Release notes 2024: Azure Health Data Services
This article describes features, enhancements, and bug fixes released in 2024 for the FHIR® service, DICOM® service, and MedTech service in Azure Health Data Services.
August 2024
Azure Health Data Services
FHIR service
Import operation error handling
- The import operation returns a HTTP 400 error when a search parameter resource is ingested via the import process. This change is intended to prevent search parameters from being placed in an invalid state when ingested with an import operation.
- The import operation will return a HTTP 400 status code, as opposed to the previous HTTP 500 status code, in cases where configuration issues with the storage account occur. This update aims to improve error handling associated with managed identities during import operations.
July 2024
Azure Health Data Services
FHIR service
Allow dates in JSON data to be treated as strings in the Convert-Data operation
It's possible for dates supplied within JSON data to be returned in a different format than what was supplied. During deserialization of the JSON payload strings that are identified as dates get converted into .NET DateTime objects. These objects then get converted back to strings before going through the Liquid template engine. This conversion can cause the date value to be reformatted and represented in the local timezone of the FHIR service.
The coercion of strings to .NET DateTime objects can be disabled using the boolean parameter jsonDeserializationTreatDatesAsStrings. When set to true, the supplied data is treated as a string and won't be modified before being supplied to the Liquid engine.
Import Operation enhancement
The FHIR service now allows ingestion of data without specifying a version at the resource level. The order of resources is maintained using the lastUpdated value. This enhancement introduces the "allowNegativeVersions" flag. Setting flag true allows the FHIR service to assign negative versions for resource records with an explicit lastUpdated value and no version specified.
Bug Fixes
- Fixed inclusion of soft deleted resources when using _security:not search parameter When using the _security:not search parameter in search operations, IDs for soft-deleted resources were being included in the search results. We have fixed the issue so that soft-deleted resources are now excluded from search results.
- Exporting Data as SMART User Exporting data as a SMART user no longer requires write scopes. Previously, it was necessary to grant "write" privileges to a SMART user for exporting data, which implied higher privilege levels. To initiate an export job as a SMART user, ensure the user is a member of the FHIR export role in RBAC and requests the "read" SMART clinical scope. Updating Status Code from HTTP 500 to HTTP 400
- Updating Status Code from HTTP 500 to HTTP 400 During a patch operation, if the payload requested an update for a resource type other than parameter, an internal server error (HTTP 500) was initially thrown. This has been updated to throw an HTTP 400 error instead.
Performance enhancement
Query optimization is added when searching FHIR resources with a data range. This query optimization will help with efficient querying as one combined CTE is generated.
May 2024
Azure Health Data Services
FHIR service
Scaling enhancement to the Import operation
The scaling logic for import operations is improved, enabling multiple jobs to be executed in parallel. This change impacts audit logs for the import operation. Audit logs for individual import jobs have multiple rows, with each row corresponding to an internal processing job.
Bug fixes
- Fixed: HTTP status code for long-running requests. FHIR requests that take longer than 100 seconds to execute return an HTTP 408 status code instead of HTTP 500.
- Fixed: History request in bundle. Prior to the fix, history request in a bundle returned HTTP status code 404.
Stand-alone FHIR converter (preview)
The stand-alone FHIR converter API available for preview is decoupled from the FHIR service and packaged as a container (Docker) image. In addition to enabling you to convert data from the source of record to FHIR R4 bundles, the FHIR converter offers:
- Bidirectional data conversion from source of record to FHIR R4 bundles and back. For example, the FHIR converter can convert data from FHIR R4 format back to HL7v2 format.
- Improved experience for customization of default Liquid templates.
- Samples that demonstrate how to create an ETL (extract, transform, load) pipeline with Azure Data Factory (ADF).
To implement the FHIR converter container image, see the FHIR converter GitHub project.
April 2024
DICOM service
Enhanced Upsert operation
The enhanced Upsert operation enables you to upload a DICOM image to the server and seamlessly replace it if it already exists. Before this enhancement, users had to perform a Delete operation followed by a STOW-RS to achieve the same result. With the enhanced Upsert operation, managing DICOM images is more efficient and streamlined.
Expanded storage for required attributes
The DICOM service allows users to upload DICOM files up to 4 GB in size. No single DICOM file or combination of files in a single request is allowed to exceed this limit.
FHIR service
The bulk delete operation is generally available
The bulk delete operation allows deletion of FHIR resources across different levels, enabling healthcare organizations to comply with data retention policies while providing asynchronous processing capabilities. The benefits of the bulk delete operation are:
- Execute bulk delete at different levels: The bulk delete operation allows you to delete resources from the FHIR server asynchronously. You can execute bulk delete at different levels:
- System level: Enables deletion of FHIR resources across all resource types.
- Individual resource type: Allows deletion of specific FHIR resources.
- Customizable: Query parameters allow filtering of raw resources for targeted deletions.
- Async processing: The operation is asynchronous, providing a polling endpoint to track progress.
Learn more:
March 2024
DICOM service
Integration with Azure Data Lake Storage is generally available
Azure Data Lake Storage integration for the DICOM service in Azure Health Data Services is generally available. The DICOM service provides cloud-scale storage for medical imaging data using the DICOMweb standard. With the integration of Azure Data Lake Storage, organizations can enjoy full control over their imaging data and increased flexibility for accessing and working with that data through the Azure storage ecosystem and APIs.
By using Azure Data Lake Storage with the DICOM service, organizations are able to:
- Enable direct access to medical imaging data stored by the DICOM service using Azure storage APIs and DICOMweb APIs, providing more flexibility to access and work with the data.
- Open medical imaging data up to the entire ecosystem of tools for working with Azure storage, including AzCopy, Azure Storage Explorer, and the Data Movement library.
- Unlock new analytics and AI/ML scenarios by using services that natively integrate with Azure Data Lake Storage, including Azure Synapse, Azure Databricks, Azure Machine Learning, and Microsoft Fabric.
- Grant controls to manage storage permissions, access controls, tiers, and rules.
Learn more:
- Manage medical imaging data with the DICOM service and Azure Data Lake Storage
- Deploy the DICOM service with Azure Data Lake Storage
FHIR service
Bundle parallelization (GA)
Bundles are executed serially in FHIR service by default. To improve throughput with bundle calls, we enabled parallel processing.
Learn more:
Import operation accepts multiple resource types in single file
Import operation allowed to have resource type per input file in the request parameters. With this enhance capability, you can pass multiple resource types in single file.
Bug fixes
Fixed: Import operation ingests resources with the same resource type and lastUpdated field value. Before this change, resources executed in a batch with the same type and
lastUpdatedfield value wasn't ingested into the FHIR service. This bug fix addresses the issue. See PR#3768.Fixed: FHIR search with 3 or more custom search parameters. Before this fix, a FHIR search query at the root with three or more custom search parameters resulted in HTTP status code 504. See PR#3701.
Fixed: Improve performance for bundle processing. Updates to the task execution method, enabling bundle processing performance improvement. See PR#3727.
February 2024
FHIR service
Counting all versions of resources is enabled
The query parameter _summary=count and _count=0 can be added to the _history endpoint to get a count of all versioned resources. This count includes historical and soft deleted resources.
Revinclude search can reference all resources with wildcard character
The FHIR service supports wildcard searches with revinclude. Add *.* to the query parameter in a revinclude query to direct the FHIR service to reference all resources mapped to the source resource.
Bug fixes
Fixed: Improve FHIR query response time with performance enhancements. To improve performance, a missing modifier can be specified for a search parameter that is used for sort. See PR#3655.
Fixed: Import operation honors ingestion of non-sequential resource versions. Before this change, incremental mode in the
importoperation assumed versions are sequential integers. After this bug fix, versions can be ingested in nonsequential order. See PR#3685.
January 2024
DICOM service
Bulk update of files
The bulk update operation allows you to change imaging metadata for multiple files stored in the DICOM service. For example, bulk update enables you to modify DICOM attributes for one or more studies in a single, asynchronous operation. You can use an API to perform updates to patient demographics and avoid the cost of repeating time-consuming uploads.
Beyond the efficiency gains, the bulk update capability preserves a record of the changes in the change feed and persists the original, unmodified instances for future retrieval.
Learn more:
FHIR service
Selectable search parameters (preview)
The selectable search parameter capability available for preview allows you to customize and optimize searches on FHIR resources. The capability lets you choose which inbuilt search parameters to enable or disable for the FHIR service. By enabling only the search parameters you need, you can store more FHIR resources and potentially improve performance of FHIR search queries.
Learn more:
Integration of the FHIR service with Azure Active Directory B2C
Healthcare organizations can use the FHIR service in Azure Health Data Services with Azure Active Directory B2C (Azure AD B2C). Organizations gain a secure and convenient way to grant access to the FHIR service with fine-grained access control for different users or groups, without creating or comingling user accounts in their organization’s Microsoft Entra ID tenant. With this integration, organizations can:
- Use additional identity providers to authenticate and access FHIR resources with SMART on FHIR scopes.
- Manage and customize user access rights or permissions with SMART on FHIR scopes that support fine-grained access control, FHIR resource types and interactions, and a user’s underlying privileges.
Related content:
- Use Azure Active Directory B2C to grant access to the FHIR service
- Configure multiple service identity providers for the FHIR service
- Troubleshoot identity provider configuration for the FHIR service
- Enable SMART on FHIR for the FHIR service
- Sample: Azure ONC (g)(10) SMART on FHIR
Request up to 100 TB of storage
The FHIR service can store and exchange large amounts of health data, and each FHIR service instance has a storage limit of 4 TB by default. If you have more data, you can ask Microsoft to increase storage up to 100 TB for your FHIR service.
With more storage, organizations can handle large data sets to enable analytics scenarios. For example, you can use more storage to manage population health, conduct research, and gain new insights from health data. Plus, more storage enables Azure API for FHIR customers with high-volume data (greater than 4 TB) to migrate to the FHIR service in Azure Health Data Services.
To request storage greater than 4 TB, create a support request on the Azure portal and use the issue type Service and Subscription limit (quotas).
Note
Due to an issue with billing metrics for storage, customers who opt for more than 4 TB of storage capacity won't be billed for storage until the issue is resolved.