Cyber-Medicine: Microsoft’s Advancements in Healthcare AI Technology

The buzz around AI advancements has been buzzing for years, but it’s really been ramping up in the past year, especially in healthcare spaces. Microsoft in particular has made some waves with the announcement of several of its AI platform developments. A few of these platforms have recently moved forward into the testing stage, with healthcare organizations putting them to use in direct patient care. Let’s take a look at what some of these programs have to offer, and how they may be able to improve the healthcare experience for both patients and providers.

Azure AI Catalog

Microsoft’s Azure AI is a catalog of foundational AI models that healthcare organizations can customize and adapt to their specific needs. Essentially, they can use the framework of Microsoft’s models to develop AI programs that can incorporate imaging, clinical records, and genomic data while skipping the leg work it would normally take to create a program like that. Healthcare organizations typically aren’t willing the spend the time and resources necessary to build AI platforms from scratch, but with Azure AI they could bypass many of those upfront costs and jump straight into fine-tuning a product that fits their needs.

Image Source: https://gdcitsolutions.com/resources/tech-articles/azure-ai-enterprise-intelligence/

Some Examples of the AI Models Available in the Azure Catalog:

§ MedImageInsight uses image analysis to streamline workflows and build tools specific to their specialty and practice. It can flag abnormalities, generate preliminary reports, and automatically route images to the right specialists. In a hospital setting, this could fast-track diagnosis, potentially leading to faster treatment and better patient outcomes.

§ MedImageParse provides precise image segmentation, meaning the program can segment an image into regions or differentiate areas by color, shape, or texture. In healthcare, this can be used for tumor segmentation or organ delineation, allowing for targeted cancer and other disease detection and diagnosis.

§ CXRReportGen uses current and previous chest X-ray images and patient information to highlight AI-generated findings directly on the images. This could quickly alert providers to potential problems, leading to decreased turnaround times and faster diagnosis.

Microsoft Fabric

Healthcare data is a vast landscape of information spread across several different platforms. We have some big sources of information, like the CDC, NIH, and WHO, who share their information, but there’s still no centralized location for health data. Microsoft Fabric aims to solve that by pulling healthcare data from a variety of sources and centralizing it in one place for analysis. There is a suite of features that allow providers and organizations to grab data from anywhere and analyze it for various uses, like care planning, diagnosis, and identifying gaps in care.

Image Source: https://www.comset.co.uk/business-analytics-news/more-new-features-for-microsoft-fabric-as-it-approaches-go-live/

Here are a few examples:

§ Conversational Data Integration can send audio files, transcripts, and clinical notes to the Microsoft Fabric platform. The data collected can be used with Fabric or Azure AI models to analyze it or combine it with other information to get comprehensive patient insights.

§ Social Determinants of Health (SDOH) Data Analysis takes national and international information from a variety of sources and analyzes it, allowing organizations to spot health trends and find areas where the industry isn’t serving the public as it should.

§ Medicare/Medicaid Claims Data can be analyzed and combined with SDOH data to identify gaps in care and keep healthcare costs down.

§ Care Management Analysis can identify high-risk patients, optimize treatment plans, and improve care coordination, allowing providers to be more efficient and proactive in developing treatment and care plans.

Ambient Listening, A Collaboration with EPIC for Nursing Support

Ambient listening is an AI-powered software that records audio data during patient care and sends it to the patient’s chart. It will create a draft of nursing documentation that the nurse then reviews for accuracy before it officially becomes a part of the patient’s medical record. The goal is to cut down the time that nurses have to spend documenting, giving them more time to spend directly with the patient. As a nurse, I need to note that this software is not a replacement for safe staffing and absolutely cannot be used as an excuse to overload nurses with too many patients. Nurses have a lot more responsibilities than just documenting, and AI will never replace clinical nursing judgment.

Image Source: https://www.epic.com/epic/post/epic-and-microsoft-bring-gpt-4-to-ehrs/

DAX Copilot

Microsoft has developed a healthcare agent service within its DAX Copilot AI. Providers can use the software to assist with appointment scheduling, match patients to clinical trials, and even assist with patient triage. Much like Ambient Listening, it’s meant to free up time providers spend on administrative tasks, giving them more face time with patients and assisting them in providing better direct patient care.

Security

The AI discussion inevitably always includes potential security threats, and that conversation is vitally important in healthcare applications. In recent years, there have been reports of many major healthcare providers subject to data breaches and cyber-attacks, putting patient privacy at risk. AI could potentially increase that risk, but AI tools can also be used to further protect sensitive data.

Microsoft claims that they “[aim] to meet the Health Insurance Portability and Accountability Act (HIPAA), Health Information Trust Alliance (HITRUST), and General Data Protection Regulation (GDPR) requirements, with role-based access control, data governance, and data lineage.” They’ve published a set of principles to guide their AI development and have teamed up with OpenAI to monitor and research potential cyber threats to their software.

Final Thoughts

Microsoft is a powerhouse in healthcare technology, and these advancements could drive us forward by leaps and bounds. The hope is that these technologies can improve the process of providing healthcare, leading to better health and quality of life. It is not without its challenges, and safety and security must always be at the forefront of technological development, but the future of healthcare AI seems optimistic.

https://news.microsoft.com/2024/10/10/microsoft-expands-ai-capabilities-to-shape-a-healthier-future/

https://www.9news.com/article/news/local/st-joseph-hospital-denver-ai-nurses/73-f58d6c39-a402-4560-b93a-53b1298cf37e

https://www.microsoft.com/en-us/security/blog/2024/02/14/cyber-signals-navigating-cyberthreats-and-strengthening-defenses-in-the-era-of-ai/

Access the article here