Microsoft announced that its agent orchestrator for cancer care management is now available in the Azure AI Foundry Agent Catalog.
Azure’s AI Foundry platform allows developers to design, manage, customize and support enterprise-ready AI agents and apps.
An agent orchestrator is a system that allows multiple AI agents to work together to complete a task.
The tech giant’s oncology-focused agent orchestrator highlights pre-set agents with multi-agent orchestration and open-source customization options that permit developers and researchers to construct agents.
The agents coordinate multidisciplinary, multimodal healthcare data workflows, including tumor boards and streamline deployment into healthcare enterprise productivity tools like Microsoft Teams and Word.
Microsoft claims the agent orchestrator could lower administrative roadblocks and transform care delivery.
In a statement, Microsoft said that “modular, general reasoners, as well as specialized, multimodal AI agents, work together to address tasks that would take hours, to augment clinician specialists with customized cutting-edge agentic AI effectively.”
The healthcare agent orchestrator can also oversee the analysis and reasoning of diverse healthcare data types, including digital imaging and communications in medicine (DICOM files) and pathology (whole-slide images), genomics data and clinical notes from electronic health records.
In a statement, Microsoft explained that each agent is provided with advanced AI models from Azure AI Foundry, integrating general-purpose reasoning capabilities with “health-specific modality models to drive actionable insights grounded in multimodal clinical data.”
According to Microsoft, early development partnerships featured the integration of multi-agent workflow into Teams chats. For instance, group chats allowed conversations between multiple human experts and specialized healthcare AI agents linked to specific healthcare data.
The capabilities of healthcare agent orchestrator include:
-
Orchestrating agentic capabilities that can reason over intricate EHR data and expand protracted duties such as building a chronological patient timeline and determining cancer stage.
-
Supplying tools that link enterprise healthcare data via Microsoft Fabric and the Fast Healthcare Interoperability Resources (FHIR) data service.
-
Permitting developers to create, customize, and adjust each agent with their models, tools, instructions and data sources.
-
This allows for interoperability and integration into active workflows, such as distribution to familiar tools like Teams, Word, PowerPoint, and Microsoft 365 Copilot.
-
Supplying explainability capabilities in agentic AI-generated outputs, grounding responses to the source EHR data and adoption in high-stakes healthcare environments.
According to Microsoft, researchers and developers at Stanford University, Johns Hopkins, Providence Genomics, Mass General Brigham, and the University of Wisconsin School of Medicine (UW) and Public Health are investigating the healthcare agent orchestrator to study how agentic AI could deliver value to complex clinical tasks, including cancer care.
“Stanford Medicine sees 4,000 tumor board patients a year and our clinicians are already using foundation model generated summaries in tumor board meetings today (via a PHI safe instance of GPT on Azure),” Mike Pfeffer, chief information officer at Stanford Health Care and Stanford School of Medicine, said in a statement.
“The new healthcare agent orchestrator has the power to streamline this existing workflow by reducing fragmentation and enabling new insights from data elements that were challenging to search, such as trial eligibility criteria, treatment guidelines and real-world evidence. Stanford Health Care is excited to research further the potential of using the healthcare agent orchestrator to build the first generative AI agent solution used in a production setting for real-world care for our cancer patients.”
THE LARGER TREND
Earlier this month, Microsoft added Elon Musk’s AI startup xAI’s Grok 3 to its Azure platform. Grok 3 and Grok 3 mini will be available through Azure’s AI Foundry platform.
Grok 3, xAI’s flagship model, comprises LLMs focused on pushing AI innovation and accelerating scientific discovery.
The companies said the model was trained on “xAI’s Colossus supercluster with 10x the computing power of prior leading models.”
In 2023, Blue Shield of California announced a multi-year cloud development plan with Microsoft to offer its members an integrated data hub, dubbed the Experience Cube, that runs on Microsoft’s Azure cloud platform.
The Experience Cube utilized Azure’s analytics and storage capabilities to connect near real-time member, provider and payer data, aiming to create more personalized services for members and drive insights for providers.
The platform was first used as an integrated digital health record that includes a member’s lab results, health conditions, emergency room visits, medications, plan coverage and other information related to their healthcare.
That same year, Japanese ICT provider Fujitsu unveiled a new cloud-based platform that allows for the secure collection, storage and leverage of health and health-related data.
Based on Microsoft Azure, the new platform features automatic conversion of medical data from EMR systems to conform with current HL7 FHIR standards for easy utilization and data exchange.
Additionally, it can convert patient-encoded personal health information to non-personally identifiable information, which hospitals and other medical institutions can use to select appropriate medications and treatment methods for patients, promoting personalized healthcare.
Credit: Source link
