5 Quick Things You Didn’t Know About the Microsoft Cloud for Healthcare

Last week, I and 15 other Microsoft Certified Trainers (MCTs) partook in an event at Microsoft surrounding their new Microsoft Cloud for Healthcare (MC4H) platform. Since it was announced earlier this year, there have been tons of advancements in Microsoft’s healthcare offerings that cover technical needs from providers to payers to research. I learned quite a bit at this event and so I wanted to quickly share some things I found interesting.

1: MC4H Encompasses Dynamics, the Power Platform, and Azure

When you first look at some of the articles about MC4H, it may be unclear as to which part of the Microsoft stack it belongs. Well, MC4H actually encompasses templates, solutions, APIs, and services that span Dynamics 365, the Power Platform, Microsoft 365, and Azure.

Here’s the current list of solutions in MC4H:

Note that some facets of MC4H require certain licensing (such at licenses to the Power Platform or Dynamics) whereas the Azure ones just require that you have an Azure Subscription and can be deployed independently.

2: Dynamics Could Replace your EMR and Other Medical Information Systems

Dynamics, Microsoft’s bespoke ERP/CRM platform, is extended through MC4H templates that allow organizations to easily build health-focused solutions, including patient outreach, care management, virtual visits, and more. This is all deployable through the Cloud Solution Center.

Once deployed, you’ll see the Healthcare Administration features in Dynamics. This is all available out of the box with the MC4H Dynamics package. I took this for a spin and it feels very much like a mature EMR platform like Epic or Cerner.

You’ll also be able to use some template applications in Power Apps. You can customize these apps to create your very own patient portal, hospital website, scheduling system, and more.

3: Dataverse Allows us to Quickly Derive Insights and Enrich our Data

Dataverse sits within the Power Platform and allows us to create connections to various data sources, which then flow through a shared data model (called the “Common Data Model”). This is especially useful if you’re using multiple diverse sources of data.

For example, your EMR platform (Dynamics, Cerner, Epic) often serves as a main data entry point for health organizations. This data can then be enriched with other data (IoMT, third-party data, -omics, etc.) to allow for deeper insights into your patients, research, and healthcare operations.

The Common Data Model is extended in MC4H with the Healthcare Data Model and allows us to have a semantic representation of our data that can flow between source systems and other tools such as Power Apps, Power BI, and the Azure Healthcare APIs.

4: Integrating with FHIR, DICOM, and MedTech is Easier on Azure

In the healthcare space, there are standards such as FHIR and DICOM that dictate how data can be transmitted, organized, and stored. In Azure, the Heath Data Services workspace, we can provision PaaS-based versions of these standards that are deployed as secure API endpoints in the cloud.

FHIR (Fast Healthcare Interoperability Resources, “/ˈfī(ə)r/”) is a global standard that allows for data exchange between systems using this common data model and web standards. For example, two EMR systems could communicate and share data through the common FHIR standard, which will standardize the data attributes and can be transferred through a FHIR API service.

The FHIR API is a HTTP RESTful service that has a standard set of functionality to create, retrieve, and update resources — and resources can be anything from patients to medications to procedures and more.

With X-rays, MRIs, CT scans, etc., DICOM (Digital Imaging and Communications in Medicine, “/dī-käm/”) is the standard format of choice. The DICOM API serves as a PaaS DICOMweb offering that will house study images and allows us to retrieve them with ease. It is the international standard to transmit, store, retrieve, print, process and display medical imaging information…and it’s now easily deployable on Azure.

For IoMT (“Internet of Medical Things”) data, the MedTech service enables us to get data from diverse medical devices and convert it into FHIR data. We can stream data through Event Hubs from virtually any device, normalize it (map the data points to standard features), and then record the data in our FHIR service. This makes it easier than ever to get medical device data into the patient record (or other systems) so that we can use it for diagnoses, research, AI, and more.

I’ll mentioned again that these services are cost-effective, scalable, and secure — all right out of the box.

5: There’s a Special NLP API for Healthcare

In addition to the suite of pre-trained AI models in Azure’s Cognitive Services (which include computer vision, speech recognition, anomaly detection, translation, and more), they have a specially-trained model for medical text. With entity extraction, relation extraction, entity linking, and assertion detection, this model can easily pick out medical phrases and jargon that would otherwise be missed by normal NLP. Many of the items it detects are related to an entry from the Unified Medical Language System (UMLS), which makes it extremely useful for structuring otherwise messy text data.

To learn more about the Text Analytics for Health API , click here.

Ready To Get Started?

As I hinted at the beginning of this post, I am a Microsoft Certified Trainer and my company, <Tuple>, specializes in architecting cloud solutions in healthcare and life sciences. Reach out and let’s discover how services in MC4H can help your organization grow to be more data-driven in your patient care and research!

Resources

Stay Curious…

--

--

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Colby T. Ford, PhD

Cloud genomics and AI guy and aspiring polymath. I am a recovering academic from machine learning and bioinformatics and I sometimes write things here.