frankkramer-lab / covid19.miscnn Goto Github PK
View Code? Open in Web Editor NEWRobust Chest CT Image Segmentation of COVID-19 Lung Infection based on limited data
License: GNU General Public License v3.0
Robust Chest CT Image Segmentation of COVID-19 Lung Infection based on limited data
License: GNU General Public License v3.0
From what I understand, the "iterations" parameter is the steps per epoch. In your script, you fixed that as 150. During my training, I tried to leave that as default, hence None, hoping that it will run through all the created patches. However, it showed 10 as the steps per epoch during the training. Does this mean the network only trains on 10*batch_size(=2) = 20 patches per epoch? This is way too few.
Do I misunderstand something here? Thanks a lot for your help!
Dear @muellerdo
Thanks for using our dataset. It would be highly appreciated if you could add following citation in your paper.
@article{COVID-19-SegBenchmark,
title={Towards Efficient COVID-19 CT Annotation: A Benchmark for Lung and Infection Segmentation},
author={Ma Jun and Wang Yixin and An Xingle and Ge Cheng and Yu Ziqi and Chen Jianan and Zhu Qiongjie and Dong Guoqiang and He Jian and He Zhiqiang and Ni Ziwei and Yang Xiaoping},
journal={arXiv preprint arXiv:2004.12537},
year={2020}
}
Best regards,
Jun
Hello, I would like to ask you, what kind of running program do you use? I ran on the server and trained 5 times for 3 days. The speed is too slow, and you trained 1000 times, so can you tell me what you used to run the program method.
To-do:
Implement GIF animation of 3D volume via matplotlib
Hi,
I have a confusion in the use of the GPUs. the function Neural_Network use le variable gpu_number (version 0.34), but if the GPU=0 is busy and I get an error. The last version of miscnn use the variable multi_gpu=boolean.
Could you explain to me how to set the use of a particular gpu.
Thank
https://gitee.com/junma11/COVID-19-CT-Seg-Benchmark
Some kind of rank list for more transparency?
Organization of challenge?
-> more data required
Hi,
I've noticed something that Patchwise Overlap is 80x80x40 in the paper but it is 80x80x30 in the code. If overlap shape is not half of patch shape , is that okey ?
IEEE Transactions on Medical Imaging:
Special Issue on "Imaging-based Diagnosis of COVID-19"
IEEE Journal of Biomedical and Health Informatics:
Special Issue on "AI-driven Informatics, Sensing, Imaging and Big Data Analytics for Fighting the COVID-19 Pandemic"
Medical Image Analysis:
Special Issue on "Intelligent Analysis of COVID-19 Imaging Data"
To-do:
Calculate Dice Similarity Coefficient, Sensitivity and Specificity for each class.
Average (mean) Lung left and Lung Right to class "Lungs"
Hi @frankkramer
When I check the dataset, There are 2 part coranacases and radiopedia. For the radiopedia part the images set up 0-255. But for the coronacases part HU range -1250 to 250. I wonder how do you overcome this problem.
Hi,
I have a question regarding the patch shape in the script run_miscnn.py. It is set as (160, 160, 80). Is the z-axis, 80 represents the number of slices in the patch? However, in your source code of Miscnn, you wrote:
patch_shape (integer tuple): Size and shape of a patch. The variable has to be defined as a tuple.
For Example: (64,128,128) for 64x128x128 patch cubes.
Be aware that the x-axis represents the number of slices in 3D volumes.
Can you clarify this for me?
Best wishes,
LHJ
Hey guys,
Great work. I am trying to reproduce the results from your paper.
Here is the error I am getting
2020-07-22 15:59:09.423261: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1
Traceback (most recent call last):
File "scripts/run_preprocessing.py", line 25, in <module>
from miscnn.evaluation.cross_validation import split_folds
ImportError: cannot import name 'split_folds
Please have a look and suggest any fix
Thanks
The contradictory required version of numpy & miscnn, numpy & tensorflow made it impossible to run the code. Furthermore, NIFTI_interface is not defined. I managed to get into the source code but can't debug it due to lack of software coding knowledge.
Hi @muellerdo
I am using your MIScnn framework from installing directly from Github. I am not using (pip installation). But in cross validation file I can't file "load_csv2fold". So in run_miscnn.py line: 32 give error.
Add 2 functions to CV of MIScnn:
Sir, firstly it is great repository and I want to say thank you.
Could you give more information about how did you slice 3D images into patches (I believe patch means slices of 2D image which is extracted from 3D images, am I wrong)?
Issue:
arm64 architecture needs tensorflow-macos, which is currently not recognized by aucmedi
Solution:
aucmedi-macos that depends on tensorflow-macos
File "scripts/run_miscnn.py", line 120, in
save_models=False)
File "/home/ubuntu/anaconda3/lib/python3.7/site-packages/miscnn/evaluation/cross_validation.py", line 166, in run_fold
iterations=iterations, callbacks=cb_list)
File "/home/ubuntu/anaconda3/lib/python3.7/site-packages/miscnn/neural_network/model.py", line 201, in evaluate
max_queue_size=self.batch_queue_size)
File "/home/ubuntu/anaconda3/lib/python3.7/site-packages/tensorflow/python/keras/engine/training.py", line 108, in _method_wrapper
return method(self, *args, **kwargs)
File "/home/ubuntu/anaconda3/lib/python3.7/site-packages/tensorflow/python/keras/engine/training.py", line 1098, in fit
tmp_logs = train_function(iterator)
File "/home/ubuntu/anaconda3/lib/python3.7/site-packages/tensorflow/python/eager/def_function.py", line 780, in call
result = self._call(*args, **kwds)
File "/home/ubuntu/anaconda3/lib/python3.7/site-packages/tensorflow/python/eager/def_function.py", line 840, in _call
return self._stateless_fn(*args, **kwds)
File "/home/ubuntu/anaconda3/lib/python3.7/site-packages/tensorflow/python/eager/function.py", line 2829, in call
return graph_function._filtered_call(args, kwargs) # pylint: disable=protected-access
File "/home/ubuntu/anaconda3/lib/python3.7/site-packages/tensorflow/python/eager/function.py", line 1848, in _filtered_call
cancellation_manager=cancellation_manager)
File "/home/ubuntu/anaconda3/lib/python3.7/site-packages/tensorflow/python/eager/function.py", line 1924, in _call_flat
ctx, args, cancellation_manager=cancellation_manager))
File "/home/ubuntu/anaconda3/lib/python3.7/site-packages/tensorflow/python/eager/function.py", line 550, in call
ctx=ctx)
File "/home/ubuntu/anaconda3/lib/python3.7/site-packages/tensorflow/python/eager/execute.py", line 60, in quick_execute
inputs, attrs, num_outputs)
tensorflow.python.framework.errors_impl.UnknownError: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above.
[[node functional_1/conv3d/Conv3D (defined at /home/ubuntu/anaconda3/lib/python3.7/site-packages/miscnn/neural_network/model.py:201) ]] [Op:__inference_train_function_9673]
Function call stack:
train_function
2020-12-07 06:04:13.638644: W tensorflow/core/kernels/data/generator_dataset_op.cc:103] Error occurred when finalizing GeneratorDataset iterator: Failed precondition: Python interpreter state is not initialized. The process may be terminated
Executing scripts/download_data.py
gives following error
$ python3 scripts/download_data.py
INFO: Downloading Volumes
COVID-19-CT-Seg_20cases.zip?download=1: 0%| | 1.20k/1.11G [00:00<225:54:25, 1.36kB/s]
INFO: Downloading Segmentations
Lung_and_Infection_Mask.zip?download=1: 0%| | 1.20k/11.7M [00:00<2:24:22, 1.35kB/s]
INFO: Obtain sample list from the volumes ZIP file
Traceback (most recent call last):
File "scripts/download_data.py", line 85, in <module>
with zipfile.ZipFile(path_vol_zip, "r") as zip_vol:
File "/home/cinnamon/Repositories/miniconda3/envs/covid/lib/python3.6/zipfile.py", line 1131, in __init__
self._RealGetContents()
File "/home/cinnamon/Repositories/miniconda3/envs/covid/lib/python3.6/zipfile.py", line 1198, in _RealGetContents
raise BadZipFile("File is not a zip file")
zipfile.BadZipFile: File is not a zip file
How to overcome this issue
Hello, I'm sorry to disturb you, but I encountered some problems while running the program you wrote. I hope you can answer.
In the process of running run_miscnn.py, it always shows that there is no imaging.nii.gz file. I have copied it from other sources. If it exists, it still reports this error. So I want to ask, did you pass other files? We have processed the data and look forward to your reply. Greatful.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ 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. ๐๐๐
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 โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.