ZeyuGaoAi/CLAM_PreProcessing

CLAM-based preprocessing with more fundation models

stars

commits

Python

primary language

Dec 12, 2024

updated

README

CLAM_PreProcessing

CLAM-based preprocessing with more foundation models

Version Name	Link
uni_v1	https://huggingface.co/MahmoodLab/UNI
conch_v1	https://huggingface.co/MahmoodLab/CONCH
conch_v1_5	https://huggingface.co/MahmoodLab/TITAN
h_optimus_0	https://huggingface.co/bioptimus/H-optimus-0
virchow2	https://huggingface.co/paige-ai/Virchow2
gigapath	https://huggingface.co/prov-gigapath/prov-gigapath
hibou_l	https://huggingface.co/histai/hibou-L

Patch Tessellation

python create_patches_fp.py --source /home/shared/su123/TCGA_ORI/TCGA-OV/ \
       --save_dir /home/shared/su123/TCGA_Embed/TCGA-OV/clam_gen_1024   \
       --step_size 1024 --patch_size 1024 --seg --patch --stitch --patch_level 0 --preset tcga.csv

Feature Extraction

set up model_name: uni_v1, conch_v1, conch_v1_5, h_optimus_0, virchow2, gigapath, hibou_l

python extract_features_fp.py --data_h5_dir /home/shared/su123/TCGA_Embed/TCGA-RCC/clam_gen_1024/ \
    --anno_dir /home/z/zeyugao/dataset/WSIData/TCGA-RCC/annotation/ \
    --data_slide_dir /home/shared/su123/TCGA_ORI/TCGA-RCC/ \
    --csv_path /home/shared/su123/TCGA_Embed/TCGA-RCC/clam_gen_1024/process_list_autogen.csv --model_name virchow2\
    --feat_dir /home/shared/su123/TCGA_Embed/TCGA-RCC/clam_gen_1024/virchow2/ \
    --suffix "_0_1024" --patch_size 1024

Contributors

57 commits

18 commits

4 commits

4 commits

Similar repos

mahmoodlab/CLAM

Open source tools for computational pathology - Nature BME

1,729

shared contributors

pathology-data-mining/Mussel

Computational pathology toolkit: tiling, feature extraction, and annotation for whole-slide images…

shared contributors

birkhoffkiki/PrePATH

PrePATH: A parallel toolkit for Preprocessing WSI

WonderLandxD/PIANO

An easy-to-use PyTorch library for Pathology Image ANalysis Orchestrator (PIANO), including…

Eitan177/TridentEdited

AtlasAnalyticsLab/AtlasPatch

AtlasPatch: An Efficient and Scalable Tool for Whole Slide Image Preprocessing

mahmoodlab/TRIDENT

Toolkit for large-scale whole-slide image processing.

626

IUCompPath/aegis

Multiple Instance Learning framework for multi-modal classification with histopathology data

Languages

Python

79.4%

Jupyter Notebook

16.2%

Shell

4.5%

ZeyuGaoAi/CLAM_PreProcessing

CLAM-based preprocessing with more fundation models

stars

commits

Python

primary language

Dec 12, 2024

updated

README

CLAM_PreProcessing

CLAM-based preprocessing with more foundation models

Version Name	Link
uni_v1	https://huggingface.co/MahmoodLab/UNI
conch_v1	https://huggingface.co/MahmoodLab/CONCH
conch_v1_5	https://huggingface.co/MahmoodLab/TITAN
h_optimus_0	https://huggingface.co/bioptimus/H-optimus-0
virchow2	https://huggingface.co/paige-ai/Virchow2
gigapath	https://huggingface.co/prov-gigapath/prov-gigapath
hibou_l	https://huggingface.co/histai/hibou-L

Patch Tessellation

python create_patches_fp.py --source /home/shared/su123/TCGA_ORI/TCGA-OV/ \
       --save_dir /home/shared/su123/TCGA_Embed/TCGA-OV/clam_gen_1024   \
       --step_size 1024 --patch_size 1024 --seg --patch --stitch --patch_level 0 --preset tcga.csv

Feature Extraction

set up model_name: uni_v1, conch_v1, conch_v1_5, h_optimus_0, virchow2, gigapath, hibou_l

python extract_features_fp.py --data_h5_dir /home/shared/su123/TCGA_Embed/TCGA-RCC/clam_gen_1024/ \
    --anno_dir /home/z/zeyugao/dataset/WSIData/TCGA-RCC/annotation/ \
    --data_slide_dir /home/shared/su123/TCGA_ORI/TCGA-RCC/ \
    --csv_path /home/shared/su123/TCGA_Embed/TCGA-RCC/clam_gen_1024/process_list_autogen.csv --model_name virchow2\
    --feat_dir /home/shared/su123/TCGA_Embed/TCGA-RCC/clam_gen_1024/virchow2/ \
    --suffix "_0_1024" --patch_size 1024