AWS March 2024 Roundup
Noteworthy AWS blog posts and platform updates from March 2024 that benefit healthcare and life sciences applications:
Amazon GuardDuty EC2 Runtime Monitoring is now generally available (March 29, 2024)
With this new capability, GuardDuty adds visibility into EC2 instance-specific runtime behavior, such as file access, process execution, command line arguments, and network connections. Previously, GuardDuty runtime monitoring was only available for EKS and ECS clusters. While currently limited to instances running the Amazon Linux 2 or Amazon Linux 2023 operating systems, AWS notes that Ubuntu support will be available in the “near future.” A list of potential findings generated by GuardDuty is in the documentation.
Securing generative AI: data, compliance, and privacy considerations (March 27, 2024)
Whether you’re using public generative AI services or training your own models, this blog post discusses various scopes and “regulatory themes” such as protecting personal data, transparency, human oversight, classification, profiling, and AI safety.
AWS HealthImaging enables the import of large DICOM objects and high-throughput JPEG 2000 (HTJ2K) transfer syntaxes (March 8, 2024)
”AWS HealthImaging is a HIPAA-eligible service that empowers healthcare providers and their software partners to store, analyze, and share medical images at petabyte scale. Customers can now import and store large DICOM instances, such as those generated by digital pathology systems. With this release, HealthImaging also adds support for importing DICOM objects with pixel data encoded in any of the High-throughput JPEG 2000 (HTJ2K) transfer syntaxes that were recently added to the DICOM standard. Further, HealthImaging now supports import of data in the JPEG Lossless, Nonhierarchical (Process 14) transfer syntax.”
AWS announces Aurora MySQL integration with Amazon Bedrock for Generative AI (March 8, 2024)
”Amazon Aurora ML exposes models as SQL functions, allowing you to use standard SQL to pass data to models and return model output as query results. It provides simple, optimized, and secure integration between Aurora and AWS machine learning services without having to build custom integrations or move data around.”
AWS HealthLake expands search parameter support (March 7, 2024)
”AWS HealthLake added 28 new search parameters needed for payor workflows. HealthLake is a HIPAA-eligible service offering healthcare companies a complete view of individual and patient population health data using Fast Healthcare Interoperable Resources (FHIR) API based transactions to securely store and transform their data into a queryable format at petabyte scale. With this release, customers can now use over 1,400 search parameters to enrich their healthcare software applications.”
