Additional file 1 of Common clinical blood and urine biomarkers for ischemic stroke: an Estonian Electronic Health Records database study

Additional file 1: Table S1. Abbreviations used in the article. Table S2. Clinical parameters mentioned in the article. Text S3. Raw data extraction workflow. Text S4. Semi-automatic pipeline for cleaning raw tabular EHR data for downstream analyses of IS. Fig. S5. Sources for phenotype information from the EstBB. Text S6. Filters used in Association Rule Mining (ARM) to identify potentially interesting rules for further testing. Text S7. R script to transform numerical clinical data for Logistic Regression (LR). Text S8. R script used for Cox Proportinal Hazards (CPH) model and Kaplan-Meier (K-M) graphs. Text S9. Deep neural networks (DNN) implementation. Table S10. Five association rules (ARMs) identified. Fig. S11. Lowess curve examples. Table S12. Summary of lowess. Table S13. Summary of Logistic Regression (LR). Fig. S14. Kaplan-Meier graphs for CPs with P