Image Segmentation
Flair
Keras
tensorflow
medical-imaging
white-matter-hyperintensities
mri
deep-learning
neurology
multiple-sclerosis
Instructions to use Bawil/wmh_leverage_normal_abnormal_segmentation with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Flair
How to use Bawil/wmh_leverage_normal_abnormal_segmentation with Flair:
from flair.models import SequenceTagger tagger = SequenceTagger.load("Bawil/wmh_leverage_normal_abnormal_segmentation") - Keras
How to use Bawil/wmh_leverage_normal_abnormal_segmentation with Keras:
# Available backend options are: "jax", "torch", "tensorflow". import os os.environ["KERAS_BACKEND"] = "jax" import keras model = keras.saving.load_model("hf://Bawil/wmh_leverage_normal_abnormal_segmentation") - Notebooks
- Google Colab
- Kaggle
Download transunet/tables/comprehensive_results.csv from Bawil/wmh_leverage_normal_abnormal_segmentation: direct link, hf CLI and curl.
- Browser
- Download file 511 Bytes
-
https://huggingface.co/Bawil/wmh_leverage_normal_abnormal_segmentation/resolve/f4d9c79455fcf4b46331a1ac32fcc23a0646afd4/transunet/tables/comprehensive_results.csv
- Command line
-
hf download hf://Bawil/wmh_leverage_normal_abnormal_segmentation@f4d9c79455fcf4b46331a1ac32fcc23a0646afd4/transunet/tables/comprehensive_results.csv
-
curl -L -o comprehensive_results.csv https://huggingface.co/Bawil/wmh_leverage_normal_abnormal_segmentation/resolve/f4d9c79455fcf4b46331a1ac32fcc23a0646afd4/transunet/tables/comprehensive_results.csv
511 Bytes
| Scenario,Accuracy,Precision,Recall,Specificity,Dice,IoU,HD95,ASSD | |
| Binary Classification (Processed),0.986004278526543,0.3461836987030803,0.9649564156425705,0.9997302921073945,0.5095598516138679,0.3418854773044586,, | |
| Three-class Classification (Processed),0.9946109937584918,0.6029034779921095,0.8342410787285695,0.9987378954133063,0.6999530360371915,0.5384059548377991,, | |
| Statistical Analysis,Dice p=0.0000,Dice t=7.9888,Dice Δ=0.1449,Dice ES=0.4778,IoU p=0.0000,IoU Δ=0.1457,HD95 Δ=3.0926px,ASSD Δ=-1.2769px | |