Pretrained nnU-Net Model for 3D semantic image segmentation of Lipoma and Atypical Lipomatous Tumor
Description
Pretrained nnU-Net model used in the study: Multi-Center External Validation of an Automated Method Segmenting and Differentiating Atypical Lipomatous Tumors from Lipomas Using Radiomics and Deep-Learning on MRI (Note, ref will be included soon)
The model is build with the nnU-Net framework v1 and comprises an ensemble of five 3d models. In order to use this model, first follow instructutions on setting up the nnU-Net framework. Next, download, extract and move the files provided to your RESULTS_FOLDER, defined here.
Once your data and model is prepared you can run inference with:
nnUNet_predict -i FOLDER_WITH_TEST_CASES -o OUTPUT_FOLDER_MODEL1 -tr nnUNetTrainerV2 -ctr nnUNetTrainerV2CascadeFullRes -m 3d_fullres -p nnUNetPlansv2.1 -t Task803_LipomaALT
Note, we also trained five 2d models, but nnU-Net did not select them for ensembling so they are not provided here.
Files
Task803_LipomaALT.zip
Files
(1.7 GB)
| Name | Size | Download all |
|---|---|---|
|
md5:f20df58a75c0ed31946bcee60ac7c954
|
1.7 GB | Preview Download |
Additional details
References
- F. Isensee, P. Jaeger, S. Kohl, J. Petersen, and K. Maier-Hein, "nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation," Nature Methods, vol. 18, pp. 1–9, Feb. 2021, doi: 10.1038/s41592-020-01008-z.