mvm7 / EsophagusSegmentation Public

Notifications You must be signed in to change notification settings
Fork 0
Star 1

CNN architectures for multiclass semantic segmentation of esophageal diseases

1 star 0 forks Branches Tags Activity

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 5 Commits
models		models
LICENSE		LICENSE
README.md		README.md
main.py		main.py
modifyImage.py		modifyImage.py
predict.py		predict.py
prepareData.py		prepareData.py
train.py		train.py

Repository files navigation

Training results (50 epochs)

Model	accuracy
FCN	0.889
SegNet	0.906
U-Net	0.901
VGG16 U-Net	0.962
ResNet101 U-Net	0.977
VGG16 UNet 3+	0.977
ResNet101 UNet 3+	0.978

Results. Model: VGG16 UNet 3+

Comparison of result (left image) and true segmentation (right image). Model: VGG16 UNet 3+

About

CNN architectures for multiclass semantic segmentation of esophageal diseases

deep-learning keras cnn fcn unet segnet semantic-segmentation unet3plus multiclass-segmentation

Report repository

Releases

No releases published

Packages

No packages published

Languages

Python 100.0%