This Repository consists of various Models using UNET for Performing Bio-Medical Image Segmenation of the covid-19lesion from The CT-SCAN
The Dataset: https://www.kaggle.com/maedemaftouni/large-covid19-ct-slice-dataset is being built by curating data from 7 public datasets. Three of these datasets had shared COVID-19 lesion masks. This dataset merges the COVID-19 lesion masks and their corresponding frames of these 3 public datasets, with 2729 image and ground truth mask pairs. All different types of lesions are mapped to white color for consistency across datasets.
THE UNET MODELS and their respective lesion segmentation results are:
1 UNET
2 RESUNET
3 VGG16UNET
4 VGG19UNET