Microsoft publicizes latest health-care AI tools

Microsoft on Thursday announced latest health-care data and artificial intelligence tools, including a set of medical imaging models, a health-care agent service and an automatic documentation solution for nurses. 

The tools aim to assist health-care organizations construct AI applications quicker and save clinicians time on administrative tasks, a significant explanation for industry burnout. Nurses spend as much as 41% of their time on documentation, based on a report from the Office of the Surgeon General. 

“By integrating AI into health care, our goal is to cut back the strain on medical staff, foster the collective health team collaboration, enhance the general efficiency of healthcare systems across the country,” Mary Varghese Presti, vp of portfolio evolution and incubation at Microsoft Health and Life Sciences, said in a prerecorded briefing with reporters. 

The brand new tools are the most recent example of Microsoft’s efforts to ascertain itself as a pacesetter in health-care AI. Last October, the corporate unveiled a series of health features across its Azure cloud and Fabric analytics platform. It also acquired Nuance Communications, which offers speech-to-text AI solutions for health care and other sectors, in a $16 billion deal in 2021.

Lots of the solutions Microsoft announced on Thursday are within the early stages of development or only available in preview. Health-care organizations will test and validate them before the corporate rolls them out more broadly. Microsoft declined to share what these latest tools will cost.

Health-care AI models 

Roughly 80% of hospital and health system visits include an imaging exam because doctors often depend on images to assist treat patients.

Microsoft is launching a set of open-source multimodal AI models that may analyze data types beyond just text, corresponding to medical images, clinical records and genomic data. Health-care organizations can use the models to construct latest applications and tools.

For instance, digitizing a single pathology slide can require greater than a gigabyte of storage, so many existing AI pathology models have trained on small pieces of slides at a time. Microsoft and Windfall Health & Services built a whole-slide model that improves on mutation prediction and cancer subtyping, based on a paper published within the peer-reviewed journal Nature.

Now, health systems can construct on it and fine-tune it to satisfy their needs. 

“Getting a whole-slide foundation model for pathology has been a challenge previously … and now we’re actually in a position to do it,” Sara Vaezy, chief strategy and digital officer at Windfall, told CNBC in an interview. “It was really form of a game changer.” 

The models can be found within the model catalog inside Azure AI Studio, which serves as Microsoft’s generative AI development hub. 

Health-care agent service

Microsoft also announced a latest way for health systems to construct AI agents.

AI agents vary in complexity, but they will help users answer questions, automate processes and perform specific tasks. 

Through Microsoft Copilot Studio, these organizations can create agents equipped with health-care-specific safeguards. When a solution comprises a reference to clinical evidence, for example, the source is shown, and a note indicates if the reply is AI-generated. Fabrications and omissions are also flagged, Microsoft said. 

For instance, a health-care organization could construct an AI agent to assist doctors discover relevant clinical trials for a patient. Microsoft said a physician could type the query, “What clinical trials for a male 55-year-old with diabetes and interstitial lung disease?” and receive a listing of potential options. It could save the doctor the effort and time of finding each trial. 

AI agents that will help patients answer basic questions have been popular among the many health systems which have already begun testing the service, Hadas Bitran, general manager of health AI at Microsoft Health and Life Sciences, said in a Q&A with reporters. Agents that will help doctors answer questions on recent guidelines and patients’ history are also common, she added.

Microsoft’s health-care agent service is offered in a preview capability starting Thursday.

Bringing automated documentation to nurses

In August, Microsoft announced that the following phase of its partnership with Epic Systems can be dedicated to constructing an AI-powered documentation tool for nurses, and the corporate detailed those plans on Thursday. 

Epic is a health-care software vendor that houses the electronic health records of greater than 280 million people within the U.S. It has a yearslong relationship with Microsoft. 

Microsoft’s Nuance already offers an automatic documentation tool for doctors called DAX Copilot, which it unveiled last yr. It allows doctors to consensually record their visits with patients, and AI routinely transforms them into clinical notes and summaries.

Ideally, this implies doctors haven’t got to spend time typing out these notes themselves each time they see a patient. 

The technology has exploded in popularity this yr. Nuance announced that DAX Copilot was generally available inside Epic’s electronic health record in January – a coveted stamp of approval throughout the health-care industry. Integrating a tool like DAX Copilot directly into doctors’ EHR workflow means they will not need to change apps to access it, which helps save time and reduces administrative workload.

But to date, DAX Copilot has only been available to doctors. Microsoft said that is changing. It’s constructing an analogous tool optimized for nurses.

“The nursing workflow could be very different from that of physicians, and any solution developed for nurses must integrate with the way in which they work,” Presti said through the briefing. “Our team has spent hours shadowing nurses during their shifts to see how they perform their tasks and to find where the best points of friction exist throughout their day.”  

Microsoft is working with organizations like Stanford Health Care, Northwestern Medicine and Tampa General Hospital to develop it.