Ignora esplorazione

Cloud migration for medical imaging data using Azure Health Data Services and IMS

Data di pubblicazione: 23 marzo, 2022

Sr. Program Manager, Health and Life Sciences, Microsoft

A technician uses touch screen monitor to examine performance test results in clean room of an Renault racing factory.

This blog post is co-authored by Vittorio Accomazzi, Chief Technical Officer (CTO) at IMS.

This blog is part of a series in collaboration with our partners and customers leveraging the newly announced Azure Health Data Services. Azure Health Data Services, a platform as a service (PaaS) offering designed to support Protected Health Information (PHI) in the cloud, is a new way of working with unified data—providing care teams with a platform to support both transactional and analytical workloads from the same data store and enabling cloud computing to transform how we develop and deliver AI across the healthcare ecosystem.

The first implementation of digital imaging techniques in clinical use started in the 1970s. Since then, the medical imaging industry has grown exponentially—over the last two and a half decades, there has been a significant development in image acquisition solutions, which has boosted image quality and adoption in different clinical applications. Healthcare is projected to deliver the greatest industry-specific CAGR of 36 percent out to 2025 (Global healthcare data is forecasted to reach 2.3 zettabytes* in this coming year alone) and medical imaging data represents approximately 80 – 90 percent of that growth1.

While the amount of data generated by the medical imaging industry has continued to grow, the solutions for storing and handling this data remain archaic and on-premises due to limited products with insufficient computing power, storage size, and continuously outdated hardware. In addition, the lack of interoperability of these on-premises systems with other types of clinical data solutions and increasing workloads within imaging departments resulted in a big struggle to achieve predictive diagnosis and improved outcomes for patients. Bringing health data into the cloud has been met with challenges ranging from concerns about the security and privacy of the data to a lack of understanding of the opportunities it opens.

For the most part, interoperability in the health industry has also been limited and focused on clinical data. However, other types of health data such as imaging, IoT, and unstructured data also play a critical role in getting a full view of the patient, thereby contributing to better patient diagnosis and care.

This is why Microsoft has released Azure Health Data Services which aims to support the combining clinical, imaging, and MedTech data in the cloud using global interoperability standards like Fast Healthcare Interoperability Resources (FHIR®) and Digital Information Communication in Medicine (DICOM). The DICOM service within Azure Health Data Services allows standards-based communication with any DICOMweb™ enabled systems such as medical imaging systems, vendor-neutral archives (VNAs), picture archiving, and communication systems (PACS), etc. The goal is to fully leverage the power of the cloud infrastructures for medical images, creating a service that is fast, highly reliable, scalable, and designed for security.

Within the DICOM service, QIDO, WADO, and STOW protocols support query, retrieve, and storage of DICOM objects, while custom tags allow for user-defined, searchable tags. You can also use DICOMcast as a single source to query for cross-domain scenarios. The DICOMcast injects DICOM metadata into the FHIR service, or FHIR server, allowing a single source of truth for both clinical data and imaging metadata.

Once imaging data is persisted in the cloud, there is also a need for seamless integration of workloads into the cloud with minimum disruption and without extra investment in devices and software. In order to enable customers currently relying on DICOM DIMSE to be able to smoothly adopt cloud-based imaging storage and solutions powered by our DICOM service.

IMS collaborated with Microsoft to leverage its cloud technologies for IMS to provide a solution for this challenge resulting in a powerful tool that migrates medical imaging data from legacy workstations to the cloud using Azure Health Data Services. IMS selected Microsoft Azure because it has the most comprehensive offering and active road map to support the transition of healthcare to the cloud.

Using CloudSync as a synchronization tool

It was apparent from the beginning that creating a simple protocol converter or gateway to push images from on-premises to the cloud was not an optimal solution: since the data will flow only in one direction (from a healthcare organization to the cloud for storage, archival or advanced analytics). With that, the institution would be missing most of the benefits, such as calling back the image set into the existing on-premises viewer after performing annotations, running cloud-enabled AI models, or advanced analytics. On the other hand, having access to prior imaging studies of the patients during the current visit also plays a vital role in validating abnormal conditions over time for better clinical outcomes.

To bridge this gap, IMS designed and developed CloudSync, which is a software-only DICOM device that actively synchronizes the on-premises archive (or multiple archives) with an Azure DICOMweb endpoint. CloudSync allows the data to flow both ways and furthermore allows the implementation of business logic for the proactive staging of patient historical imaging data for immediate access, thereby reducing the latency experienced by the user.

This synchronization allows integration of organizations’ existing on-prem solutions with Azure Health Data Services and machine learning environments so that they can store, archive, slice-and-dices their data for superior cohort management. With the possibility to conveniently connect to Microsoft Power BI and Azure Synapse Analytics through Azure Health Data Services, institutions can curate their datasets, develop and deploy models, monitor their performance, perform advanced analytics on Azure Machine Learning Pipeline and push results back into their clinical workflow.

Key features of CloudSync include:

  • Synchronize medical DICOM images from on-premises archives to the cloud using Azure Health Data Services: Enable collaboration among multiple on-prem devices by connecting all of them in one point for ease of access by everyone.
  • Eliminate network latency while fetching medical imaging data: Proactively push prior medical images of the patient from the cloud to the on-prem devices based on the patient’s schedule and have them ready during the patient’s visit.
  • Migrate imaging data from legacy workstations to the cloud: Enable seamless and effortless integration of on-premises imaging workstations with the cloud.

CloudSync workflow leveraging Azure Health Data Services.

CloudVue: A one-stop-shop for medical image viewing

To fully leverage the power of Azure, IMS also provides a zero-footprint diagnostic viewer called CloudVue. CloudVue allows users to safely review the data stored in the Azure DICOMweb archive on any device making it possible to access imaging data from anywhere. On top of the standard security mechanisms, CloudVue also encrypts the data during transmission.

In addition to providing the standard viewing features and tools of a web viewer, CloudVue also provides:

  • Organizations with the ability to grant granular secure access to specific medical imaging data for distribution such as authorizing users to access specific studies in the archive, and not the entire repository. Therefore, the organization can safely grant access to referring physicians and even patients.
  • The ability to deliver and improve AI workloads on Azure: CloudVue can store and handle data originating from AI predictions and track user behavior at the same time so it can determine if the prediction is correct. Therefore, CloudVue is implementing a positive feedback loop to monitor and improve AI over time.
  • Annotation capabilities for images used in AI modeling.

CloudVue is the perfect companion for CloudSync in allowing users to take advantage of storing data in Azure using Azure Health Data Services.

CloudVue workflow leveraging Azure Health Data Services.

Within the current healthcare market space, one of the biggest challenges facing radiologists, clinicians, and care teams while making the diagnosis is the easy availability of a complete history of the patient—while a radiologist might have access to the medical images, not being able to query and find a patient’s medical history, medications and other lab work in the same place, makes predictive diagnosis difficult and time-consuming. With Azure Health Data Services, all of this data can now be accessed together, and using DICOMcast, a new feature, clinical information can sit alongside metadata from medical images, making them searchable quickly. This technology used alongside IMS’s CloudSync and CloudVue can change how radiologists interact with medical images and give them the ability to use the data they have for diagnosis and research.

The beginning of next-gen medical imaging viewing

IMS has created a solution for the medical imaging industry that allows health organizations to take full advantage of Azure Health Data Services at their own pace while:

  • Avoiding any disruptions to the current workflow.
  • Maintaining the current investments in devices and software.

Do more with your data with Microsoft Cloud for Healthcare

With Azure Health Data Services, health organizations are empowered to transform their patient experience, discover new insights with the power of machine learning and AI, and manage PHI data with confidence. Enable your data for the future of healthcare innovation with Microsoft Cloud for Healthcare.

We look forward to being your partner as you build the future of health.


1“Driving Data Growth in Healthcare: Challenges and Opportunities for IT,” IDC