Skip to content

Commit

Permalink
Merge pull request #340 from sreevidya-16/main
Browse files Browse the repository at this point in the history
Pneumonia Detection using X-Ray Images
  • Loading branch information
TAHIR0110 authored Aug 8, 2024
2 parents 2b2c641 + 78fb476 commit cc2d6ac
Show file tree
Hide file tree
Showing 11 changed files with 2,437 additions and 0 deletions.
2 changes: 2 additions & 0 deletions Pneumonia Detection using X-Ray Images/Dataset/Readme.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,2 @@
# Pneumonia Detection using X-Ray images
## Detaset link: https://www.kaggle.com/datasets/paultimothymooney/chest-xray-pneumonia
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.

Large diffs are not rendered by default.

1,136 changes: 1,136 additions & 0 deletions Pneumonia Detection using X-Ray Images/Models/Pneumonia Detection via CNN.ipynb

Large diffs are not rendered by default.

91 changes: 91 additions & 0 deletions Pneumonia Detection using X-Ray Images/Readme.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,91 @@
# <p align = "center">Pneumonia Detection using X-Ray images </p>
## <p align = "center">Aim of the project: To detect the pneumonica of the patient using the X-Ray images dataset.</p>
## Libraries and Frameworks used:
#### 1. Pandas
#### 2. Numpy
#### 3. Matplotlib
#### 4. Tensorflow
#### 5. Keras
#### 6. OpenCV
#### 7. Imutils
#### 8. Pathlib

## Deep Learning Algotithms used:
#### 1. EfficientNet50
#### 2. MobileNetV2
#### 3. InceptionV3
#### 4. Normal Convolutional Neural Network

## Evaluation Metrics used:
#### 1. Mean Absolute Error

## Loss Functions Used:
#### 1. Softmax
#### 2. Rectified Linear Unit

## <p align = "center">Images</p>



### <p align = "center">Visualization Phase using Matplotlib</p>

![Grayscale visuals](https://github.com/PiyushBL45t/DL-Simplified/assets/75735209/72b48b73-0e39-45a1-a42a-573ba954f99e)


![Matshow Image](https://github.com/PiyushBL45t/DL-Simplified/assets/75735209/5b8751c5-3584-4d60-b181-875735236d97)





## Loss and Accuracy Statistics
<p align = "center">Loss and Accuracy Graphs for EfficientNEt50 Algorithm</p>

#### Loss Graph
![Loss Graph EfficientNet50](https://github.com/PiyushBL45t/DL-Simplified/assets/75735209/69e94b4e-5b48-4a3d-a340-d92d72441534)

#### Accuracy Graph
![Accuracy Graph Efficientnet50](https://github.com/PiyushBL45t/DL-Simplified/assets/75735209/8a86826b-5dff-4e84-b679-092ecb122f70)


<p align = "center">Loss and Accuracy Graphs for CNN Algorithm</p>

#### Loss Graph

![Loss Graph CNN](https://github.com/PiyushBL45t/DL-Simplified/assets/75735209/23a28c42-5abe-4e29-9837-7e487f07b8ff)


#### Accuracy Graph

![Accuracy Graph CNN](https://github.com/PiyushBL45t/DL-Simplified/assets/75735209/2c3d878e-bb68-4e7c-875d-f426ecf6d531)


## <p align = "center">Results</p>

<div align = "center">

![CNN Result](https://github.com/PiyushBL45t/DL-Simplified/assets/75735209/4b2462bf-760a-48bb-960f-90dce1525307)

</div>

## Accuracy and training time comparison of all the Deep Learning Algorithms
| | Normal CNN | EfficientNet50 | InceptionV3 | MobileNetV2 |
|-------------|------------|----------------|-------------|-------------|
|Accuracy | 97% | 77% | 86% | 93% |
|Eval Metrics | None | MAE | MAE | None |
|Train Time | 15 mins | 20 mins | 20 mins | 20 mins |
|Epochs | 10 | 10 | 10 | 10 |


## Techniques used for data preprocssing:
1. Image Prepreocssing
2. Data Preprocessing
3. Image visualizations
4. Setting up the path for defective and non defective images
5. Data Labellling and annotations
6. Building the CNN model
7. Using the pretrained models and fine tuning them
8. Converting the image data to numpy arrays
9. Checking the loss and accuracy graphs for each DL algorithm used


0 comments on commit cc2d6ac

Please sign in to comment.