CTcue version update: 4.6.0

The latest version of CTcue is ready for rollout: v4.6.0. Below is a more detailed overview of some of the bigger changes in this version, besides some minor improvements and bug fixes. To read the full release notes, click here. 

More powerful add term search 

Over time, we have introduced many new features and functionalities to CTcue, which enables users to increasingly meet their data needs. For new users, however, the sheer amount of options can be overwhelming: which category should I select to find certain data? Where do you find specific filters?  

To help new users find their way building queries, we have been working to make the add term search more powerful. Users can now describe what they are looking for and CTcue will provide suggestions of matching categories and filters. If you wish to use the suggestion, then you can select the suggestion and add it to your query. When multiple suggestions are provided, you can select the ones you want to use and add them in bulk. After adding the suggestions, you can continue building the query with more terms and filters. This will not only help new users, but also make it faster for experienced users to build their queries.  

Users who wish to work the way they did before can still do so: there is still the option to select the categories from the overview and build the query from scratch yourself. We have redesigned the layout to keep a better overview of the available categories. To that end we have removed the “recently used categories” and “recently searched” sections from the view. 

For now we support a number of common types of input. We will look to further improve this “text to query” functionality in the near future, so that more and more types of input are supported. If your search input doesn't generate any results, feel free to submit your input string to our support team, so we can implement support for this later. 

Some tips and tricks when you start using text to query: 

• You may use quoted text to search for specific terms without the system interpreting this further 
• When typing DBC names, they must match exactly 
• If you are not getting any results, try following the pattern: category – field – value 

Various patient finder improvements 

We have implemented a number of improvements to the patient finder to help speed up the process and make cooperation with others easier. Here is a brief overview of the changes: 

• To provide more overview when you have many groups in your criteria, you can now collapse and expand groups. There is an arrow in the top right corner of each group that allows you to toggle between the two states. 
• To allow users to reuse specific groups of terms from other projects, we have added the option to import groups. This means you no longer have to import the individual terms and build the group from scratch with the relations between the terms. Groups are imported from the same section as “import term”, so that you can import single terms and groups at the same time. To help keep an overview, we have now also divided the terms by inclusion, exclusion and questions (data collector only) and added expand/collapse options for these sections. 
• The patient cohorts table now uses lazy loading to reduce waiting times. This means that the table will be displayed as soon as the first 10 patients are loaded. The remaining patients will be loading in the background. Just beware when doing bulk validation that more patients may still be loading in the background. 
• You can now jump to a specific page in the table when there are more than 10 pages in the table. This may be helpful when you come back to the cohorts table at a later time to continue reviewing patients where you left off before. 
• The comments column now displays the latest comment for a patient instead of only the total count. If there is a specific comment that you want to display in the table, then you can pin that comment. 
• If you have a group with multiple terms, then you can filter in the cohorts table for the number of matching terms within that group. This may help speed up the validation process. For example, if you added 2 types of medication in a group and want to directly bulk include patients with results for both medications, but want to review the patients with results for only one of the medications. To use this filter, make sure to add the column “group summary” to your cohorts table. 

• You can now decide if your manually added patients should be added to the inclusion cohort or the exclusion cohort. A dropdown menu has been included on the screen to select which cohort the patients will be added to. 
• Previously you could only export the included and/or excluded patients, but now you can also export the patients listed in the search results tab. This may be helpful if you want to validate patients offline or by someone without access to CTcue. 
• When you duplicate a project, you can now choose to also include the patient cohorts and/or data validation table (Data Collector only). This means you don't need to search and validate the data from scratch again in the duplicate project. 

New category: care plans 

Care plans is a new category we are introducing to cover prescribed and planned care activities and medication for patients. At the moment only medication plans (Dutch: kuren) are covered, but we plan to add other care activities in the future.

We would like to highlight some of the new filters in this category that have been requested by our users:

• Care plan description - the name or description of a care plan
• Template - the template code (Dutch: VCMO) that is used to prescribe a set of medication
• Care plan type - eg. cytostatical

Reports Plus changes 

Reports Plus is still an experimental feature that is being tested with a few specific hospitals before we do a general roll out. For Reports Plus we have made a number of changes to make it easier to use. 

For most users the focus should be on the relevance and less so on the specific confidence scores. That is why we have moved the confidence sliders to a filter and hidden the specific confidence scores in the results table, unless users specifically enable the confidence filter and change the default sliders. 

When a concept has not been found a report, it is important for users to know whether the report just hasn't been processed yet, or whether the report was processed but the concept was not found – either because it's legitimately not a concept, or because the AI pipeline mistakenly didn't pick it up. Therefore we have added a clear indication at the top of the report warning users when a report was not yet processed.