Driving a unified AI experience at RSNA 2025: Microsoft Dragon Copilot expands to radiologists, transforming the reporting workflow

Every patient image tells a story, and radiologists play a critical role in unlocking insights from the pixels to inform downstream care. By delivering timely, accurate, and complete reports, radiology teams help bridge the gap between diagnosis and action, accelerating time to treatment and supporting better outcomes. Microsoft is deepening its commitment to healthcare by working closely with radiologists to deliver a unified AI experience for their reporting workflow. As radiologists face increasing demands for speed, accuracy, and efficiency in reporting, this is a pivotal moment for helping them unlock new levels of productivity and efficiency through generative, multimodal, and agentic AI.

At the 2025 Radiological Society of North America Conference and Annual Meeting (RSNA 2025), Microsoft is extending the value of its flagship AI clinical assistant, Microsoft Dragon Copilot, specifically for radiologists. Now in preview, Dragon Copilot integrates directly into familiar workflows to help radiology teams focus on their interpretations.*

const currentTheme =
localStorage.getItem(‘msxcmCurrentTheme’) ||
(window.matchMedia(‘(prefers-color-scheme: dark)’).matches ? ‘dark’ : ‘light’);

// Modify player theme based on localStorage value.
let options = {“autoplay”:false,”hideControls”:null,”language”:”en-us”,”loop”:false,”partnerName”:”cloud-blogs”,”poster”:”https://www.microsoft.com/en-us/industry/blog/wp-content/uploads/2025/11/Thumbnail-3.jpg”,”title”:””,”sources”:[{“src”:”https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/638700-DragonCopilotforradiology-0x1080-6439k”,”type”:”video/mp4″,”quality”:”HQ”},{“src”:”https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/638700-DragonCopilotforradiology-0x720-3266k”,”type”:”video/mp4″,”quality”:”HD”},{“src”:”https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/638700-DragonCopilotforradiology-0x540-2160k”,”type”:”video/mp4″,”quality”:”SD”},{“src”:”https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/638700-DragonCopilotforradiology-0x360-958k”,”type”:”video/mp4″,”quality”:”LO”}],”ccFiles”:[{“url”:”https://www.microsoft.com/en-us/industry/blog/wp-json/msxcm/v1/get-captions?url=https%3A%2F%2Fwww.microsoft.com%2Fcontent%2Fdam%2Fmicrosoft%2Fbade%2Fvideos%2Fproducts-and-services%2Fen-us%2Findustry%2F638700-dragoncopilotforradiology%2F638700-DragonCopilotforradiology_cc_en-us.ttml”,”locale”:”en-us”,”ccType”:”TTML”}]};

if (currentTheme) {
options.playButtonTheme = currentTheme;
}

document.addEventListener(‘DOMContentLoaded’, () => {
ump(“ump-69273a72d8574”, options);
});

Serving as a companion for PowerScribe One, Microsoft’s leading reporting solution and trusted by thousands of radiologists, it delivers a unified workflow that streamlines reporting, surfaces information, and automates tasks.

Delivering generative, multimodal, and agentic AI without disrupting existing workflows

Every day, radiologists face a relentless tide of images, data, and documentation—each with varying degrees of complexity and urgency. And every moment a radiologist spends wrestling with fragmented technology is a missed opportunity to make a difference when it matters most.

This challenge extends into the AI landscape where fragmentation occurs when tools address narrow problems but create extra steps and poor integration into broader clinical workflows. Yet, hope is on the horizon. With many radiologists relying on the proven capabilities of PowerScribe One, like speech recognition to accurately dictate their report content and generative AI for generating draft impressions, organizations are looking for ways to integrate AI without altering the entire reporting experience.

For radiologists, Dragon Copilot works with PowerScribe One to deliver generative, multimodal, and agentic AI in a secure, scalable, and extensible way. It brings cloud-native features and AI directly into the radiologist’s workflow. By integrating with PowerScribe One, Dragon Copilot can further streamline report creation without interrupting the radiologist’s interpretation.

As we embrace the next frontier of AI, we know that having cloud-based solutions that work seamlessly with our existing products and systems is paramount. Having Dragon Copilot as a companion for PowerScribe One gives me confidence that I can test and benefit from the latest AI advancements with minimal disruptions and distractions.

—Sean Cleary, MD, Vice Chair of Informatics for Imaging Sciences University of Rochester Medical Center

AI capabilities tailored for radiologists

Prior report summarization: Radiologists often rely on prior reports to provide essential context for interpreting current studies. Prior report summarization in Dragon Copilot distills relevant prior reports and associated metadata into concise bullets, helping improve speed and accuracy in diagnostics. It assists the radiologists in interpretation of the current exam by clearly highlighting findings from prior reports requiring follow-up or ongoing attention.

Chat with credible sources: During an interpretation, radiologists may have questions or need to do additional research on a specific topic. A chat experience in Dragon Copilot routes questions to the appropriate agent or plugin to deliver relevant, reliable responses backed by credible sources with patient context—helping radiologists work more efficiently and intelligently without having to toggle between windows and other applications.

Report optimization for billing and beyond: Dragon Copilot can use third-party AI insights from partners to check the report content to help radiologists improve accuracy and quality, helping to prevent downstream issues like claim denials. By surfacing important details and helpful reminders directly inside the reporting workflow, radiology teams can optimize their reports for billing and beyond.

AI draft report content: With Dragon Copilot, organizations can accelerate the path to intelligent, draft-first reporting by integrating self-developed and third-party multimodal AI models. Radiologists can receive AI-generated draft report content from image analysis, which they can then review, test, and validate.

Accelerating innovation with new models and a rich partner ecosystem

Many organizations are experimenting with testing, fine-tuning, and building AI models tailored to their specific needs. With the rise of multimodal AI models that can analyze medical imaging, genomics, clinical records, and more, customers and partners are leading the next wave of innovation and exploring new use cases.

Collaborating on this work allows us to assess models like CXRReportGen that support grounded report generation based on image analysis from chest X-rays. With these new capabilities comes the challenge of how to surface model output into clinical workflows in ways that can add value. This has given us the opportunity to collaborate on AI interoperability with Microsoft in concert with other partners like Epic and Sectra to test how multimodal AI can work across systems to optimize clinical workflows and improve patient care.

—Dr. Richard Bruce, Professor and Vice Chair of Informatics, Department of Radiology at the University of Wisconsin School of Medicine and Public Health

Microsoft Foundry offers a growing catalog of more than 50 first- and third-party healthcare AI models across modalities like radiology, pathology, dermatology, and protein analysis. Key models include MedImageInsight for image embedding, MedImageParse for segmentation, and CXRReportGen for drafting imaging reports.

Building on this foundation, Microsoft now provides premium versions of two proprietary AI models—CXRReportGen Premium and MedImageInsight Premium—trained on high-value datasets for superior accuracy and task-specific performance. These models allow developers and ISVs to create advanced imaging solutions without starting from scratch, supporting use cases such as image quality checks, image-to-image search, exam parameter classification, and metadata analysis.

Most importantly, our partners are also pushing the boundaries of what’s possible and offering radiology-specific AI apps and agents that can surface directly in Dragon Copilot.

• Lunit, a global leader in AI-powered cancer screening and diagnostics, brings advanced AI-powered image analysis for mammography exams to Dragon Copilot, allowing radiologists to receive real-time insights directly within their reporting workflow. Lunit’s comprehensive breast screening data assists radiologists in classifying and correlating abnormalities. By integrating Lunit’s algorithms, Dragon Copilot helps surface critical findings and optimize report accuracy without disrupting clinical interpretation—supporting diagnostic efficiency and better patient care.
• Zotec extends the value of Dragon Copilot by integrating revenue cycle management and billing intelligence directly into the radiologist’s workflow, helping ensure reports are complete, compliant, and satisfy quality measures before final sign-off. This seamless connection reduces the risk of missed billing information and the need for addenda, saving time and improving reimbursement accuracy. By making Zotec’s expertise available, Dragon Copilot empowers radiology teams to focus on clinical care while automating complex administrative tasks.

Lunit and Zotec are part of a broader ecosystem of partners bringing new AI innovations to Dragon Copilot customers. The collaborative innovation between Microsoft and its partners keeps solutions adaptable, secure, and tailored to the evolving needs of healthcare professionals.

Expanding what cloud, data, and AI can do for radiology

The power of our ecosystem extends beyond Dragon Copilot. At RSNA 2025, we’re extending that innovation across the broader Microsoft Azure ecosystem, showcasing partners who are transforming imaging workflows and data management in the cloud.

• Merge by Merative, from first click to final read, accelerates imaging workflows in the cloud.
• CitiusTech helps organizations consolidate their medical imaging on Azure with their digital imaging and communications in medicine (DICOM) migration suite.
• Qumulo allows for intelligent retention and lifecycle management of imaging data in one unified file system on Azure.
• Milvue delivers both automated X-ray measurements and pathology detection for chest and appendicular skeleton imaging, supporting both pediatric and adult patients.

Learn more about AI adoption in radiology

• See Dragon Copilot and our radiology solutions, including partner innovations, live at RSNA booth #1311.
• Not attending RSNA 2025? Contact your account executive to schedule a demo.
• For advice on how you can accelerate AI adoption in your organization, explore the 2025 AI in Healthcare Decision Brief.

---

*Dragon Copilot for radiology is currently in preview for PowerScribe One customers. 

The post Driving a unified AI experience at RSNA 2025: Microsoft Dragon Copilot expands to radiologists, transforming the reporting workflow appeared first on Microsoft Industry Blogs.

Source: Microsoft Industry Blog