Awesome Open Source
Awesome Open Source

Pneumonia Detection from Chest X-Ray Images using Transfer Learning

Domain             : Computer Vision, Machine Learning
Sub-Domain         : Deep Learning, Image Recognition
Techniques         : Deep Convolutional Neural Network, ImageNet, Inception
Application        : Image Recognition, Image Classification, Medical Imaging


1. Detected Pneumonia from Chest X-Ray images using Custom Deep Convololutional Neural Network and by retraining pretrained model “InceptionV3” with 5856 images of X-ray (1.15GB).
2. For retraining removed output layers, freezed first few layers and fine-tuned model for two new label classes (Pneumonia and Normal).
3. With Custom Deep Convololutional Neural Network attained testing accuracy 89.53% and loss 0.41.


GitHub Link      : Detection of Pneumonia from Chest X-Ray Images(GitHub)
GitLab Link      : Detection of Pneumonia from Chest X-Ray Images(GitLab)
Portfolio        : Anjana Tiha's Portfolio


Dataset Name     : Chest X-Ray Images (Pneumonia)
Dataset Link     : Chest X-Ray Images (Pneumonia) Dataset (Kaggle)
                 : Chest X-Ray Images (Pneumonia) Dataset (Original Dataset)
Original Paper   : Identifying Medical Diagnoses and Treatable Diseases by Image-Based Deep Learning
                   (Daniel S. Kermany, Michael Goldbaum, Wenjia Cai, M. Anthony Lewis, Huimin Xia, Kang Zhang)
Dataset Details
Dataset Name            : Chest X-Ray Images (Pneumonia)
Number of Class         : 2
Number/Size of Images   : Total      : 5856 (1.15 Gigabyte (GB))
                          Training   : 5216 (1.07 Gigabyte (GB))
                          Validation : 320  (42.8 Megabyte (MB))
                          Testing    : 320  (35.4 Megabyte (MB))

Model Parameters
Machine Learning Library: Keras
Base Model              : InceptionV3 && Custom Deep Convolutional Neural Network
Optimizers              : Adam
Loss Function           : categorical_crossentropy

For Custom Deep Convolutional Neural Network : 
Training Parameters
Batch Size              : 64
Number of Epochs        : 30
Training Time           : 2 Hours

Output (Prediction/ Recognition / Classification Metrics)
Accuracy (F-1) Score    : 89.53%
Loss                    : 0.41
Precision               : 88.37%
Recall (Pneumonia)      : 95.48% (For positive class)

Sample Output:
See More Images
Confusion Matrix:
Confusion Matrix

Tools / Libraries

Languages               : Python
Tools/IDE               : Anaconda
Libraries               : Keras, TensorFlow, Inception, ImageNet


Duration                : October 2018 - Current
Current Version         : v1.0.0.3
Last Update             : 12.16.2018

Get A Weekly Email With Trending Projects For These Topics
No Spam. Unsubscribe easily at any time.
jupyter-notebook (6,156
tensorflow (2,138
keras (761
convolutional-neural-networks (450
imagenet (109
medical-imaging (95
keras-tensorflow (70
image-recognition (69
inception (28
medical-image-processing (27
inceptionv3 (21
medical-image-analysis (21