Microsoft Azure launches medical data-mining tool

  • Small
  • Medium
  • Large

Microsoft announced July 8 a new artificial intelligence tool for its cloud platform Azure that allows developers to analyze unstructured medical data, including clinical notes, clinical trial protocols and medical publications. 

Microsoft's Text Analytics for health allows researchers, data analysts and medical professionals to detect words and phrases from unstructured text and connect them to relevant healthcare and biomedical concepts, such as diagnoses, medication names and treatments. Users can extract more than 100 types of personally identifiable information, including protected health information, in unstructured text. 

The AI tool also links entities to medical ontologies and coding systems, such as the Unified Medical Language System, to detect connections between medical concepts mentioned in text. For example, the tool can be used to find the relationship between a medication name and the dosage associated with it.  

Medical researchers and data analysts can use the tool to create analytics on historical medical data, develop prediction models and match patients to clinical trials. Seattle-based Allen Institute for AI used it to develop a COVID-19 search engine allowing researchers to more quickly analyze coronavirus information. The institute, founded by Microsoft co-founder Paul Allen in 2014, partnered with the University College London to review medical research reports. 

"We have been partnering with engineers at Microsoft and data scientists to build a 'living' reviews system – that automatically identifies relevant research for reviews as they are published," James Thomas, a professor at University College London, stated in a news release. "Text Analytics for health provides a powerful tool for extracting insights from clinical literature, with rich support for a wide range of healthcare terminology so that we can more quickly and accurately identify relevant information." 

Copyright © 2022 Becker's Healthcare. All Rights Reserved. Privacy Policy. Cookie Policy. Linking and Reprinting Policy.


Featured Whitepapers

Featured Webinars