https://github.com/hasibzunair/uniformizing-3d

[MICCAI'2020 PRIME] Uniformizing Techniques to Process CT scans with 3D CNNs for Tuberculosis Severity Estimation.
https://github.com/hasibzunair/uniformizing-3d

3d-data classification computed-tomography convolutional-neural-networks deep-learning information-retrieval volumetric-data

Last synced: 4 days ago
JSON representation

[MICCAI'2020 PRIME] Uniformizing Techniques to Process CT scans with 3D CNNs for Tuberculosis Severity Estimation.

• Host: GitHub
• URL: https://github.com/hasibzunair/uniformizing-3d
• Owner: hasibzunair
• Created: 2019-08-04T00:16:29.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2022-08-11T16:40:59.000Z (over 2 years ago)
• Last Synced: 2023-04-13T07:20:44.372Z (over 1 year ago)
• Topics: 3d-data, classification, computed-tomography, convolutional-neural-networks, deep-learning, information-retrieval, volumetric-data
• Language: Jupyter Notebook
• Homepage:
• Size: 23.6 MB
• Stars: 43
• Watchers: 4
• Forks: 5
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Intro

This is official code of MICCAI'2020 PRIME workshop paper: 


*Uniformizing Techniques to Process CT scans with 3D CNNs for Tuberculosis Prediction* ([Paper](https://link.springer.com/chapter/10.1007%2F978-3-030-59354-4_15), [arXiv](https://arxiv.org/abs/2007.13224))

### Virtual Presentation at MICCAI'2020 PRIME

[![IMAGE ALT TEXT HERE](https://img.youtube.com/vi/IP6poudyny4/0.jpg)](https://www.youtube.com/watch?v=IP6poudyny4)

### Citation

If you use this code or models in your scientific work, please cite the following paper:

```bibtex

@inproceedings{zunair2020uniformizing,

  title={Uniformizing Techniques to Process CT Scans with 3D CNNs for Tuberculosis Prediction},

  author={Zunair, Hasib and Rahman, Aimon and Mohammed, Nabeel and Cohen, Joseph Paul},

  booktitle={International Workshop on PRedictive Intelligence In MEdicine},

  pages={156--168},

  year={2020},

  organization={Springer}

}

```

### Data Processing Method

Data uniformizing methods







### 3D Convolutional Neural Network







### Results







### Dependencies

*    Ubuntu 14.04

*    Python 3.6

*    Tensorflow: 2.0.0

*    Keras: 2.3.1

### Environment setup

You can create the appropriate conda environment by running

`conda env create -f environment.yml`

### Directory Structure & Usage

First, get the data from [here](https://www.imageclef.org/2019/medical/tuberculosis). Then:

* Run notebooks in order

* `others`: Contains helper codes to preprocess and visualize samples in dataset.

### Demo

A 🤗 Spaces demo for detecting pneumonia from CT scans using our [method](https://keras.io/examples/vision/3D_image_classification/) is available [here](https://huggingface.co/spaces/keras-io/3D_CNN_Pneumonia). Demo built by [Faizan Shaikh](https://github.com/faizankshaikh).

### This is an extension of previous work

More details at this [link](https://github.com/hasibzunair/tuberculosis-severity)

```bibtex

Zunair,  H.,  Rahman,  A.,  Mohammed,  N.:   Estimating  Severity  from  CT  Scans

of  Tuberculosis  Patients  using  3D  Convolutional  Nets  and  Slice  Selection.   In:

CLEF2019  Working  Notes.  Volume  2380  of  CEUR  Workshop  Proceedings.,

Lugano, Switzerland, CEUR-WS.org

(September 9-12 2019) 

```

Previous paper published in CEUR-WS. Paper can be found at [CLEF Working Notes 2019](http://www.dei.unipd.it/~ferro/CLEF-WN-Drafts/CLEF2019/paper_77.pdf) under the section ImageCLEF - Multimedia Retrieval in CLEF.

### License

Your driver's license.