search

johnnydfci / DL-parenchymal

Deep Learning-selected Liver Parenchyma for Categorizing Hepatic Steatosis on Multinational Chest CT
Apache License 2.0
3 stars 1 forks source link

WARNING! Cannot run postprocessing because the postprocessing file is missing. Make sure to run consolidate_folds in the output folder of the model first! #2

Open juoliii opened 3 days ago

juoliii commented 3 days ago

WARNING! Cannot run postprocessing because the postprocessing file is missing. Make sure to run consolidate_folds in the output folder of the model first!

johnnydfci commented 2 days ago

Thank you for your comments. This is a warning but not an error. I assume the test segmentation can still be inferred by the nnUNet model. Thus, please proceed to run the liver_seg_curated_into_top1_mask.ipynb notebook for post-processing the segmented mask to retain the largest connected component. For further details, please refer to the Implementation Steps documentation.

juoliii commented 2 days ago

Thank you for your comments. This is a warning but not an error. I assume the test segmentation can still be inferred by the nnUNet model. Thus, please proceed to run the liver_seg_curated_into_top1_mask.ipynb notebook for post-processing the segmented mask to retain the largest connected component. For further details, please refer to the Implementation Steps documentation.

but i could not generate the /test_label_in_nii/* file，and liver_seg_various_accuracy_compute.ipynb can not run, please tell me how to fix？

johnnydfci commented 2 days ago

Hi there,

The images segmented by nnUNet are stored in nnUNet_raw_data_base/nnUNet_test_data/test_seg_in_nii/raw. After post-processing, the images are saved in nnUNet_raw_data_base/nnUNet_test_data/test_seg_in_nii.

To compare the automated segmentation with manual segmentation, you need to download the manual segmentation files into the directory: nnUNet_raw_data_base/nnUNet_test_data/test_label_in_nii. These manual segmentation datasets are freely available online:

Google Drive: Curated CT images and liver segmentations Baidu Wangpan: Link (Passcode: 2022) Let me know if you need further assistance!

juoliii commented 2 days ago

Hi there,

The images segmented by nnUNet are stored in nnUNet_raw_data_base/nnUNet_test_data/test_seg_in_nii/raw. After post-processing, the images are saved in nnUNet_raw_data_base/nnUNet_test_data/test_seg_in_nii.

To compare the automated segmentation with manual segmentation, you need to download the manual segmentation files into the directory: nnUNet_raw_data_base/nnUNet_test_data/test_label_in_nii. These manual segmentation datasets are freely available online:

Google Drive: Curated CT images and liver segmentations Baidu Wangpan: Link (Passcode: 2022) Let me know if you need further assistance!

ok i see , thank you