Machine learning models for predicting severe COVID-19 outcomes in hospitals

This is the Githup Repository for the paper Prediction of Covid-19 associated sentinel events based on laboratory values via machine learning in hospitals by Wendland etal. [1] containing the code to reconstruct our results and additional supplementary files. In our work we construct machine-learning based models for the prediction of in-hospital mortality, transfer to intensive-care unit (ICU) and mechanical ventilation of hospitalized Covid-19 patients. We use age, biological sex and averaged covariates of the first 48 or 24 hours after admission to a hospital as (possible) covariates.

Important: The patient data used in our analysis is not included in this repository because we are not allowed to publicly share these data due to german law.

Content of the repository

• In the folder Code_Paper you can find the code to reproduce our findings
• The folder Tests contain complete csv files with all wilcoxon-rank-sum-tests and t-tests for all endpoints including missing rates for each feature
• The lab_abbreviations.csv file contains a complete list of the full names of all laboratory values

Explanations regarding the Code for the paper

• To install all package dependencies we recommend to create a anaconda environment via the environment.yml file using "conda env create -f environment.yml"

List of Package dependencies

• pandas [2]
• numpy [3]
• matplolib [4]
• seaborn [5]
• sklearn [6]
• scipy [7]
• statsmodels [8]
• xgboost [9]
• skopt [10]
• tidyverse [11]
• proc [12]

References

[1] Wendland P, Schmitt V, Zimmermann J, Schenkel-Häger C, Kschischo M. "Machine learning models for predicting severe COVID-19 outcomes in hospitals". 2022

[2] McKinney W. Data Structures for Statistical Computing in Python, Austin, Texas: 2010, p. 56–61. https://doi.org/10.25080/Majora-92bf1922-00a.

[3] Harris CR, Millman KJ, van der Walt SJ, Gommers R, Virtanen P, Cournapeau D, et al. Array programming with NumPy. Nature 2020;585:357–62. https://doi.org/10.1038/s41586-020-2649-2.

[4] Hunter JD. Matplotlib: A 2D Graphics Environment. Comput Sci Eng 2007;9:90–5. https://doi.org/10.1109/MCSE.2007.55.

[5] Waskom M. seaborn: statistical data visualization. JOSS 2021;6;3021. https://doi.org/10.21105/joss.03021.

[6] Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, et al. Scikit-learn: Machine Learning in Python 2018.

[7] Virtanen P, Gommers R, Oliphant TE, Haberland M, Reddy T, Cournapeau D, et al. SciPy 1.0: fundamental algorithms for scientific computing in Python. Nat Methods 2020;17:261–72. https://doi.org/10.1038/s41592-019-0686-2.

[8] Seabold S, Perktold J. Statsmodels: Econometric and Statistical Modeling with Python, Austin, Texas: 2010, p. 92–6. https://doi.org/10.25080/Majora-92bf1922-011.

[9] Chen T, Guestrin C. XGBoost: A Scalable Tree Boosting System. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco California USA: ACM; 2016, p. 785–94. https://doi.org/10.1145/2939672.2939785.

[10] Head T, MechCoder, Louppe G, Iaroslav Shcherbatyi, Fcharras, Zé Vinícius, et al. Scikit-Optimize/Scikit-Optimize: V0.5.2. Zenodo; 2018. https://doi.org/10.5281/ZENODO.1207017.

[11] Wickham H, Averick M, Bryan J, Chang W, McGowan L, François R, et al. Welcome to the Tidyverse. JOSS 2019;4:1686. https://doi.org/10.21105/joss.01686.

[12] Robin X, Turck N, Hainard A, Tiberti N, Lisacek F, Sanchez J-C, et al. pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics 2011;12:77. https://doi.org/10.1186/1471-2105-12-77.

Contact: Philipp Wendland - [email protected]