https://github.com/devidduma/ms-detect
Multiple sclerosis detection from MRI images.
https://github.com/devidduma/ms-detect
Last synced: about 1 month ago
JSON representation
Multiple sclerosis detection from MRI images.
- Host: GitHub
- URL: https://github.com/devidduma/ms-detect
- Owner: devidduma
- License: apache-2.0
- Created: 2025-01-11T09:30:04.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-01-11T09:39:35.000Z (over 1 year ago)
- Last Synced: 2025-01-11T10:20:04.368Z (over 1 year ago)
- Language: Jupyter Notebook
- Size: 0 Bytes
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# ms-detect
Multiple sclerosis detection from MRI images.
#### Dataset
- Dataset from [eHealth lab](http://www.medinfo.cs.ucy.ac.cy/index.php/facilities/32-software/218-datasets) of Department of Computer Science at University of Cyprus.
#### Recipe
- Pre-processing: apply continuous or discrete wavelet transformations to MRI images.
- Binary classification with CNN: predict whether the input image is healthy or unhealthy using a CNN built with Keras framework.
#### Presentations
- Wavelet Transformations: ([ipynb](presentations/preprocessing/mswavelet_presentation.ipynb))
- Batch Wavelet Transformations: ([ipynb](presentations/preprocessing/dataset_preprocessing_batch.ipynb))
- CNNNet results: ([CNNNet ipynb](presentations/cnn/cnnnet_results.ipynb))
#### Results
val_loss: 0.4376 - val_accuracy: 0.8297 - val_mse: 0.1396 - val_mae: 0.2852