These repository includes the Jupyter notebooks to derive logistic regression models that predict Borrelia diagnosis using clinical and laboratory data of patients.
-
2018-2019-Clinical-Malaria-Cleaned.ipynb: This notebook constructs a clinical-only prediction model using data from 2018 and 2019, including patients who were currently diagnosed positive with malaria
-
2018-2019-Clinical-No-Malaria-Cleaned.ipynb: This notebook constructs a clinical-only prediction model using data from 2018 and 2019, removing all patients who were currently diagnosed positive with malaria.
-
2019-Clinical-Cleaned.ipynb: This notebook constructs a clinical-only prediction model using data from 2019, for all patients (malaria negative and postive)
-
2019-Clinical+Lab-Cleaned.ipynb: This notebook constructs a clinical + lab prediction model using data from 2019, for all patients (malaria negative and postive)
-
BestModel_2018train_2019test.ipynb: This notebook constructs a clinical-only model from 2018 data and validates it on 2019 data, for all patients (malaria negative and postive)