Hey. I am currently trying to make a 3d printable model of the human bronchial airway using CT data. If anyone can tell me the process for it?
Thanks in advance.
You may be able to segment airways using Threshold effect then cut off irrelevant parts using Scissors effect.
If a single global threshold cannot capture everything then you can put seeds inside airways and outside and use “Grow from seeds” effect to create a complete segmentation. The advantage of this approach is that you can add further seeds until you get a segmentation with sufficient details.
If you have trouble obtaining good segmentation then post screenshots highlighting the issues.
1 Like
Also, from where we can get detailed CT dicom data of lungs online? I need airways atleast till 7th generation (G7). Please help
TCIA contains lots of lung images (with ground truth segmentations available in Medical Decathlon). However, it is always the best to ask data from your clinical collaborators - they can also validate quality of your processed data and results, keep your focus on the right clinical priorities, etc.
We have sample CT chest data available in 3d slicer but it is not in DICOM format. I downloaded an extension named ‘Airway Segmentation CLI’ for segment airways using non-dicom data. I have some doubts on how to use that module. Please help
I recommend using the other airway segmentation module, not the CLI (it is easier), it will create a model from the label map as well
When I am doing with that using sample CT chest data available in 3-D slicer, this is what it is showing with other airway segmentation extension-
It is a bug when using non dicom images as they don’t have certain tags, you can use this fix to bypass it and make it work , simply replace the file located in Slicer_install_dir\lib\Slicer-4.10\qt-scripted-modules\AirwaySegmentation.py with that one after backing up the original.
‘AirwaySegmentation.py’ is not showing in the folder you mentioned. I have downloaded the extension in the 3-D slicer.
Sir have you created your lung airways model?
You may want to try to use the “airway segmentation” in Lung CT Analyzer → Lung CT Segmenter module.
Sir this module is not available in 3D slicer 5.3.0 version.
Thanks & Regards:
Sameer Verma,
Ph.D. (Research Scholar),
National Institute of Technology, Rourkela
Odisha (India)-769008.
Phone: +917042554240
I can not test it now, but you could switch to 5.4.0 where it is available.
Hello everyone,
I’d like to suggest a workflow that has proven effective for me:
- Use the Lung CT segmenter to quickly and precisely isolate the airway.
- Remove all other segmentations such as PA, PV, Thoracic cavity, left lung, right lung, etc.
- Create a new segmentation, which I refer to as “Tracheal Wall”, and duplicate the airway using logical operators.
- Apply the margin effect to expand your Tracheal Wall by 0.5 cm.
- Subtract the original airway segmentation from the Tracheal Wall using logical operations. This will result in a hollowed airway.
- Export a STL file using the Segmentation module
While you can opt to use the hollow effect directly, I’ve found that it doesn’t produce a satisfactory “air space” in the more distal bronchus.
I`ve been using it for some airway surgery planing
And also some realistic printed models for segmentectomy robotic surgery training
It works well for me… I hope this is helpful
3 Likes
Sir I have downloaded the slicer 5.4.0 version but could not able to generate lung airways model upto 6th generation. Can you please help?
Thanks & Regards:
Sameer Verma,
Ph.D. (Research Scholar),
National Institute of Technology, Rourkela
Odisha (India)-769008.
Phone: +917042554240
The details of airway segmentation very much depend on the starting material you are working with. Please use thin sliced CT (1 mm) for good results.
Sir can you please provide some open source links to download CT scan dicom file for lungs airways (thin sliced).
Thanks & Regards:
Sameer Verma,
Ph.D. (Research Scholar),
National Institute of Technology, Rourkela
Odisha (India)-769008.
Phone: +917042554240
We could probably share the lung023 volume from the Monailabel training dataset if that is ok @diazandr3s is there an official link to that data?
In the meantime you can use:
https://drive.google.com/drive/folders/1h_UQLL71yrXwqkUtNZ4a6NtvEobAuPyR?usp=sharing
1 Like
As @lassoan already posted here, either TCIA or Medical Segmentation Decathlon (MSD) datasets are available to download.
MSD datasets are not in DICOM format though.
For lung airway segmentation, you could use MONAI Label to create your AI model.
Thank you so much sir for providing the google drive link it worked for me in generating airways model.
Can you please tell how I can import this 3D slicer file into the Ansys fluent software for the computational simulation purpose.