Classical machine learning to classify SARs-CoV-2 vs. Mock in NHP
When referencing this work, please cite:
Chu, W. T., Castro, M. A., Reza, S., Cooper, T. K., Bartlinski, S., Bradley, D., Anthony, S. M., Worwa, G., Finch, C. L., Kuhn, J. H., Crozier, I., & Solomon, J. (2023). Novel machine-learning analysis of SARS-CoV-2 infection in a subclinical nonhuman primate model using radiomics and blood biomarkers. Scientific Reports, 13(1), 19607. https://doi.org/10.1038/s41598-023-46694-9
- Determine features most relevant to prediction of SARS-CoV-2 vs. Mock
- Build a ML model for automatic classification of SARS-CoV-2 vs. Mock
- Build a foundation for future work in severity classification & translation to humans
- 12 SARs-CoV-2 & 8 Mock Cynomolgus monkeys
- 4 time points: BL, 2, 4, & 6 days post-infection
- Radiomics measures calculated off of CT scan of lung & whole body
- Clinical pathology and immunology measures calculated off of blood sample analyses
- Preprocessing
a) Reshape & reformat radiomics data
- Calculate change from baseline
b) Reshape & reformat clinical pathology data
- Calculate change from baseline
c) Reshape & reformat immunology data
- Calculate change from baseline
d) Merge radiomics and clinical pathology data\ - Run exploratory analyses
- Classic statistics
- Data Visualization
- Feature Selection
- Relevance threshold (f-stat, MI, chi2)
- Minimum redundancy, maximum relevance (mRMR)
- Machine Learning
- Models
- Evaluation of performance
- Effect of confounding variables
- Comparison of model performance
- Peng, H., Long, F., & Ding, C. (2005). Feature selection based on mutual information: Criteria of max-dependency, max-relevance, and min-redundancy. IEEE Transactions on Pattern Analysis and Machine Intelligence, 27(8), 1226–1238. https://doi.org/10.1109/TPAMI.2005.159