Chest Xray image analysis using Deep Transfer Learning technique with Pytorch.
The maxpool-5 Layer pretrained VGGNet-16 CNN model via Finetuning the convnet with SGD optimizer and Batch Normalization for the classification of Normal vs Opacity vs Cardiomegaly
Fine Tuning : Instead of random initializaion, we initialize the network with a pretrained network, like the one that is trained on imagenet 1000 dataset. Rest of the training looks as usual.
Feature Extractor(Option Field Created): where we will freeze the weights for all of the network except that of the final fully connected layer,which can be done setting up feature_extract=True in set_require_grad function .
openi.nlm.nih.gov has a large base of Xray,MRI, CT scan images publically available.Specifically Chest Xray Images have been scraped, Normal and Nodule labbeled images are futher extrated for this task.
1.Go to import data folder – Download Data using this command in your CLI.
python scraper.py <path/to/folder/to/save/images>
This downloads the images and saves corresponding disease label in json format.
2.Now , the following Importdata/Making dataset-For three labels.ipynb notebook for Data processing and generate
-
All images for training will be stored in Train Folder
-
All images for testing will be stored in Test Folder with 3 cateogries of each .
3.Finally , Run the Vision_Transfer_Learning.ipynb
Accuracy : 71.4221%