Part of the xLUNGS project. This directory contains code and the TLOS dataset associated with the article
H. Baniecki, B. Sobieski, P. Bombiński, P. Szatkowski, P. Biecek. Hospital Length of Stay Prediction Based on Multi-modal Data towards Trustworthy Human-AI Collaboration in Radiomics. International Conference on Artificial Intelligence in Medicine, 2023. https://arxiv.org/abs/2303.09817
@inproceedings{baniecki2023hospital,
title = {{Hospital Length of Stay Prediction Based on Multi-modal Data
towards Trustworthy Human-AI Collaboration in Radiomics}},
author = {Hubert Baniecki and Bartlomiej Sobieski and Przemysław Bombiński
and Patryk Szatkowski and Przemysław Biecek},
booktitle = {International Conference on Artificial Intelligence in Medicine},
year = {2023}
}
The preprocessed TLOS dataset in the tabular form is available in data under the CC-BY-NC-ND-4.0 license. Raw X-ray data including images and radiology reports cannot be openly shared due to privacy concerns. For inquires, please write at [email protected].
Exemplary code for performing the described analysis is available in code under the MIT license. It is accompanied by the key metadata checkpoints in results. For inquires, please write at [email protected].
The analysis was performed using the following software:
- R v4.2.1
survex
v0.2.2 updated at https://github.com/ModelOriented/survex/tree/xlungs-trustworthy-los-predictionmlr3proba
v0.4.17 accessed from https://github.com/mlr-org/mlr3probamboost
v2.9.7 updated with the following fix boost-R/mboost#118
- Python v3.8.13
pyradiomics
v3.0.1pydicom
v2.3.0
This work was financially supported by the Polish National Center for Research and Development grant number INFOSTRATEG-I/0022/2021-00
, and carried out with the support of the Laboratory of Bioinformatics and Computational Genomics and the High Performance Computing Center of the Faculty of Mathematics and Information Science, Warsaw University of Technology.