The project was accepted as a short paper at the Federated Conference on Computer Science and Information Systems (FedCSIS) International Workshop on Artificial Intelligence in Machine Vision and Graphics (AIMaViG), and submitted for inclusion in the IEEE Xplore Digital Library and published electronically in a volume of Annals of Computer Science and Information Systems” (ACSIS).
This repository contains the data set, preprocessing pipeline, CNN architecture implementations, model training and evaluation pipelines, and a simple web application for visualizing the results.
Please find my Final Degree Project in my portfolio here: https://uberg.me/portfolio/
If you wish to cite the paper, please do it as such:
@inproceedings{Uberg_2022,
author = {Uberg, Lewi L. and Kadry, Seifedine},
doi = {10.15439/2022f69},
url = {https://doi.org/10.15439%2F2022f69},
year = {2022},
month = {sep},
pages = {201-204},
publisher = {{IEEE}},
title = {Analysis of Brain Tumor Using {MRI} Images},
booktitle = {Annals of Computer Science and Information Systems}
}
Please see CITATION.cff for the full citation information.
Lie Uberg, L. (2022). Tumorclass.info (Version 0.1.0) [Computer software]. https://github.com/lewiuberg/tumorclass.info
@software{Lie_Uberg_Tumorclass_info_2022,
author = {Lie Uberg, Lewi},
license = {MIT},
month = {5},
title = {{Tumorclass.info}},
url = {https://github.com/lewiuberg/tumorclass.info},
version = {0.1.0},
year = {2022}
}
The preprocessing pipeline made for the original REMBRANDT dataset can be found in the preprocessing notebook.
The augmentation pipeline made for the manually labeled REMBRANDT dataset can be found in the augmentation notebook.
The different model training pipelines can be found in the notebooks directory.
The runs from the main.py script can be found in the app directory. This application utilizes the many of the files and subdirectories in project root directory. In order to run a local version of the application, clone this repository. After cloning, Poetry can be used to install dependencies. Since the main.py
file is not located in the project root, you may need to export the python path using this command:
export PYTHONPATH=${PWD}/:${PWD}/app/
With a terminal open at the root of the project directory, run the following command to start the application:
poetry run python app/main.py
When the application is running, you can access it at the following URL http://localhost:8000