A Bayesian analysis of heart rate variability changes over acute episodes of bipolar disorder

This codebase was developed by Filippo Corponi and Bryan M. Li. It is part of the npj Mental Health Research paper "A Bayesian analysis of heart rate variability changes over acute episodes of bipolar disorder". If you find this code or any of the ideas in the paper useful, please consider starring this repository and citing:

@article{corponi2024bayesian,
  title = {A Bayesian analysis of heart rate variability changes over acute episodes of bipolar disorder},
  author = {Corponi, F. and Li, B.M. and Anmella, G. \etal},
  journal = {npj Mental Health Research},
  volume = {3},
  page = {44},
  year = {2024},
  doi = {10.1038/s44184-024-00090-x},
  url = {https://doi.org/10.1038/s44184-024-00090-x}
}

Setup

• Create a new conda environment with Python 3.10.

  conda create -n hrv python=3.10

• Activate hrv virtual environment

  conda activate hrv

• Install all dependencies and packages.

  pip install -r requirements.txt
  pip install -e .

dataset/README.md details the structure of the dataset.

Preprocessing

The commands below preprocess the data (see manuscript for details). Please see --help for all available options.

python preprocess_ds.py --output_dir data/preprocessed/unsegmented --overwrite --overwrite_spreadsheet

python segment.py --output_dir data/preprocessed/sl300_ss60 --extract_features hrv --hrv_extractor flirt --segment_length 300 --step_size 60 --overwrite --use_empatica_ibi

python build_hrv_dataset.py --dataset data/preprocessed/sl300_ss60 --output_dir runs

Bayesian Analysis

The command below runs the HRV analysis (see manuscript for details). Please see --help for all available options.

python bayesian_analysis.py --working_dir runs