Google has released a Cloud Healthcare API that it says will improve the ability of the medical sector to run machine learning on healthcare data sets.
The API released as a beta, allows Google Cloud Platform applications to ingest medical data sets, with open source adapters built into the API supporting interoperability with a range of new and established medical data standards.
Users of the Healthcare API, a beta release of which was announced this week, will be able to move their existing data formats onto Google Cloud Platform (GCP) allowing them to run data cleansing, analytics and machine learning on it.
The API taps emerging standards in healthcare data interoperability such as Fast Healthcare interoperability Resources (FHIR); an emerging standard for exchanging health records and improving interoperability across networks/systems.
(GCP is the latest to make a sustained push to deliver a FHIR-friend API: recent upgrades to Red Hat’s Fuse Online also included a preview, as cloud companies aim for a growing piece of the pie of the healthcare market, which – as cloud security improves – is increasingly looking at the value-added possible from running Machine Learning or AI on large datasets in the public cloud).
See also: Red Hat Rolls Out Array of New Integration Tools
The API includes support – via an open source “adapter” – for DICOMweb; a REST API used for storing querying and images made available under DICOM, the established standard for storing and exchanging medical images and their metadata across a wide range of modalities, including radiology, cardiology, ophthalmology, and dermatology. GCP said: “The DICOMweb support in Cloud Healthcare API allows existing imaging devices, PACS solutions, and viewers to interact with the Cloud Healthcare API. ”
Content from our partners
“This can be done either directly or via open source adapters designed to support existing DICOM DIMSE protocols. This allows customers to scalably store their medical imaging data and connect their data to powerful tools for analytics and machine learning,” GCP said.
Ilia Tulchinsky Head of Engineering and Joe Corkery, MD Head of Product at Google Cloud wrote in a blog that: “Our primary goal with Cloud Healthcare API has been to advance data interoperability by breaking down the data silos that exist within care systems. The API enables healthcare organizations to ingest and manage key data—and better understand that data through the application of analytics and machine learning in real time, at scale.”
Another essential communication modality in clinical systems is HL7v2 which facilitates a REST interface for the data input, sending and searching.
GCP said it has also integrated this communication module with an open source adapter, allowing it to run within Google Kubernetes Engine “to provide rapid provisioning, communicates over Cloud Pub/Sub to deliver horizontal scalability, and connects with Cloud VPN to enable transport security.”
Currently Google is working with healthcare providers and customers like the American Cancer Society who use the GCP’s machine learning to identify patterns in digital pathology images. While Corkery and Tulchinsky write that: “Stratus Medicine will review a serverless architecture for generating real-time clinical predictions using Cloud Healthcare API to feed FHIR and DICOM data into Cloud Machine Learning Engine.”
