PSYCOP, Aarhus Psychiatry Research

Machine learning utilities for the PSYCOP project.

🌲 Monorepo for the PSYCOP research project. Prediction of disease from Electronic Health Records at Aarhus University.

Generation of train-test-split.

Converting irregularly spaced time series, such as eletronic health records, into dataframes for tabular classification.

CARE-ML: Predicting the use of restraint on psychiatric inpatients using EHRs and ML. Developed by sarakolding and signekb for their Master's Thesis.

Code for "Lexical Stability of Psychiatric Clinical Notes from Electronic Health Records over a Decade"

Examining the stability of diagnoses across the LPR2-LPR3 transition