DeLLiriuM: A large language model for delirium prediction in the ICU using structured EHR

This reporistory contains the code for running the experiments in the paper "DeLLiriuM: A large language model for delirium prediction in the ICU using structured EHR".

Steps

1. Install the required packages:

   pip install -r requirements.txt

2. Download the public datasets

3. Run the codes on the datasets folder to pre-process the data:

4. Run the codes on the data_processing folder to process and compile all data to train and evaluate the models

5. Run the codes on the models folder to train and evaluate the models

6. Run the codes on the analyses folder to perform the analyses of the results