This is a console app from python to do tooth segmentation The only need to run is tooth_segmenter.py It has 2 arguments:

1. --path-file : string => path to STL / VTP file
2. --type : 'max'|'min' => Type for maxillary or mandibular

The output is VTP file which is the segmented input file in saved folder Take the out_refined.vtp file

example run: "C:/Program Files/Python39/python.exe" "d:/NyeMan/KULIAH S2/Thesis/MeshSegNet-master/MeshSegNet-master/for-app/tooth_segmenter.py" --path-file "D:\NyeMan\KULIAH S2\Thesis\3Shape new-20220223T024758Z-001\3Shape new\1.MNF Perempuan -18thn\MNF UpperJawScan.stl" --type max

dependencies:

• pytorch
• vedo
• scipy
• sklearn
• pygco
• numpy
• scikit-learn
• vedo
• pygco
• pandas
• Cython

How to setup and run:

1. if there is no venv folder, run py -3 -m venv venv
2. activate the venv with command venv\Scripts\activate
3. after the venv active, install flask witch command pip install flask
4. install Microsoft Visual for C++ 14.0 or higher
5. install dependencies: pip3 install numpy Cython scipy sklearn visdom scikit-learn pandas vedo pygco torch==1.8.2+cu102 if your PC does not have CUDA (GPU), remove +cu102
6. run below command: set FLASK_ENV=development (optional) set FLASK_APP=app.py flask run
7. if there is still error, run pip install torchvision==0.9.2+cu102 torchaudio===0.8.2 -f https://download.pytorch.org/whl/lts/1.8/torch_lts.html if your PC does not have CUDA (GPU), remove +cu102
8. do step number 6 again.

API:

• segment///<path-to-file (absolute)> Call the api like this: http://127.0.0.1:5000/segment/max/d/D:\NyeMan\KULIAH S2\Thesis\3Shape new-20220223T024758Z-001\3Shape new\1.MNF Perempuan -18thn\MNF UpperJawScan.stl