data architecture
How can a modern healthcare data structure benefit from AI? With the February 2026 release of several new AI charbots from the major AI vendors, a new level of AI capability has been reached. AI is now writing the next generation of AI instead of humans. However, AI’s ability to improve patient outcomes and experience is constrained by legacy data system in EHRs. Interoperability planning coupled with information models is one means by which Healthcare IT can evolve beyond siloed, function specific data architectures to ones that ingest data from data exchange partners in real time to support reasoning and autonomous action. The use of standards based data exchange and libraries of discoverable APIs can now supplement the functions of the siloed off the shelf EHR systems.
No longer does an IT department need a staff of 20 programmers to write a new app or an entire system. It can now be done in a matter of minutes by AI. However, the testing of the generated code is still a challenge. No healthcare organization will ever release untested code to a live operational system. AI testing of AI software will be a key field in the future.
The development of Implementation Guides for information exhcange is a key step to unlocking silos. Properly governed and tested AI apps can enable health systems to achieve reliable clinical data management. AI.
Artificial intelligence (AI) use in healthcare has moved from theoretical to transformative in just a few years. AI is helping us reimagine and reengineer every aspect of healthcare – from administrative burden to diagnostics and patient outcomes. AI is revealing new, previously unseen insights and discoveries. But this journey of learning to fly with wings of AI is fraught with danger. AI is known to hallucinate for reasons no one understands. AIs must watch out for other AIs. None the less, AI can ingest enterprize data and information models such as the FHIM, data standards such as FHIR, and standardized reference terminologies, and easily generate code to do whatever is needed.
“According to the Data Management Body of Knowledge (DMBOK), Data Architecture ‘includes specifications used to describe existing state, define data requirements, guide data integration, and control data assets as put forth in a data strategy.’ Data Architecture bridges business strategy and technical execution”.
Data Architecture can be synthesized into the following components:
• Data Architecture Outcomes: Models, definitions, and data flows on various levels, usually referred as Data Architecture artifacts.
• Data Architecture Activities: Forms, deploys, and fulfills Data Architecture intentions.
• Data Architecture Behaviors: Collaborations, mindsets, and skills among the various roles that affect the enterprise’s Data Architecture.”
Realizing the target state
Data is one the most important assets of an organization. The understanding and planning of data assets is crucial to the future success of any organization. Essential to realizing proper utilization of data, Data Architecture is the process of defining how data is stored, used, and processed. The Data Architect is responsible for defining the target state.The Data Architect breaks the subject down by going through 3 architectural processes:
-
1
Conceptual
Represents all business entities.
-
2
Logical
Represents the logic of how entities are related.
-
3
Physical
The actual data stored on the storage media, the realization of the data for a specific function (storage, history, integration, updates, search, reporting, analytics).