Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/priyavrat-misra/xrays-and-gradcam
Classification and Gradient-based Localization of Chest Radiographs using PyTorch.
https://github.com/priyavrat-misra/xrays-and-gradcam
cnn covid-19 deep-learning densenet121 early-stopping fine-tuning gradcam imbalanced-data localization oversampling pneumonia pytorch-implementation radiographs resnet18 transfer-learning vgg16 xrays
Last synced: about 1 month ago
JSON representation
Classification and Gradient-based Localization of Chest Radiographs using PyTorch.
- Host: GitHub
- URL: https://github.com/priyavrat-misra/xrays-and-gradcam
- Owner: priyavrat-misra
- License: mit
- Created: 2021-04-15T14:40:54.000Z (over 3 years ago)
- Default Branch: master
- Last Pushed: 2021-05-15T11:59:31.000Z (over 3 years ago)
- Last Synced: 2024-01-25T07:08:52.673Z (8 months ago)
- Topics: cnn, covid-19, deep-learning, densenet121, early-stopping, fine-tuning, gradcam, imbalanced-data, localization, oversampling, pneumonia, pytorch-implementation, radiographs, resnet18, transfer-learning, vgg16, xrays
- Language: Jupyter Notebook
- Homepage:
- Size: 3.89 MB
- Stars: 44
- Watchers: 3
- Forks: 16
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- fucking-awesome-readme - priyavrat-misra/xrays-and-gradcam - TOC for easy navigation. Clear introduction of what the project does. Complete list of all files in the repo and what their function is. Usage section. Clean and concise tables. (Examples)
- awesome-readme - priyavrat-misra/xrays-and-gradcam - TOC for easy navigation. Clear introduction of what the project does. Complete list of all files in the repo and what their function is. Usage section. Clean and concise tables. (Examples)
- awesome-readme - priyavrat-misra/xrays-and-gradcam - TOC for easy navigation. Clear introduction of what the project does. Complete list of all files in the repo and what their function is. Usage section. Clean and concise tables. (Examples)
README
# Classification and Gradient-based Localization of Chest Radiographs
## Contents
- [Introduction](#introduction)
- [Overview](#overview)
- [Steps](#steps)
- [Results](#results)
- [Usage](#usage)
- [Conclusions](#conclusions)
- [References](#references)
## Introduction
> A team of radiologists from New Orleans studied the usefulness of Chest Radiographs for diagnosing COVID-19 compared to the reverse-transcription polymerase chain reaction (RT-PCR) and found out they could aid rapid diagnosis, especially in areas with limited testing facilities [[1]](https://pubs.rsna.org/doi/10.1148/ryct.2020200280 "A Characteristic Chest Radiographic Pattern in the Setting of the COVID-19 Pandemic").
> Another study found out that the radiographs of different viral cases of pneumonia are comparative, and they overlap with other infectious and inflammatory lung diseases, making it hard for radiologists to recognize COVID‐19 from other viral pneumonia cases [[2]](https://pubs.rsna.org/doi/10.1148/rg.2018170048 "Radiographic and CT Features of Viral Pneumonia").
> This project aims to make the former study a reality while dealing with the intricacies in the latter, with the help of Deep Learning.
## Overview
> The project uses the COVID-19 Radiography Database [[3]](https://www.kaggle.com/tawsifurrahman/covid19-radiography-database) as it's dataset.
> It has a total of `21165` Chest X-Rays (CXRs) belonging to 4 different classes (`COVID-19`, `Lung Opacity`, `Normal` and `Viral Pneumonia`).
> Three top scoring CNN architectures, __VGG-16__ [[4]](https://arxiv.org/abs/1409.1556 "Very Deep Convolutional Networks for Large-Scale Image Recognition"), __ResNet-18__ [[5]](https://arxiv.org/abs/1409.1556 "Deep Residual Learning for Image Recognition
") and __DenseNet-121__ [[6]](https://arxiv.org/abs/1608.06993 "Densely Connected Convolutional Networks"), trained on the ImageNet Dataset [[7]](http://image-net.org/), were chosen for __fine-tuning__ on the dataset.
> The results obtained from the different architectures were then evaluted and compared.
> Finally, with the help of __Gradient weighted Class Activation Maps__ (Grad-CAM) [[8]](https://arxiv.org/abs/1610.02391 "Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization") the affected areas in CXRs were localized.
* ___Note:___ The dataset and the trained models can be found in [here](https://drive.google.com/drive/folders/14L8wd-d2a3lvgqQtwV-y53Gsnn6Ud2-w?usp=sharing).
## Steps
> 1. [Dataset Exploration](./1_data_exploration.ipynb "1_data_exploration.ipynb")
> 2. [Split the dataset](./split_dataset.py "split_dataset.py")
> |Type|COVID-19|Lung Opacity|Normal|Viral Pneumonia|Total|
> |:-|-:|-:|-:|-:|-:|
> |Train|3496|5892|10072|1225|20685|
> |Val|60|60|60|60|240|
> |Test|60|60|60|60|240|
> 3. [Fine-tune VGG-16, ResNet-18 and DenseNet-121](./2_finetune_models.ipynb "2_finetune_models.ipynb")
> 1. [Define Transformations](./utils.py#L15-L33)
> 2. [Handle imbalanced dataset with Weighted Random Sampling (Over-sampling)](./2_finetune_models.ipynb "2_finetune_models.ipynb/")
> 3. [Prepare the Pre-trained models](./networks.py "networks.py")
> 4. [Fine-tune step with Early-stopping](./utils.py#L83-L151)
> - |Hyper-parameters||
> |:-|-:|
> |Learning rate|`0.00003`|
> |Batch Size|`32`|
> |Number of Epochs|`25`|
> - |Loss Function|Optimizer|
> |:-:|:-:|
> |`Categorical Cross Entropy`|`Adam`|
> 5. [Plot running losses & accuracies](./plot_utils.py#L8-L42)
> - |Model|Summary Plot|
> |:-:|:-:|
> |VGG-16||
> |ResNet-18||
> |DenseNet-121||
> 4. [Results Evaluation](./3_evaluate_results.ipynb "3_evaluate_results.ipynb")
> 1. [Plot confusion matrices](./plot_utils.py#L45-L69)
> 2. [Compute test-set Accuracy, Precision, Recall & F1-score](./utils.py#L64-L80)
> 3. [Localize using Grad-CAM](./grad_cam.py)
> 5. [Inference](./overlay_cam.py)
## Results
VGG-16
ResNet-18
DenseNet-121
|__Pathology__|
|:-|
|COVID-19|
|Lung Opacity|
|Normal|
|Viral Pneumonia|
|Accuracy|Precision|Recall|F1-Score|
|-:|-:|-:|-:|
|0.9956|0.9833|1.0000|0.9916|
|0.9582|0.8833|0.9464|0.9138|
|0.9622|0.9667|0.8923|0.9280|
|0.9913|0.9833|0.9833|0.9833|
|Accuracy|Precision|Recall|F1-Score|
|-:|-:|-:|-:|
|0.9871|0.9667|0.9830|0.9748|
|0.9664|0.8667|1.0000|0.9286|
|0.9664|1.0000|0.8823|0.9375|
|0.9957|1.0000|0.9836|0.9917|
|Accuracy|Precision|Recall|F1-Score|
|-:|-:|-:|-:|
|0.9957|0.9833|1.0000|0.9916|
|0.9623|0.9167|0.9322|0.9244|
|0.9623|0.9500|0.9047|0.9268|
|0.9957|0.9833|1.0000|0.9916|
|TL;DR|
|:-|
|Train set|
|Test set|
|Total Correct Predictions|Total Accuracy|
|-:|-:|
|20362|98.44%|
|229|__95.42%__|
|Total Correct Predictions|Total Accuracy|
|-:|-:|
|20639|99.78%|
|230|__95.83%__|
|Total Correct Predictions|Total Accuracy|
|-:|-:|
|20540|99.30%|
|230|__95.83%__|
Confusion Matrices
- __Localization with Gradient-based Class Activation Maps__
> |||||
> |:-:|:-:|:-:|:-:|
> |_COVID-19 infected CXR_|_VGG-16_|_ResNet-18_|_DenseNet-121_|
## Usage
> - Clone the repository
> ```bash
> git clone 'https://github.com/priyavrat-misra/xrays-and-gradcam.git' && cd xrays-and-gradcam/
> ```
> - Install dependencies
> ```bash
> pip install -r requirements.txt
> ```
> - Using `argparse` script for inference
> ```bash
> python overlay_cam.py --help
> ```
> ```
> usage: GradCAM on Chest X-Rays [-h] [-i IMAGE_PATH]
> [-l {covid_19,lung_opacity,normal,pneumonia}]
> -m {vgg16,resnet18,densenet121}
> [-o OUTPUT_PATH]
>
> Overlays given label's CAM on a given Chest X-Ray.
>
> optional arguments:
> -h, --help show this help message and exit
> -i IMAGE_PATH, --image-path IMAGE_PATH
> Path to chest X-Ray image.
> -l {covid_19,lung_opacity,normal,pneumonia}, --label {covid_19,lung_opacity,normal,pneumonia}
> Choose from covid_19, lung_opacity, normal &
> pneumonia, to get the corresponding CAM. If not
> mentioned, the highest scoring label is considered.
> -m {vgg16,resnet18,densenet121}, --model {vgg16,resnet18,densenet121}
> Choose from vgg16, resnet18 or densenet121.
> -o OUTPUT_PATH, --output-path OUTPUT_PATH
> Format: " + + .jpg"
> ```
> - An example
> ```bash
> python overlay_cam.py --image-path ./assets/original.jpg --label covid_19 --model resnet18 --output-path ./assets/dense_cam.jpg
> ```
> ```
> GradCAM generated for label "covid_19".
> GradCAM masked image saved to "./assets/res_cam.jpg".
> ```
## Conclusions
> - DenseNet-121 having only `7.98 Million` parameters did relatively better than VGG-16 and ResNet-18, with `138 Million` and `11.17 Million` parameters respectively.
> - Increase in model's parameter count doesn’t necessarily achieve better results, but increase in residual connections might.
> - Oversampling helped in dealing with imbalanced data to a great extent.
> - Fine-tuning helped substantially by dealing with the comparatively small dataset and speeding up the training process.
> - GradCAM aided in localizing the areas in CXRs that decides a model's predictions.
> - The models did a good job distinguishing various infectious and inflammatory lung diseases, which is rather hard manually, as mentioned earlier.
## References
> - [1] David L. Smith, John-Paul Grenier, Catherine Batte, and Bradley Spieler. [A Characteristic Chest Radiographic Pattern in the Setting of the COVID-19 Pandemic](https://pubs.rsna.org/doi/10.1148/ryct.2020200280). Radiology: Cardiothoracic Imaging 2020 2:5.
> - [2] Hyun Jung Koo, Soyeoun Lim, Jooae Choe, Sang-Ho Choi, Heungsup Sung, and Kyung-Hyun Do. [Radiographic and CT Features of Viral Pneumonia](https://pubs.rsna.org/doi/10.1148/rg.2018170048). RadioGraphics 2018 38:3, 719-739.
> - [3] Tawsifur Rahman, Muhammad Chowdhury, Amith Khandakar. [COVID-19 Radiography Database](https://www.kaggle.com/tawsifurrahman/covid19-radiography-database). Kaggle.
> - [4] Karen Simonyan, Andrew Zisserman. [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://arxiv.org/abs/1409.1556). arxiv:1409.1556v6.
> - [5] Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun. [Deep Residual Learning for Image Recognition](https://arxiv.org/abs/1512.03385). arxiv:1512.03385v1.
> - [6] Gao Huang, Zhuang Liu, Laurens van der Maaten, Kilian Q. Weinberger. [Densely Connected Convolutional Networks](https://arxiv.org/abs/1608.06993). arxiv:1608.06993v5.
> - [7] Deng, J. et al., 2009. [Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition](http://image-net.org/). pp. 248–255.
> - [8] Ramprasaath R. Selvaraju, Michael Cogswell, Abhishek Das, Ramakrishna Vedantam, Devi Parikh, Dhruv Batra. [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://arxiv.org/abs/1610.02391). arXiv:1610.02391v4.