How to reach me: Connect with me on Twitter @Aakash Mohole | Linkedin @Aakash Mohole
🟢 Click on the image to see the demo video:
Overview of AE-CNN: Our proposed framework consists of three main blocks namely encoder, decoder, and classifier. The figure shows the autoencoder based convolutional neural network (AE-CNN) model for disease classification. Here, autoencoder reduces the spatial dimension of the input image of size 1024 × 1024. The encoder produces a latent code tensor of size 256 × 256 and decoder reconstructs back the image. This latent code tensor is passed through a CNN classifier for classifying the chest x-rays. The final loss is the weighted sum of the reconstruction loss by the decoder and classification loss by the CNN classifier.
Function: The encoder compresses the input data into a latent-space representation. In the context of an image, it extracts important features and reduces the dimensionality.
Architecture: Typically consists of convolutional layers followed by pooling layers. For example, in ResNet-50, the initial layers can be considered part of the encoder as they extract features from the input image.
Function: The decoder reconstructs the input data from the latent-space representation. This is used in tasks where output images are required (e.g., image generation or segmentation).
Architecture: Typically consists of upsampling layers (like transposed convolutions) that increase the dimensionality of the latent representation back to the original input size.
Function: An autoencoder is a type of neural network used to learn efficient codings of input data. It consists of two parts: the encoder and the decoder. The goal is to compress the input into a latent-space representation and then reconstruct the output as closely as possible to the original input.
Architecture: Combines both the encoder and decoder. The encoder reduces the input to a latent space, and the decoder reconstructs the input from this latent space.
ResNet-50 (Residual Network):
ResNet-50 is a deep convolutional neural network that is 50 layers deep. It is well-known for its ability to handle the vanishing gradient problem, which is common in very deep networks. This is achieved through the introduction of residual blocks.
The ResNet-50 model's convolutional layers act as an encoder, extracting high-level features from chest X-ray images.
After feature extraction, a global average pooling layer and a fully connected layer are added to classify the image as COVID-19 positive or negative.
The model is trained on a labeled dataset of chest X-ray images, with labels indicating the presence or absence of COVID-19.
The model is evaluated on a separate test set to measure its accuracy, sensitivity, specificity, and other relevant metrics.
To run the web application locally, follow these steps:
Clone this repository
git clone https://github.com/aakashmohole/COVID-19-PneumoScan-Chest-X-ray-AI.gitInstall the necessary dependencies using
pip install -r requirements.txtRun the Streamlit application using
cd appstreamlit run app.pyAccess the application in your web browser at http://localhost:5000.