Category | Badge |
---|---|
License | |
CI/CD | |
Code Coverage | |
Code Quality |
- In this package, we present the Modified Detecting Deviating Cells (MDDC) algorithm for adverse event identification.
- For a certain time period, the spontaneous reports can be extracted from the safety database and depicted as an
$I \times J$ contingency table, where:-
$I$ denotes the total number of AEs -
$J$ denotes the total number of drugs - With cell counts
$n_{ij}$ the total number of reported cases corresponding to the$j$ -th drug and$i$ -th AE
-
- We are interested in which (AE, drug) pairs are signals. The signals refer to potential adverse events that may be caused by a drug.
- In the contingency table setting, the signals refer to the cells with
$n_{ij}$ abnormally higher than the expected values. - Rousseeuw and Bossche (2018) proposed the Detecting Deviating Cells (DDC) algorithm for outlier identification in a multivariate dataset.
- The original DDC algorithm assumes multivariate normality of the data and selects cutoff values based on this assumption. We modify the DDC algorithm to better suit the discrete nature of adverse event data in pharmacovigilance that clearly do not follow a multivariate normal distribution.
- Our Modified Detecting Deviating Cells (MDDC) algorithm has the following characteristics:
- It is easy to compute.
- It considers AE relationships.
- It depends on data-driven cutoffs.
- It is independent of the use of ontologies.
- The MDDC algorithm has five steps, with the first two steps identifying univariate outliers via cutoffs, and the next three steps evaluating the signals via the use of AE correlations. The algorithm can be found at MDDC algorithm.
-
Anran Liu Email: anranliu@buffalo.edu
-
Raktim Mukhopadhyay Email: raktimmu@buffalo.edu
-
Marianthi Markatou Email: markatou@buffalo.edu
Raktim Mukhopadhyay
Email: raktimmu@buffalo.edu
The documentation is hosted on Read the Docs
at - https://mddc.readthedocs.io/en/latest/
pip install MDDC
For installing the development version, please download the code files from the master branch of the Github repository.
Please note that installation from Github might be buggy, for the latest stable release please download using pip
.
For downloading from Github, use the following instructions:
git clone https://github.com/rmj3197/MDDC.git
cd MDDC
pip install -e .
Please refer to the Contributing Guide.
The code of conduct can be found at Code of Conduct.
This project uses the GPL-3.0 license, with a full version of the license included in the repository.
The work has been supported by Food and Drug Administration, and Kaleida Health Foundation.
Rousseeuw, P. J., & Bossche, W. V. D. (2018). Detecting deviating data cells. Technometrics, 60(2), 135-145.