INTEGRAL-Radiomics

This repository contains the model and code for using the INTEGRAL-Radiomics screen-detected pulmonary nodule malignancy model published in:

Warkentin MT, Al-Sawaihey H, Lam S, et al Radiomics analysis to predict pulmonary nodule malignancy using machine learning approaches Thorax Published Online First: 09 January 2024. doi: 10.1136/thorax-2023-220226

If you have any comments or questions, please file an Issue.

Usage

To use the INTEGRAL-Radiomics model, you simply need to install the R package integralrad and it handles all the necessary R and Python dependencies to perform the PyRadiomics feature extraction and model predictions.

# Install `integralrad`
remotes::install_github("mattwarkentin/INTEGRAL-Radiomics")

Once installed, we can use extract_radiomics(...) to perform only the feature extraction, or predict_integral_radiomics(...) to execute the entire prediction pipeline.

library(integralrad)

# Path to CSV with required columns (see `?predict_integral_radiomics` for details)
input <- "path/to/csv"

preds <- predict_integral_radiomics(input)

Command-Line Interface

We have also included a command-line interface to simplify getting predictions using the INTEGRAL-Radiomics model. For input, simply pass in a path to a CSV file (--input) that contains the relevant columns described in ?predict_integral_radiomics(). You must also provide the path to where the output CSV should be saved on disk (--output).

After installing the package (as described above), you simply need to install the integral-radiomics CLI by running the following R code:

integralrad::install_integralrad_cli()

After restarting your shell, you should be able to run the following:

integral-radiomics --input=<path-to-input> --output=<path-to-output>

Citation

If you use this model, please cite the following article:

Warkentin MT, Al-Sawaihey H, Lam S, et al Radiomics analysis to predict pulmonary nodule malignancy using machine learning approaches Thorax Published Online First: 09 January 2024. doi: 10.1136/thorax-2023-220226

@article {Warkentinthorax-2023-220226,
  author = {Matthew T Warkentin and Hamad Al-Sawaihey and Stephen Lam and Geoffrey Liu and Brenda Diergaarde and Jian-Min Yuan and David O Wilson and Sukhinder Atkar-Khattra and Benjamin Grant and Yonathan Brhane and Elham Khodayari-Moez and Kiera R Murison and Martin C Tammemagi and Kieran R Campbell and Rayjean J Hung},
  title = {Radiomics analysis to predict pulmonary nodule malignancy using machine learning approaches},
  elocation-id = {thorax-2023-220226},
  year = {2024},
  doi = {10.1136/thorax-2023-220226},
  publisher = {BMJ Publishing Group Ltd},
  issn = {0040-6376},
  URL = {https://thorax.bmj.com/content/early/2024/01/08/thorax-2023-220226},
  eprint = {https://thorax.bmj.com/content/early/2024/01/08/thorax-2023-220226.full.pdf},
  journal = {Thorax}
}