Listen to the article
6 minutes

This audio is automatically generated. Please let us know if you have any comments.

Microsoft revealed a host of new artificial intelligence capabilities for healthcare organizations Thursday, including a product to help companies create their own AI agents.

The tech giant also announced basic models for medical imaging and a health data analytics platform, as well as details of its plans to create an AI documentation product for nurses.

Healthcare organizations have shown growing interest in adopting AI tools, even as some experts and lawmakers raise questions about their safe and equitable use. Tech companies say the products have the potential to help providers manage their workloads and alleviate burnout.

“We are at an inflection point. AI breakthroughs are changing, improving the way we work and live,” said Kees Hertogh, vice president of healthcare and life sciences product marketing, during a press briefing. “The integration of AI into healthcare has significantly improved patient care and reignited the joy of practicing medicine for clinicians. »

AI Agents in Healthcare

Microsoft’s agent service would allow companies to create AI tools with predefined models and data sources that could be used for appointment scheduling, clinical trial matching and patient triage , Hertogh said.

The service is currently in public preview, which allows broader access to the tools and allows organizations to provide feedback on the product, according to a spokesperson.

In one example, a doctor could ask an AI agent to find clinical trials for a 55-year-old patient with diabetes and interstitial lung disease.

The tools could also be aimed at patients. The Cleveland Clinic used the agent service to create tools for patients to ask health questions and navigate the health system’s services, according to a spokesperson.

The product allows organizations to create agents with healthcare-specific functionality using information from credible sources, which aims to improve safety, said Hadas Bitran, partner general manager of health AI at Microsoft Health and Life Sciences.

The tech company is also releasing additional application programming interfaces, or APIs, in private preview that can help verify model results, such as features to detect omissions or link responses to base data , she said.

Imaging models

Microsoft also revealed base models, or systems built on large data sets that can be used for a number of tasks, focused on medical imaging.

The models, developed with partners including Providence Health System and digital pathology company Paige.ai, would allow healthcare organizations to create their own AI tools without the heavy data and computing resources needed to create them from from zero.

Basic models include MedImageInsight, which enables image analysis that can be used to automatically send scans to specialists or flag abnormalities for review.

MedImageParse is intended for image segmentation, which could be used to segment tumors or describe organs at risk before radiotherapy for cancer patients.

The third model, CXRReportGen, creates reports based on chest X-rays, which Microsoft says could speed up image analysis and improve radiologists’ diagnostic accuracy.

Health data analytics platform

The tech giant said healthcare-specific data tools are now generally available in Microsoft Fabric, the company’s analytics product. The platform allows organizations to ingest, store and analyze health data.

“Analytics enrichments can help them improve reporting with details such as the geographic distribution of patients, their age, gender, etc., as well as the status of patient outcomes and satisfaction,” Hertogh said.

Organizations will also be able to use other types of data in public preview, such as conversational insights from Microsoft’s DAX Copilot AI documentation tool, public social determinants of health information, and CMS claims data. Also in a public preview, the tech giant released healthcare security models in Microsoft Purview, the company’s data security and governance offering.