This repository is committed to ongoing research focused on developing models for the cleansing of pulsatile signals, specifically through the integration of artifact augmentation. An research paper, which delineates the methodologies employed alongside artifact generation, is presently being drafted. The paper is scheduled to be submitted to arXiv by June 2024. Following its registration, a link to the document will be made accessible. Please ensure to carefully review the license terms, as they are included.
Please note that this version is primarily for developmental use and may contain bugs or incomplete features. Users are advised to proceed with caution and are encouraged to contact the authors for more information or to discuss potential collaborations.
Unauthorized Use Prohibited: Unauthorized use, duplication, or distribution of this software and its associated documentation is strictly prohibited without prior written permission from the author(s). Please contact the author(s) to obtain permission before using this software in any manner not explicitly authorized.
If you plan to use our simulator or model structure for your research or if you need specific details about the implementation, please contact the author(s) before proceeding. This will ensure you have the most up-to-date information and guidance. Contact details can be found below:
-
Email: [[email protected], [email protected]]
-
Institution: [Department of Cancer AI and Digital Health, Graduate School of Cancer Science & Policy, National Cancer Center, KOREA]
J. Kim initiated this project by establishing the concept, conducting literature research, and drafting codes for artifact generation. Subsequently, K. Park conducted a proof of concept, significantly enhancing the generation process and contributing to major improvements in its sophistication.
This project is organized into several directories, each serving a specific purpose in the research and development process:
Contains essential libraries and functions needed for the project:
artifact_augmentation.py
: Functions for augmenting artifacts during model training.artifact_simulation.py
: Functions required by the Artifact Simulator.
Contains scripts for training different models:
ModelStructure_DI.py
: Basic structure for the DI model.train_DI.py
: Training script for the DI model.train_DI_D.py
: Training script for the DI-D model variant.train_DI_A.py
: Training script for the DI-A model variant.ModelStructure_DA.py
: Basic structure for the DA model.train_DA.py
: Training script for the DA model.train_DA_D.py
: Training script for the DA-D model variant.train_DA_A.py
: Training script for the DA-A model variant.
Contains sample files for the simulator:
ABP_60s_sample.npy
: Sample file for artifact simulation.
Houses the artifact simulator:
Example_for_Artifact_Simulation.ipynb
: Jupyter notebook demonstrating artifact simulation.