mahmoodlab / toad Goto Github PK
View Code? Open in Web Editor NEWAI-based pathology predicts origins for cancers of unknown primary - Nature
Home Page: http://toad.mahmoodlab.org
License: GNU Affero General Public License v3.0
AI-based pathology predicts origins for cancers of unknown primary - Nature
Home Page: http://toad.mahmoodlab.org
License: GNU Affero General Public License v3.0
Hi, thank you for creating such an amazing works. I'm trying to create a visualization heatmaps using the model from toad, what I understand is you create it using the heatmaps code in the CLAM repository. I already try it, but it didn't works, can you help me to figure out why, or maybe you have the procedures to do it?
my problem when directly apply the create heatmaps in clam is the "sex" problem which is not used/available in CLAM package
First of all, I would like to appreciate all the hard work that has been put into this research article.
Moving to the problem. So, I was going through the GDC tool for downloading Diagnostic Slides for patients however I was not able to find the label site label i.e primary or metastatic? I have read your paper multiple times to understand how to extract labels but the instructions are not very clear and also went through the documentation of the TCGA GDC portal but was unable to find it.
Therefore could you point out on how you were able to extract labels typically site i.e primary pr metastatic?
I would really appreciate the help. Thanks
Hi, I tried to running the command on Colab
!python create_splits.py --task dummy_mtl_concat --seed 1 --k 1
And I am getting this error:
Traceback (most recent call last):
File "create_splits.py", line 25, in
dataset = Generic_WSI_MTL_Dataset(csv_path = '/content/drive/MyDrive/Colab Notebooks/MTDL/TOAD/dataset_csv/dummy_dataset.csv',
File "/content/drive/My Drive/Colab Notebooks/MTDL/TOAD/datasets/dataset_mtl_concat.py", line 71, in init
slide_data = self.df_prep(slide_data, self.label_dicts, self.label_cols)
File "/content/drive/My Drive/Colab Notebooks/MTDL/TOAD/datasets/dataset_mtl_concat.py", line 133, in df_prep
data.at[i, 'label'] = label_dicts[0][key]
KeyError: 'Esophagogogastric'
Could anyone please explain why this is happening and how to resolve this error?
Could you indicate whether the label of the metastatic sample you used was consistent with the primary site or with the section location? If the primary colorectal cancer has metastasized to the liver, do you use images from the colorectal or liver? What is the label when using slices from the liver? I would appreciate it if you could answer me!
Are model weights available somewhere? I can't find them in the Mahmood Lab Huggingface profile. I would like to perform inference of this model with the WSI of a personal case diagnosed with a CUP.
Hello,thanks for your great work!
Could you tell me from which layer to extract the feature vectors for visualization?
Hello and thank you for your great work!
Could you tell me if there's any way I could get the model checkpoints of this particular "Classification of tumours metastasized to the liver" network?
Thank you!
Hi,
I have read the documentation and it is a really great project but can you add a video regarding the steps to perform in a sequential manner.
Thank you
Is it possible that detailed TCGA and CPTAC projects used as well as data cleaning procedure can be provided?
Thank you!
Hello and thank you for sharing your work!
Could you please provide the last checkpoint for the pretrained model?
Thank you in advance, Lucia
Where can I find the patient‘s data of histology slides
Hi, I tried to running the evaluation command
CUDA_VISIBLE_DEVICES=0 python eval_mtl_concat.py --drop_out --k 1 --models_exp_code dummy_mtl_sex_s1 --save_exp_code dummy_mtl_sex_s1_eval --task dummy_mtl_concat --results_dir results --data_root_dir DATA_ROOT_DIR
I changed the task bellow code
--task study_v2_mtl_sex
into
--task dummy_mtl_concat
because the study_v2_mtl_sex task couldn't found. But then I ran into this error:
Traceback (most recent call last):
File "eval_mtl_concat.py", line 122, in
model, results_dict = eval(split_dataset, args, ckpt_paths[ckpt_idx])
File "/media/chingwei/bf764154-3611-460c-9bfd-4efed447a219/chingwei/Nabila/TOAD-NGS/utils/eval_utils_mtl_concat.py", line 40, in eval
results_dict = summary(model, loader, args)
File "/media/chingwei/bf764154-3611-460c-9bfd-4efed447a219/chingwei/Nabila/TOAD-NGS/utils/eval_utils_mtl_concat.py", line 120, in summary
cls_test_error /= len(loader)
ZeroDivisionError: float division by zero
Could you please provide any guidance on how to resolve this error?
OS : ubuntu
torch : 1.10.1+cu111
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.