Data

Tools

Indexes

Applications

Experiments

Open Source Science

cxrabnormalitylocalization

🫁 Chest X-ray abnormalities localization via ensemble of deep convolutional neural networks

https://github.com/pvti/cxrabnormalitylocalization

Science Score: 26.0%

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

  • ○
    CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ○
    Academic publication links
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (9.8%) to scientific vocabulary

Keywords

abnormality-detection chest-xray efficientnet ensemble-learning medical-image-analysis vindr-cxr yolo
Last synced: 6 months ago · JSON representation

Repository

🫁 Chest X-ray abnormalities localization via ensemble of deep convolutional neural networks

Basic Info
Statistics
  • Stars: 18
  • Watchers: 1
  • Forks: 3
  • Open Issues: 0
  • Releases: 0
Topics
abnormality-detection chest-xray efficientnet ensemble-learning medical-image-analysis vindr-cxr yolo
Created almost 5 years ago · Last pushed over 1 year ago
Metadata Files
Readme License Citation

README.md

Visitors

🫁 Chest X-ray abnormalities localization via ensemble of deep convolutional neural networks

1Modelling and Simulation Centre, Viettel High Technology Industries Corporation, Vietnam  2Kings Distributed Systems, Canada  3Eutech Systems Ltd, Bangladesh  ✉Corresponding Author
Convolutional neural networks have been applied widely in chest X-ray interpretation thanks to the availability of high-quality datasets. Among them, VinDr-CXR is one of the latest public datasets including 18000 expert-annotated images labeled into 22 local position-specific abnormalities and 6 globally suspected diseases. A proposed deep learning algorithm based on Faster-RCNN, Yolov5, and EfficientDet frameworks were developed and investigated in the task of multi-class clinical detection from chest radiography. The ground truth was defined by a radiologist-adjudicated image review. Their performance was evaluated by the mean average precision. The results show that the best performance belonged to object detection models ensembled with an EfficientNet classifier, resulting in a peak mAP of 0.292. As a trade-off, ensembling object detection models was much slower, increasing computing time by 3.75, 5, and 2.25 times compared to FasterRCNN, Yolov5, and EfficientDet individually. Overall, the classifiers show constant improvement on all detector models, which is recommended for further research. All of these aspects should be considered to address real-world CXR diagnosis where the accuracy and computing cost are of concern.

👨‍⚕️ News

📋 Main results

Evaluation of the proposed framework on the VinDr-CXR test dataset.

| Detector | | Accuracy (mAP@0.4) | | | Performance | | |:------------:|:------------:|:------------------:|----------------|:-----:|:---------------------------:|:--------------------:| | | Single model | Resnet50 | EficientNet-B7 | Speed | GPU memory requirement (MB) | Training time (hour) | | YOLOv5 | 0.21 | 0.246 | 0.269 | 15 | 3291 | 7 | | FasterRCNN | 0.248 | 0.263 | 0.278 | 20 | 2076 | 9.5 | | EfficientDet | 0.269 | 0.28 | 0.273 | 9 | 3685 | 12 | | Ensemble | 0.272 | 0.285 | 0.292 | 4 | 3685 | 30.5 |

💻 Installation

Please refer to INSTALL.md for installation instructions.

🩺 Model zoo

Trained models are available in the MODEL_ZOO.md.

🔍 Dataset zoo

Please see DATASET_ZOO.md for detailed description of the training/evaluation datasets.

💉 Getting Started

Follow the aforementioned instructions to install environments and download models and datasets.

GETTING_STARTED.md provides a brief intro of the usage of builtin command-line tools.

💊 Citing

If you use this work in your research or wish to refer to the results, please use the following BibTeX entry.

BibTeX @inproceedings{pham2021chest, title={Chest x-ray abnormalities localization via ensemble of deep convolutional neural networks}, author={Pham, Van-Tien and Tran, Cong-Minh and Zheng, Stanley and Vu, Tri-Minh and Nath, Shantanu}, booktitle={2021 International Conference on Advanced Technologies for Communications (ATC)}, pages={125--130}, year={2021}, organization={IEEE} }

Owner

  • Name: _
  • Login: pvti
  • Kind: user

GitHub Events

Total
  • Watch event: 7
Last Year
  • Watch event: 7

Dependencies

src/FasterRCNN/setup.py pypi