About the "CARE - Concept Observation Tally Demographics Histograms" Web Client Plugin
This plugin is part of the CARE (Cohort Analysis & Refinement Expeditors) Tool Set developed as a result of feedback and suggestions from the i2b2 users community in the University of Massachusetts Medical School. As its name implies, the purpose of this tool set is to facilitate and expedite the analysis of the cohorts (i.e. a subset of a patient set) and therefore hopefully the refined usage of those particular cohorts in a study.
This i2b2 web client plugins enables a user to visualize the demographics of a subset of a patient set, associated (observed) with any various counts of observations of a selected concept (ontology term).
This plugin provides the following main functions:
- Generates demographic histograms on the subsets of patients associated (observed) with any various counts of observations of a selected concept (ontology term).
- Allows the usage of smaller queries of patient subgroups (from a large patient set) iteratively, in place of one single large query of that huge patient set, to reduce risk of overwhelming the server (i2b2 hive).
- Allows selection of subset of patients from a large patient set, to further reduce risk of overwhelming the server.
- Allows selection of the appropriate grouping of observation tallies against which to generate the resulting histograms.
- Provides feedbacks on current progress, elapsed time, and estimated remaining run time.
- Provides notification & suggestion in cases of paging (query-subgrouping) by the server.
- Provides termination indication & suggestion (for recovery) when the server failed.
- Provides detailed 'HELP' and 'HINT' dialogs in response to corresponding hotlinks strategically placed next to input fields that may need explanations or suggestion.
- Navigate to the "Specify Data" tab. Drag and drop a Patient Set and a concept (ontology term) onto the appropriate input boxes:#* Enter appropriate values in the 'Starting Patient' and 'Number of Patients' fields that are then presented, to specify the subset of the patient set to use. Also specify at the 'Query Subgroup Size'field the size of smaller queries of subgroups of patients to be used iteratively, to facilitate reducing the risks of overwhelming the server (i2b2 hive).
- Click on the corresponding 'HELP' hotlinks next to the 'Starting Patient' and 'Number of Patients'fields, and the 'HELP' hotlink next to the 'Query Subgroup Size' field, for detailed information; and enter corresponding appropriate values.
- Click on the corresponding 'HINT' hotlink below the 'Concepts' field for hint or suggestion for selecting concepts.
- Next, select the appropriate grouping of observation tallies against which to generate the resulting histograms at the "Output Options" tab
- Finally, select the "View Results" tab to wait for and view the generated demographic histograms.
Result Example 1
Grouping observation tallies in sequence of '1, 2, 3, 4, 5, 6, 7, 8, 9, over 9':
Result Example 2
Grouping observation tallies in sequence of '1-2, 3-4, 5-6, 7-8, over 8':
- Since it is possible for the server (i2b2 hive) to take a long time to provide all the data requested, this plugin provides occasional updates of 'Elapsed time' and 'estimated remaining run time', etc. displays while the data are being fetched. These displayed values are at best rough estimates based on occasional data coming back from the server. In the case of zero 'Query Subgroup Size' selected (i.e. no query-subgrouping), then there would be no data coming back from the server until either of the following situations:
- The whole, single, query is done.
- The server itself started paging (i.e. query-subgrouping), and the result of the 1st of such paged subqueries just arrived.
- The server got overwhelmed by all the excessive data and failed, returning an 'error'.
- In any of these cases above, the updates of 'Elapsed time' and 'estimated remaining run time' will be quite infrequent, as updates from the server (i2b2 hive) will most likely take a long time, and far and few in-between, if any.
- This plugin also tries to provide updates on the current subgroup the server (i2b2 hive) is fetching. However, in the cases when the server (i2b2 hive) itself started paging (i.e. query-subgrouping), then the current subgroup # being fetched may exceed the original stated total number of subgroups.
- This plugin has been successfully tested with the following browsers: Firefox (v31, v35), Chrome (v41), Internet Explorer 32/64 (v10, v11), Safari (v5.1.7), Opera (v28), and Vivaldi (v184.108.40.206); all on Windows 7.
- The i2b2 web client itself does not work with Opera (v12.11, v12.16).
i2b2 version compatibilities
This current version is compatible with i2b2 v.1.6-1.7.
Initial release, with options of using a subset of a large patient set; selection of observation tallies grouping for resulting histograms; user-friendly feedback (estimated run-time, etc.), helps & hints.
This plugin is published under the GNU GPL v3 license.
This software is distributed as a plugin to the i2b2 v1.6-1.7 webclient framework, and may be used free of charge. It is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
The latest version of this web client plugins can be downloaded from its Codeplex download page.