-
Notifications
You must be signed in to change notification settings - Fork 1
Home
SetAnalysis is a demo that was build to demonstrate the capabilities of iKnow in Healthcare, especially for Patient Cohort Selection and Identification. It was used and extended in subsequent demo's and poc's, and proved to be a generic tool that can be used in a wide variety of use cases. On this page, you can find more info, the actual demo project code you can download and install, and some presentations and a users guide. Please feel free to mail me if you need more info.
The SetAnalysis demo is a CSP application. You can find the latest build in the attachments section of this page. Import the build in the namespace you want to use it (where your iKnow domains are located). The build works best in Caché 2014.1 or 2014.2. You can use it in earlier versions (2013.1), but you will get some compiler errors in the REST and DeepSee SetAnalysis classes (and hence no REST or DeepSee support in SetAnalysis for 2013.1). The demo works on all newer browser versions (Internet Explorer 9,10,Edge, Firefox, Firefox Portable & Chrome). It does not work with IE8! Open the SetAnalysis.csp page in the Studio and click on the Globe, or use following URL from your browser (change ip-address, port & namespace : http://localhost:port/csp/namespace/iKnowSetAnalysis.csp You can find a users guide in the Attachment section for full explanation.
In SetAnalysis, you create sets of members (e.g. patients) that have a common characteristic. That characteristic could be something you find in their texts, or something you find in the structured data. First you select one of the characteristics, by using a plugin. When you save this in a set, the SetAnalysis will also caclulate which members are in the set.
iKnow api plugins to get iKnow elements such as Concepts (Similar, Related), CRC's, Paths, Sentences SQL plugin to select structured data from any table Generic plugin for valuelists Create sets by doing set-logic on other sets (and, or, andnot) Examine members (e.g. patients) in a set : examine the evidence (sentences, sources), delete members from a set. Members with negated sentences will be marked in Red (only negated found) or Orange (negated & non-negates found) Use sets in DeepSee as filters, rows or columns (implemented in DeepSee as a computed dimension) Use REST to interrogate SetAnalysis form other applications Create rules that can extract data from texts to a SQL table
As SetAnalysis is dominantly used in healthcare Poc's, and new prospects keep on asking stuff that is not included, the SetAnalysis demo is constantly updated. If you have an upcoming demo or Poc, and you are missing a feature, please feel free to check with us. We are always interested to find out when you have an interesting poc or use case to do.