Skip to content

implementation of unet for the segmentation the MRIs and MLP for the classification of five heart diseases

Notifications You must be signed in to change notification settings

alineyyy/deep-learning-classification-mri

Repository files navigation

All documents are defined according to the following meanings respectively:

dataprocess_seg.ipynb:

Pre-processing of the raw data before the segmentation to obtain Train_img folder, Test_img folder, Mask_LV folder and Mask_noLV folder.

segmentaion.ipynb:

Segmentation the LV part for the test set. We obtain the result Test_Mask_LV(input_mask) folder which are saved into Test_Mask_LV folder after the operation of a thresold segmentation.

dataprocess_train.ipynb:

Pre-processing of the train data before the classification. We calculatemedical metrics, and save them into DataTrain.csv.(Before calculating the medical metrics, we obtain Mask_RV folder, Mask_Myocardium folder firstly)

dataprocess_test.ipynb:

Pre-processing of the test data before the classification. We calculate the medical metrics,and save them into Test(inputmask).csv. (Before calculating the medical metrics, we obtain Test_Mask_RV folder, Test_Mask_Myocardium folder firstly)

classification.ipynb:

Train the model of classification and predicate the catagory of the patients in the test set. The results are saved into prediction.csv.

About

implementation of unet for the segmentation the MRIs and MLP for the classification of five heart diseases

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published