 10 years than where we are now. And so, everything around interoperability – around security, around identity management, differential privacy – is likely to be part of this future.
A short-term way some institutions are trying to make progress while many of these issues are being addressed is developing in-house algorithms on their own data to improve clinical care. While these algorithms will need to be updated on more diverse data and tested before they can be used in other institutions, they can lead to immediate benefits in their home institution. One example of this is the Flagler Hospital where they used their internal data to create algorithms to identify the best care pathways for managing inpatient pneumonia, sepsis, stroke, etc. Application of an AI methodology called topographic analysis, they were able to identify groups of patients who received the best care for a condition, care that resulted in better patient outcomes, lower length of hospital stay, and lower costs. Then, they updated their order sets to reflect the identified best pathway for each of these diseases. As such, they’re seeing improvement in patient outcomes and hospital length of stays. This is one of the many innovative ways that medical institutions are pushing the field forward while the data access issues are being sorted out.
