Microsoft Unveils Innovative Healthcare Tools Featuring Copilot AI Agents
Microsoft has established itself as a key player in the development of tools and services tailored for the expansive healthcare sector. As we approach the HLTH 2024 conference set to take place in Las Vegas next week, the tech giant has unveiled a series of innovative features aimed at healthcare enterprises, most of which harness the power of generative AI.
According to a recent blog entry, one of the standout features is the introduction of a healthcare agent service within Copilot Studio. More details can be found in a different blog post. Microsoft elaborates:
The healthcare agent service empowers healthcare organizations to create their own AI-enhanced agents designed for patients or medical professionals, facilitating a variety of tasks including appointment management, clinical trial alignment, patient triage, and more. Additionally, this service allows for the enhancement of an organization’s agents with extra plugins, regardless of their origin.
This innovative agent service is now available in public preview. Furthermore, Microsoft has revealed that it is implementing new clinical safeguard APIs for its healthcare offerings, available through a private preview for chosen clients. These APIs are aimed at assisting AI healthcare services in identifying inaccuracies and omissions, as well as tracing the origins of the claims made by these AI services.
Additionally, Microsoft has introduced new AI models specifically for healthcare applications, as detailed further in another blog post. These models will be accessible through Azure’s AI model catalog, with one notable model being MedImageInsight, which specializes in interpreting medical imaging data. Microsoft states:
Healthcare organizations and researchers can leverage model embeddings and create tailored adapters for their unique needs, optimizing workflows across radiology, pathology, ophthalmology, dermatology, and other fields.
Another model introduced is MedImageParse, designed to evaluate various imaging formats generated by equipment such as x-ray machines, CT scanners, MRIs, and ultrasound devices. Lastly, the model CXRReportGen is tailored specifically for analyzing chest x-ray images.
Leave a Reply