EmpiricalCalibration 3.5.1 has been released. This includes a minor quality of life improvement.
HADES-wide release 2024Q3 has been released. This is a snapshot of the HADES packages and their dependencies, thoroughly tested and confirmed to be mutually compatible. It is intended to be a stable environment for studies and execution engines. Currently, the release is only available as an renv lock file.
This also marks the transition from R 4.2.3 to R 4.4.1 as our main supported R version. We highly recommend everyone to use R 4.4.1 moving forward. HADES will likely run on other versions as well, but everything has been tested on this version. (The main reasons for the switch is that not all binaries are available for 4.2.3 anymore, meaning slower and more error-prone building of renv environments.)
SqlRender 1.19.0 has been released. This adds translation rules that ensure emulated temp tables are dropped before creating them, in case a prior run had left some orphan emulated temp tables.
Also adding tentative support for InterSystems IRIS.
PheValuator 2.2.12 has been released. It has updated code to remove temporary files in database queries.
FeatureExtraction v3.7.2 has been released, it contains a couple of bugfixes.
SqlRender 1.19.1 has just been released. This contains some bugfixes for InterSystems IRIS.
Strategus v1.1.1 is now released and is officially part of HADES. Strategus is an R package for coordinating and executing analytics using OHDSI HADES modules. Strategus allows you to define and capture your study design across the HADES packages used for characterization, population-level effect estimation and patient-level prediction. The study design choices are captures in an analysis specification document which is stored as JSON and used as an input for executing the study via Strategus. Strategus also provides utilities for uploading results of your study using the ResultModelManager package.
I’d like to personally thank @schuemie, @jpegilbert, @Chris_Knoll and @Jreps for all of their contributions in the development of Strategus. I’d also like to thanks the OHDSI community members who have provided valuable feedback while using Strategus to execute OHDSI Network Studies and via the Strategus working group.
1 Like
DataQualityDashboard v2.6.2 has been released. This release contains some bugfixes, documentation updates, and the addition of a new executeDqChecks parameter which allows users to specify the check severity level for a run of DQD.
Special thanks to Sam Hillman and @billobrien , who contributed to this release during the HADES Hackathon at this year’s Global Symposium! It was a true pleasure working with you both at the Hackathon and I’m very grateful for your contributions 
Thank you for the development of Strategus; this is really helpful.
Just curious regarding the PopulationLevelEstimationModule. Is it available yet?
Thank you
Hi @YudhaESap - glad you are finding Strategus helpful. There are currently two population level estimation modules available: CohortMethodModule and SelfControlledCaseSeriesModule.
DatabaseConnector 6.3.3 has been released. Most important change is in insertTable() for DataBricks, which is now several orders of magnitude faster.
DatabaseConnector 6.4.0 has been released. This adds support for the InterSystems IRIS database platform.
PatientLevelPrediction v6.4.0 has now been released on github and CRAN.
As you can see from the changelog this is quite a large release with many changes required to make it suitable for CRAN. But apart from that there were quite a number of new features, imputation methods, loading sklearn models (trained in python) for validating with the PLP, using a manually specified train-test split, net benefit plot and so on. See changelog for details.
Strategus v1.2.0 has been released.
This release provides better error handling and logging for study execution, result model creation & results upload.
MethodEvaluation 2.4.0 has been released. This contains several minor changes and fixes. Most importantly this updates the OHDSI Benchmark code, which was not working on more recently added database platforms (e.g. DataBricks).
EmpiricalCalibration 3.1.4 has been released. This includes some minor changes that make it more robust against extreme outliers in the negative control distribution.
SelfControlledCaseSeries 5.3.1 has been released. This contains some minor fixes for edge cases.
TreatmentPatterns 3.0.0 (CRAN) was released and is now part of HADES.
You can install TreatmentPatterns like so:
install.packages("TreatmentPatterns")
# Or from Github
remotes::install_github("darwin-eu/TreatmentPatterns")
You can either use DatabaseConnector or CDMConnector as an interface to your database.
For further reading please visit the package website.
There is no Strategus module as of yet, but is on the to-do list. Same goes for an OhdsiShinyModules module.
Thanks @schuemie and @anthonysena for all the feedback, and making this possible.
Capr v2.0.9 has been released. It contains minor bug fixes to code and documentation.
CohortDiagnostics version 3.3.1 has been released. This is a hotfix release for MS SqlServer platforms that encountered an issue with datediff calculations involving big integers.
Please check the release notes for full details of changes.
1 Like