Version 2.01 of the Canary software is now available and can be obtained by clicking the download link in the menu above.
It includes Log functionality, output criteria numbering and several bug fixes.
A User-Friendly Information Extraction Tool
Version 2.01 of the Canary software is now available and can be obtained by clicking the download link in the menu above.
It includes Log functionality, output criteria numbering and several bug fixes.
A study comparing accuracy of Canary-based NLP tool for identification of patients with diabetes who declined a recommendation of insulin therapy to several other NLP technologies has been published in the Journal of Biomedical Informatics.
Please see this page for other publications related to Canary.
We are starting a Canary NLP Tool Library – a collection of Canary language models available to everyone free of charge. We believe that it is easier to see farther ahead when standing on the shoulders of giants, and that it will therefore be helpful to everyone in Canary user community if we are share NLP tools we have built, and build on what others have shared. Canary NLP Tool Library is available at the Library menu link above.
Version 1.3 of the Canary software is now available and can be obtained by clicking the download link in the menu above.
It includes improvements to the interface as well as a new (1.08) version of the Canary Data Converter.
Version 1.2 of the Canary software is now available and can be obtained by clicking the download link in the menu above.
It includes improvements to the interface, new sample projects, updated documentation as well as a new (1.07) version of the Canary Data Converter.
A paper describing studies performed using Canary was presented at AMIA Annual Symposium 2018.
Please see this page for other publications related to Canary.
An important step for performing information extraction using the Canary software is data preparation.
To assist with this we are making available the Canary Data Converter (CDC), a companion tool to help convert data into the format supported by Canary.
This tool comes bundled with the Canary download package.
A recently-published study of the rejection of insulin therapy by diabetics has been featured in various media outlets:
The articles provide additional perspective about the outcomes of the study. An information extraction model developed using Canary was a key component of the study.
A study analyzing patient rejection of Insulin and its impact on the delayed initiation of this therapy has been published in the Diabetic Medicine journal.
The Canary NLP software was used to create information extraction models to mine unstructured electronic medical records to identify relevant records with high accuracy.
Please see this page for other publications related to Canary.
A new study looking at patient outcomes following continuation of Statin prescriptions after adverse reactions has been published in the Annals of Internal Medicine journal.
The Canary NLP software played a key role in mining and extracting the information used in this research.
Please see this page for other publications related to Canary.