Data

Tools

Indexes

Applications

Experiments

Open Source Science

casemark

landmark detection

https://github.com/ecnuacrush/casemark

Science Score: 54.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ✓
    CITATION.cff file
    Found CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ✓
    Academic publication links
    Links to: arxiv.org
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (6.6%) to scientific vocabulary
Last synced: 7 months ago · JSON representation ·

Repository

landmark detection

Basic Info
  • Host: GitHub
  • Owner: ECNUACRush
  • Language: Python
  • Default Branch: main
  • Size: 8.25 MB
Statistics
  • Stars: 2
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Created over 1 year ago · Last pushed over 1 year ago
Metadata Files
Readme Citation

README.md

CASEMark: A Novel Hybrid Architecture forAnatomical Landmark Detection in Medical Images

This is the official pytorch implementation repository of our CASEMark: A Novel Hybrid Architecture for Anatomical Landmark Detection in Medical Images

Dataset

  • We have used the following datasets:
    • ISBI2015 dataset: Ching-Wei Wang, Cheng-Ta Huang, Jia-Hong Lee, Chung-Hsing Li,Sheng-Wei Chang, Ming-Jhih Siao, Tat-Ming Lai, Bulat Ibragimov,Tomaž Vrtovec, Olaf Ronneberger, et al. A benchmark for comparison of dental radiography analysis algorithms. Medical image analysis,31:63–76, 2016
    • ISBI2023 dataset: Anwaar Khalid, M., Zulfiqar, K., Bashir, U., Shaheen, A., Iqbal, R., Rizwan, Z., Rizwan, G., Moazam Fraz, M.: Cepha29: Automatic cephalometric landmark detection challenge 2023. arXiv e-prints pp. arXiv–2212 (2022)
    • Hand X-Rays: Payer, C., ˇStern, D., Bischof, H., Urschler, M.: Integrating spatial configuration into heatmap regression based CNNs for landmark localization. Medical Image Analysis 54, 207–219 (2019)
    • Pelvic X-Rays: https://www.kaggle.com/datasets/tommyngx/cgmh-pelvisseg；Chi-Tung Cheng, Yirui Wang, Huan-Wu Chen, Po-Meng Hsiao,Chun-Nan Yeh, Chi-Hsun Hsieh, Shun Miao, Jing Xiao, Chien-Hung Liao, and Le Lu. A scalable physician-level deep learning algorithm detects universal trauma on pelvic radiographs. Nature communications, 12(1):1066, 2021.

Prerequesites

  • Python 3.9
  • MMpose 0.15

Usage of the code

  • Dataset preparation
    • The dataset structure should be in the following structure:

Inputs: .PNG images and JSON file └── <dataset name> ├── 2D_images | ├── 001.png │ ├── 002.png │ ├── 003.png │ ├── ... | └── JSON ├── train.json └── test.json - Example json format { "images": [ { "id": 0, "file_name": "0.png", "height": 512, "width": 512 }, ... ], "annotations": [ { "image_id": 0, "id": 0, "category_id": 1, "keypoints": [ 492.8121081534473, 261.62600827462876, 2, 428.0154934462112, 301.24809471563935, 2, 504.9223114040644, 234.45993184950234, 2, 470.296873380625, 182.90429052972533, 2, 456.97751121306436, 208.8105499707776, 2, 369.95414168150415, 239.07609878665616, 2, 307.83364934785106, 229.91052362204155, 2, 373.5995213621739, 353.599939601835, 2, 499.50552505239256, 453.1111418891231, 2, 493.50543334239256, 456.12341418891231, 2 ], "num_keypoints": 10, "iscrowd": 0 }, ... ]

  • Output: 2D landmark coordinates

    • Train the model

  • To train our model run sh train.sh: ```

    sh train.sh

    CUDAVISIBLEDEVICES=gpuids PORT=PORTNUM ./tools/disttrain.sh \ configfilepath numgpus ```

    • Evaluation

  • To evaluate the trained model run sh test.sh: ```

    sh test.sh

    CUDAVISIBLEDEVICES=gpuid PORT=29504 ./tools/disttest.sh configfilepath \ modelweightpath numgpus \ # For evaluation of the Head XCAT dataset, use: --eval 'MREh','MREstdh','SDR2h','SDR2.5h','SDR3h','SDR4h' # For evaluation of ISBI2023 and Hand X-ray dataset, use: # --eval 'MREi2','MREstdi2','SDR2i2','SDR2.5i2','SDR3i2','SDR4_i2' ```

Owner

  • Name: Huang Zhen
  • Login: ECNUACRush
  • Kind: user
  • Location: Shanghai
  • Company: East China Normal University

Always to be better.

Citation (CITATION.cff) 

message: "(Citaion will be upated) If you use this code, please cite it as below."
authors:
  - name: "Thanaporn Viriyasaranon, Serie Ma, and Jang-Hwan Choi"
title: "Anatomical Landmark Detection Using a Multiresolution Learning Approach with a Hybrid Transformer-CNN Model"
date-released: 2020-05-30
url: "https://github.com/seriee/Multiresolution-Learning-based-Hybrid-Transformer-CNN-Model-for-Anatomical-Landmark-Detection"

GitHub Events

Total
  • Watch event: 2
  • Push event: 1
  • Create event: 2
Last Year
  • Watch event: 2
  • Push event: 1
  • Create event: 2

Dependencies

requirements/build.txt pypi
  • numpy *
  • torch >=1.3
requirements/docs.txt pypi
  • docutils ==0.16.0
  • myst-parser *
  • sphinx ==4.0.2
  • sphinx_copybutton *
  • sphinx_markdown_tables *
requirements/mminstall.txt pypi
  • mmcv-full >=1.3.8
  • mmdet >=2.14.0
  • mmtrack >=0.6.0
requirements/optional.txt pypi
  • albumentations >=0.3.2
  • onnx *
  • onnxruntime *
  • pyrender *
  • requests *
  • smplx >=0.1.28
  • trimesh *
requirements/readthedocs.txt pypi
  • mmcv-full *
  • munkres *
  • regex *
  • scipy *
  • titlecase *
  • torch *
  • torchvision *
  • xtcocotools >=1.8
requirements/runtime.txt pypi
  • chumpy *
  • dataclasses *
  • json_tricks *
  • matplotlib *
  • munkres *
  • numpy *
  • opencv-python *
  • pillow *
  • scipy *
  • torchvision *
  • xtcocotools >=1.8
requirements/tests.txt pypi
  • coverage * test
  • flake8 * test
  • interrogate * test
  • isort ==4.3.21 test
  • pytest * test
  • pytest-runner * test
  • smplx >=0.1.28 test
  • xdoctest >=0.10.0 test
  • yapf * test
requirements.txt pypi
setup.py pypi