Announcing Innovative AI-Powered Search
In an effort to aid healthcare organizations in effectively managing the vast troves of data gathered by doctors and hospitals, Microsoft has introduced innovative data and artificial intelligence (AI) products. This move seeks to address the challenges faced in healthcare, where crucial data is dispersed across various systems and formats, with a staggering 97% of data generated by hospitals remaining untapped.
The new healthcare-specific tools have been developed within Fabric, a data and analytics platform, which enables the amalgamation of data from diverse sources, including electronic health records, medical images, lab systems, medical devices, and claims systems. This unified data platform allows organizations to streamline the data, making it readily accessible in one central location. The primary objective of these tools is to eliminate the laborious process of searching through multiple sources individually, enabling medical providers to focus on delivering quality care.
Revolutionizing Healthcare Data Management
Microsoft has been conducting trials of Fabric for healthcare with select partners, including Northwestern Medicine, Arthur Health, and SingHealth. It is now available in a preview capacity. Doug King, Chief Information Officer at Northwestern Medicine, expressed excitement about the potential benefits. He emphasized that consolidating disparate data could lead to improved healthcare systems and a more efficient patient care process. The strategic deployment of Microsoft’s technology is expected to offer various advantages, from optimizing patient flow and staffing to integrating broader population health data into patient care. Doug King believes that such advancements are pivotal in solving some of the healthcare industry’s most significant challenges.
Introducing Azure AI Health Tools
Microsoft’s commitment to healthcare innovation extends to its Azure AI services. One notable addition is the Azure AI Health Bot, a generative AI chatbot capable of sourcing information from both internal data and reputable external sources, such as the Food and Drug Administration and the National Institutes of Health. This chatbot is intended to assist staff with questions related to disease treatment and internal protocols. Patients can also access the chatbot through their patient portal to seek clarifications about symptoms and medical terminology. Linishya Vaz, Principal Product Manager at Microsoft Health and Life Sciences, highlighted the importance of built-in safeguards and the ability to verify information for credibility.
Text Analytics for Health and Azure AI Health Insights
Another valuable solution introduced by Microsoft is Text Analytics for health, a tool designed to label and extract critical medical information from unstructured data sources, including clinical documents and notes. It will be available in multiple languages, including Spanish, French, Italian, German, Portuguese, and Hebrew.
Additionally, Microsoft unveiled three new models within Azure AI Health Insights. The “patient timeline” model offers clinicians a chronological overview of a patient’s medical history by consolidating data from diverse sources. “Clinical report simplification” aims to simplify complex medical reports for better patient understanding using generative AI. The “radiology insights” model identifies errors and inconsistencies across reports and provides follow-up recommendations.
These new healthcare tools within Azure AI are expected to enhance patient experiences and enable clinicians to focus on delivering improved care. They are currently available in a preview capacity, representing a significant stride forward in healthcare data management and patient care optimization.
Source
Discover More AI Tools
Every week, we introduce new AI tools and discuss news about artificial intelligence.
To discover new AI tools and stay up to date with newest tools available, click the button.
To subscribe to the newsletter and receive updates on AI, as well as a full list of 200+ AI tools, click here.
