Microsoft has rolled out new cloud capabilities for payers to harness data for better member outreach and care management.
The tech giant also recently rolled out new AI capabilities to unlock unstructured data for advances in cancer research and clinical trial matching.
Microsoft announced its cloud service designed specifically for healthcare back in 2020 and has been adding new capabilities and features over the past three years as it pushes deeper into healthcare. Microsoft Cloud for Healthcare brings together existing services like chatbots, Microsoft 365, Teams and Azure.
Microsoft, Google and Amazon Web Services (AWS) are pushing deeper into healthcare in a battle to provide cloud computing and data storage technology to hospitals, payers and other healthcare organizations.
The new capabilities for payers include a unified member view to provide a single place to aggregate, access and utilize different types of data instead of toggling between multiple screens and systems. The tool enables payers to view complete records that combine data from various sources throughout the healthcare system, according to Tom McGuinness, corporate vice president, worldwide health at Microsoft in a blog post.
It builds upon the Microsoft Cloud for Healthcare Payor data model, previewed in 2022, and combines member information about claims, coverage, risk profile and care plans.
Microsoft Cloud also developed a care management outreach tool to automate workflows with common scenario templates, such as for patient discharge, care plan off track and diabetes.
As healthcare moves ahead into hybrid care that combines both virtual and in-person services, Microsoft also developed a feature to support remote care management. The new feature allows care teams and care managers to remotely monitor their patients’ vital health data, identify trends with that data, and manage their inventory of assigned medical devices. It builds on the capabilities of MedTech service in Azure Health Data Services as well Microsoft's independent software vendor partner, Life365, that has integration to more than 400 original equipment manufacturer (OEM) devices.
"There's a tremendous amount of information that is potentially available to individuals, if we could find it, if we could synthesize it and then if we could apply it, those are things where technology can be extremely valuable. It allows us to be able to do this at scale," said David Rhew, global chief medical officer and vice president of healthcare at Microsoft, said in an interview.
New AI tools for healthcare
Microsoft also recently rolled out new AI capabilities to leverage social determinants of health information and cancer data as part of its Azure AI Services for Health.
One new tool helps to extract social determinants of health information from unstructured text. This helps to unlock mentions of social, environmental and demographics factors from unstructured biomedical data, according to executives.
"What we're really identifying is that there are many factors out there that impact outcomes above and beyond age and family history and one of them is social determinants, such as socioeconomic status. A lot of that information is somehow buried within the text or in some different sources. One of the things that we announced was this text analytics for health care for social determinants and ethnicity as these are major factors for outcomes and they are not oftentimes readily captured and structured," Rhew said.
Healthcare organizations, providers, researchers, and pharmaceutical companies can extract insights, improve care, assess health inequity issues, track health outcomes, as well as incorporate underrepresented groups into clinical trials and research.
Microsoft also is leveraging AI for pre-built models to help healthcare organizations tackle clinical trial matching and advance cancer research by finding key cancer attributes within their patient populations.
As part of Project Health Insights, the Oncology Phenotype model helps healthcare providers to identify cancer attributes such as tumor site, histology, clinical stage, tumor, nodes, and metastasis (TNM) categories and pathologic stage TNM categories from unstructured clinical documents. The Trial Matcher model receives patients' data and clinical trials protocols and provides relevant clinical trials based on eligibility criteria.
By using AI to parse through cancer data like tumor site and histology, researchers can better identify the potential for recurrence, Rhew said.
"If we could pull all this together and enable individuals to be rapidly identified for clinical trials, we can start to put the building blocks in so that we can run these more sophisticated AI algorithms to be able to help solve these problems," he said.
Microsoft also released a new Azure Health Bot template to enable healthcare organizations to experiment with the integration of Azure OpenAI Service into their chatbots.
The feature does not aim to facilitate the bot to answer unknown queries in the medical space, according to McGuinness in the blog post, but it enables customers to access the Azure OpenAI Service API and decide how to use the model to improve their chatbots. Currently, Microsoft is previewing the new health bot template for internal testing and evaluation purposes only.
"There's been so much discussion around large language models and generative AI and now having the Azure health bot integrated with the OpenAI service so that you can actually answer not only those questions that you had previously defined, but it would allow you to be able to answer things in a way that you're searching for using validated sources," Rhew said. "These [tools] are all building blocks to help us achieve better health outcomes and better coordination of care."