AWS HealthImaging concepts - AWS HealthImaging

AWS HealthImaging concepts

The following terminology and concepts are central to your understanding and use of AWS HealthImaging.

Data store

A data store is a repository of medical imaging data that resides within a single AWS Region. An AWS account can have zero or many data stores. A data store has its own AWS KMS encryption key, so data in one data store can be physically and logically isolated from data in other data stores. Data stores support access control using IAM roles, permissions, and attribute-based access control.

For more information, see Managing data stores and Understanding storage tiers.

Image set

An image set is an AWS concept that defines an abstract grouping mechanism for optimizing related medical imaging data. When you import your DICOM P10 imaging data into an AWS HealthImaging data store, it is transformed into image sets comprised of metadata and image frames (pixel data). Importing DICOM P10 data results in image sets that contain DICOM metadata and image frames for one or more Service-Object Pair (SOP) instances in the same DICOM Series.

For more information, see Importing imaging data and Understanding image sets.


Metadata is the non-pixel attributes that exist within an image set. For DICOM, this includes patient demographics, procedure details, and other acquisition-specific parameters. AWS HealthImaging separates the image set into metadata and image frames (pixel data) so applications can access it quickly. This is helpful for image viewers, analytics, and AI/ML use cases that don't require pixel data. DICOM data normalizes at the Patient, Study, and Series levels, eliminating inconsistencies. This simplifies use of the data, increases safety, and improves access performance.

For more information, see Getting image set metadata and Metadata normalization.

Image frame

An image frame is the pixel data that exists within an image set to make up a 2D medical image. During import, AWS HealthImaging encodes all image frames in High-Throughput JPEG 2000 (HTJ2K). Therefore, image frames must be decoded prior to viewing.

For more information, see Getting image set pixel data and HTJ2K decoding libraries.