We have been very focused on issues with the Retro-CI installation that I will outline below.
The issues that came up during the last couple of weeks have given us a lot of insight into how OpenELIS is operating in a more advanced laboratory and how we can modify the system to improve the user interaction to prevent errors. In some cases, the issues pointed out are because the system is working as designed, but the user interface hides things from the user that it would be better to show. In other cases, we learned that we needed some clarification on business rules in the laboratory. Even further, there were issues that led us to decide we needed to fix the issue and improve the business rules around the process. Lastly, we worked with Justin to learn how to replicate complex and difficult issues here in the development environment in order to identify a solution.
Last period we focused on the following:
- Duplicate CD4 Absolute Counts – These test results were being duplicated in the system, and was causing issues when one value was being manually updated and the duplicate value did not update to the new result. This issue in the software has been fixed and released to production. In addition, with the help of the Retro-CI team and Justin, all data has been corrected in the system to remove all duplicated test results.
- Duplicate Test Requests – When results were imported from the analyzer prior to sample entry being performed, duplicate tests requests were created after sample entry was performed. This bug has been fixed in the software and released into production.
- CD4 Absolute Count Recalculation Error – In Imm-Heme, if a user was retesting a sample, the CD4 abs count was not being recalculated with the values from the new CD4% result. This was fixed and is now in production at Retro-CI.
- Results – Now show « en cours » at serology result when the serology result is not tested yet or is not validated instead of “X”
- Edit Patient Subject Number creates two patients pointing to the same sample in the observation history table. This has now been fixed in the software and released into production. We are now looking at how to improve the user interface to make the process of editing lab numbers and subject numbers more explicit.
- “Loading Error” bug – We found that when OpenELIS was open in more than one tab on the browser, the system became unstable and caused unexpected errors. We are working on a resolution to this in order to allow users to have OpenELIS in more than one tab for comparisons they would like to perform. Until we have a fix however, we strongly recommend against having OpenELIS open in more than one tab on the same browser. For now, a workaround to this would be to use two different browsers, such as to have OpenELIS opened in IE for one screen and have OpenELIS opened in Firefox for another screen.
- Automatic Backups – Justin is still working on getting the automatic backups setup for the system. The manual backups are working correctly and still being done. We expect to have this in place in the next week or so.
- Biological Validation slow loading – Additional work on optimizing the load and search function within the biological validation screen; Much improvement has been made here and we look forward to feedback from the users. Currently in testing and will be releasing into production soon.
- Additional minor bug fixes as discovered and prioritized
This week we will be working on the following:
- Synthesizing requirements and prioritizing work for release of CI LNSP OpenELIS – as the Retro-CI installation is now stable, this will resume and proceed forward quickly
- Ability to delete or modify the study type after sample entry – on hold for the past few weeks as focused on the migration issues. As stated before, this is a new feature requested due to users selecting the wrong study type when doing sample entry. Due to the study type effecting multiple areas of the workflow, we will be working on a good approach to solve this problem for users.
- Supporting Justin for issue fixes for Retro-CI – as needed
- Prioritizing and Estimating work/approach for resolving the next round of high priority items for Retro-CI – based on input from Dr. Kassi and Justin
0 Responses to “Progress and Priorities – Nov 28th”