What is AWS HealthImaging?
AWS HealthImaging is a cloud-native, HIPAA-compliant service from AWS for storing, transforming, and analyzing medical imaging data. Healthcare facilities, radiology providers, and medical technology companies face the challenge of managing growing volumes of DICOM image data produced by CT, MRI, X-ray, and ultrasound, while ensuring fast diagnostic access with low latencies. AWS HealthImaging solves this problem with a fully managed cloud infrastructure that stores petabytes of image data and provides millisecond retrieval times.
Internally, HealthImaging uses the HTJ2K format (High Throughput JPEG 2000), which supports both lossless and lossy compression with continuous quality scaling. DICOM files are automatically converted and optimized during ingest without losing diagnostic information. The progressive loading approach allows medical viewers to initially load a low-resolution version of an image and successively load higher resolution levels, significantly reducing perceived load times for radiologists. Images are stored encrypted using HTTPS and AWS KMS keys.
Integration with the AWS AI ecosystem is a core element of HealthImaging: via the API, image data can be fed directly into Amazon SageMaker training and inference pipelines, greatly simplifying the development and operation of radiological AI models, for example for automated findings or anomaly detection. Amazon Comprehend Medical analyzes clinical texts and DICOM metadata for medical entities, diagnoses, and ICD-10 codes. For interoperability between facilities, HealthImaging supports DICOMweb standards (STOW-RS, WADO-RS, QIDO-RS).
innFactory supports healthcare organizations in migrating from legacy PACS systems to AWS HealthImaging, integrating with existing clinical IT systems, and building AI-powered image analysis pipelines on AWS. As an AWS Reseller, we accompany the entire project from compliance review to production go-live.
Typical Use Cases
Frequently Asked Questions
What is AWS HealthImaging?
AWS HealthImaging is a cloud-native service for storing and processing medical imaging data in DICOM format. It enables centralized management of radiology, pathology, and other medical image data with millisecond retrieval times, without needing to operate your own image archiving infrastructure (PACS).
What is the HTJ2K format and why does HealthImaging use it?
HTJ2K (High Throughput JPEG 2000) is a modern image compression format used internally by AWS HealthImaging. It provides lossless and lossy compression with continuous quality scaling, enables progressive loading (first low resolution, then successively higher), and is significantly faster to decode than classical JPEG 2000. This saves storage costs and enables fast transfer at diagnostic quality.
Is AWS HealthImaging HIPAA compliant?
Yes, AWS HealthImaging is HIPAA eligible. AWS provides Business Associate Agreements (BAA) for HealthImaging. Data is encrypted at rest and in transit. The service is suitable for processing Protected Health Information (PHI) when the appropriate compliance configurations are in place.
How does HealthImaging integrate with AI services?
AWS HealthImaging is directly integrated with Amazon SageMaker for training and inference of AI models on medical images. Amazon Comprehend Medical can analyze DICOM metadata and clinical texts. Through the HealthImaging API, images can be fed directly into SageMaker pipelines, for example for radiological AI diagnostic models.
How do I migrate an existing PACS to AWS HealthImaging?
AWS provides migration paths via the AWS DataSync service and direct DICOM STOW-RS interfaces. Existing PACS systems can be connected via DICOMweb-compatible connectors. AWS Professional Services and partners like innFactory support migration and integration with existing clinical workflows.