carwatch-python

Get the Android app on the Play Store:

CARWatch is an open-source framework to support objective and low-cost assessment of cortisol samples in real-world, unsupervised environments. It is especially suitable for cortisol awakening response (CAR) research, but not limited to this application.

It consists of an Android application that schedules sampling times and tracks them by scanning a barcode on the respective sampling tube as well as a Python package that provides tools to configure studies and prepare the study materials and to process the log data recorded by the app.

Features

In order to use CARWatch, you need to prepare the study materials and configure the app. The whole workflow is illustrated in the following figure:

All these features are provided by carwatch which offers a user-friendly command-line interface (CLI) for the following tasks:

1. Setting up a CAR study.
   This includes:

   • Customizing study properties to your needs
   • Setting up your desired sampling schedule
   • Generating a QR-Code for the CARWatch app to automatically set up the study in the app

2. Creating printable labels with barcodes for objective sampling time assessment.
   This includes:

   • Customizing the numer of saliva samples per day, the number of days, and the number of study participants
   • Adding an optional evening saliva sample
   • Customize barcodes to fit your printable label templates

3. Analyzing the CARWatch log data.
   This includes:

   • Extracting the sampling timestamps from the log data
   • Extracting the self-reported awakening times (if available)
   • Merging the time information with the cortisol measures

Installation

carwatch requires Python >=3.8. First, install a compatible version of Python (e.g. using miniconda). Then, open a terminal (or Anaconda prompt) and install the carwatch package via pip:

pip install carwatch

Alternatively, you can download the package directly from the source repository on GitHub:

git clone https://github.com/mad-lab-fau/carwatch-python.git
cd carwatch-python
pip install .

For Developer

If you are a developer and want to contribute to carwatch you can install an editable version of the package from a local copy of the repository.

carwatch uses poetry to manage dependencies and packaging. Once you installed poetry, run the following commands to clone the repository, initialize a virtual environment and install all development dependencies:

git clone https://github.com/mad-lab-fau/carwatch-python.git
cd carwatch-python
poetry install

Usage

carwatch can be used both programmatically and with the provided command line interface (CLI).

The core functionalities of the carwatch package are

• creating a QR-Code for configuring the CARWatch App (Preparation),
• creating a PDF with printable barcode labels for the saliva sampling tubes (Preparation),
• and extracting the sampling times for the CARWatch app log recordings (Postprocessing).

Programmatic Usage

For the preparation steps, the study details can be specified using the Study class. Participant IDs can also be parsed from a *.csv file, when the path to it is specified as subject_path, and the corresponding column as subject_column.

Some basic examples are given below. For more information about the available parameters, please refer to the documentation of the mentioned classes.

Study Setup Example

from carwatch.utils import Study

study = Study(
    study_name="ExampleStudy",
    num_days=3,
    num_subjects=15,
    num_samples=5,
    subject_prefix="VP_",
    has_evening_sample=True,
    start_sample_from_zero=True,
)

Barcode Generation Example

For generating barcodes, the LabelGenerator class can be used, receiving a Study instance as a parameter. Your custom printing label layout can be specified using the CustomLayout class. By default, the AveryZweckform J4791 layout is used.
To start the PDF generation, call the generate method of the LabelGenerator class. The output PDF will be exported to the directory specified by output_dir (per default: the current working directory).

from carwatch.utils import Study
from carwatch.labels.print_layout import CustomLayout
from carwatch.labels.label_generator import LabelGenerator

study = Study(
    study_name="ExampleStudy",
    num_days=3,
    num_subjects=15,
    num_samples=5,
    subject_prefix="VP_",
    has_evening_sample=True,
    start_sample_from_zero=True,
)
generator = LabelGenerator(study=study, add_name=True, has_barcode=True)
layout = CustomLayout(
    num_cols=3,
    num_rows=4,
    left_margin=3,
    right_margin=3,
    top_margin=2,
    bottom_margin=2,
    inter_col=0.2,
    inter_row=0.5,
)
generator.generate(output_dir=".", debug=True, layout=layout)

QR-Code Generation Example

For generating the QR-Code, the QrCodeGenerator class can be used, again receiving a Study instance as a parameter. The saliva_distances parameter specifies the desired distances between saliva samples in minutes. The resulting QR-Code for setting up the CARWatch App will be exported to the directory specified by output_dir directory (per default: the current working directory).

from carwatch.qr_codes import QrCodeGenerator
from carwatch.utils import Study

if __name__ == "__main__":
    study = Study(
        study_name="ExampleStudy",
        num_days=3,
        num_subjects=15,
        num_samples=5,
        subject_prefix="VP_",
        has_evening_sample=True,
        start_sample_from_zero=True,
    )
    generator = QrCodeGenerator(study=study, saliva_distances=[10, 10, 10], contact_email="dum@my.com")
    generator.generate(output_dir=".")

Postprocessing Example

To be added

Command Line Interface

For the preparation steps, carwatch also provides a CLI for more convenient usage. Make sure you installed the carwatch package with pip install carwatch.

After that, you can simply run the TUI (terminal user interface) by running

prepare_study tui

in a terminal session.

This will implicitly run the scripts/prepare_study.py script, which will guide you through the preparation steps. You will then be prompted to enter all the required information step-by-step. The desired output files will be automatically generated for you.

The regular command line interface (CLI) can be used by running

prepare_study run

For more information about the prompted commands please run:

prepare_study run --help

License

This project is licensed under the MIT License. See the LICENSE file for details.

Contributing

We welcome contributions to carwatch! For more information, have a look at the Contributing Guidelines.

Citing carwatch

If you use carwatch in your work, please report the version you used in the text. Additionally, please also cite our paper published in Psychoneuroendocrinology:

Richer, R., Abel, L., Küderle, A., Eskofier, B. M., & Rohleder, N. (2023). CARWatch – A smartphone application for 
improving the accuracy of cortisol awakening response sampling. Psychoneuroendocrinology, 151, 106073. 
https://doi.org/10.1016/j.psyneuen.2023.106073

Contact

If you have any questions or feedback about CARWatch, please contact Robert Richer.