Healthcare data solutions in Microsoft Fabric
Healthcare data solutions in Microsoft Fabric are industry solutions that organizations can deploy to a Microsoft Fabric workspace. After you install the solution in the workspace, you can deploy one or more of the composable capabilities. You can choose the capabilities that you need to build your own unique healthcare solution to unify data in Fabric, including:
A customizable data model to structure data for analytics and AI/ML modeling
Transformations to ingest your raw data and transform it into the data model
Analytical solutions to use the standardized data
The following image shows the current capabilities that are available with healthcare data solutions.
Healthcare data foundations
Use healthcare data foundations to set up your healthcare data estate to run solution capabilities and change it to structure the data for analytics and AI/ML modeling. The foundation defines the lakehouse architecture to progressively improve the structure and quality of data as it flows through each layer of the architecture. As the data becomes more structured, it's accessible through traditional SQL tooling. This structured data allows for exploratory analysis on various aspects of healthcare data, including clinical, financial (claims and benefits), and administrative data modules. The included pipelines normalize the FHIR data and enrich the unstructured clinical notes. Relational FHIR normalizes, transforms, and then harmonizes the FHIR data. The data in JSON format isn't conducive for analytics, so the pipelines use relational FHIR to convert the data into tabular data. For more information about healthcare data foundations, see Healthcare data foundations.
You must deploy this capability to use other capabilities of healthcare data solutions.
The following sections describe the transformation capabilities of healthcare data solutions in Microsoft Fabric.
Azure Health Data Services - Data export
Ingest FHIR data to Fabric OneLake from the Azure Health Data Services FHIR service. The data ingestion capability of healthcare data solutions uses Fast Healthcare Interoperability Resources (FHIR®) standards to bring clinical FHIR data to OneLake. It uses the $export operation to land raw NDJSON in the FHIR landing zone. As part of this ingestion, FHIR data is structured and you can use it through OneLake and across the Microsoft Cloud. For more information, see Overview of Azure Health Data Services - Data export.
DICOM data transformation
Ingest, store, and analyze Digital Imaging and Communications in Medicine (DICOM) medical imaging data to OneLake. The DICOM data transformation capability allows you to ingest, store, and manage medical imaging data, such as X-rays, computed tomography (CT) scans, and magnetic resonance imaging (MRI) scans in OneLake. This imaging data allows for collaboration, research and development, and AI innovation for health and life science use cases. For more information, see Overview of DICOM data transformation.
Combining the detailed information in the DICOM images with the FHIR clinical data accelerates time-to-value to help you conduct exploratory analysis and run large-scale imaging analytics and radiomics.
OMOP transformations
Prepare data for standardized analytics through Observational Medical Outcomes Partnership (OMOP) open community standards. This capability provides researchers in the OMOP community access to OneLake's expansive scale and the AI capabilities of the Fabric platform. People can use the provided notebooks to construct statistical models, conduct population distribution studies, and use Power BI reports to visually compare various interventions and their effects on patient outcomes. The OMOP analytics capability brings the OMOP Common Data Model (CDM) to Fabric and offers a start-to-finish solution for doing clinical research by using clinical data that originated in FHIR format. For more information, see Overview of OMOP transformations.
CMS claims data transformations (preview)
Use the claims data transformations pipeline to bring your Centers for Medicare & Medicaid Services Claim and Claim Line Feed (CMS CCLF) data to OneLake. For more information, see Overview of CMS claims data transformations (preview).
SDOH datasets - transformations (preview)
Use the SDOH datasets transformations pipeline to bring your social determinants of health (SDOH) datasets to OneLake. For more information, see Overview of SDOH datasets - Transformations (preview).
Discover and build cohorts (preview)
Explore multimodal data by using natural language queries and build cohorts for downstream research and AI innovation. For more information, see Overview of discover and build cohorts (preview).
The following sections describe the analytics capabilities of healthcare data solutions in Microsoft Fabric.
Dynamics 365 Customer Insights - Data preparation
Allow the connection between Dynamics 365 Customer Insights - Data and your OneLake instance in Fabric for creating patient or member segments for your outreach. This capability allows for the creation of patient or member lists for outreach purposes. The capability also flattens some FHIR resources so that you have easier access to them and so that you experience efficient patient segmentation for outreach initiatives. For more information, see Overview of Dynamics 365 Customer Insights - Data preparation.
Patient outreach analytics (preview)
Bring your marketing data from Dynamics 365 to OneLake, harmonize them with patient data in FHIR, and use Power BI templates to improve patient engagement. For more information, see Overview of patient outreach analytics (preview).
Care management analytics (preview)
Gain insights into high-risk and rising-risk patients, allowing for timely interventions with appropriate care plan actions. For more information, see Overview of care management analytics (preview).
Unstructured clinical notes enrichment (preview)
To enrich the clinical notes, you can add structure to the unstructured clinical notes by using Azure AI Text Analytics for health. This capability uses Azure AI Language’s Text Analytics for health service for data extraction and structuring, which enhances their analytical potential. When the system extracts key FHIR entities from unstructured clinical notes, you can create structured data from these clinical notes. Then, you can analyze the structured data to gain insights, predictions, and quality measures to improve patient health outcomes. For more information, see Overview of unstructured clinical notes enrichment (preview).