Four Deep Learning Models for COVID-19 X-ray Classification
In the context of the ongoing COVID-19 pandemic, accurate and timely diagnosis is critical to controlling the spread of the virus. X-ray images have proven to be an important tool for detecting COVID-19-induced pneumonia. In this project, I applied several deep learning models to the COVID19-Xray-Dataset to predict COVID-19 cases from X-ray images. The models employed include Convolutional Neural Networks (CNN), VGG16, VGG19, and MobileNetV3.
The dataset consists of labeled X-ray images categorized into COVID-19 positive and negative cases. The images underwent pre-processing to ensure uniform size and normalization, suitable for input into deep learning models.
The study used the following models for training and evaluation:
-
Convolutional Neural Network (CNN): A simple CNN architecture was designed with multiple convolutional and pooling layers followed by fully connected layers for classification.
-
VGG16: A deep convolutional network with 16 layers, VGG16 is known for its high performance on image classification tasks.
-
VGG19: Similar to VGG16 but with a deeper architecture consisting of 19 layers, VGG19 often improves performance by capturing more complex patterns.
-
MobileNetV3: MobileNetV3 is a lightweight, efficient architecture designed for mobile and edge devices, making it well-suited for environments with limited computational resources. It strikes a balance between model size and accuracy.
Each model will be trained on the same pre-processed dataset, with hyperparameters such as learning rate and batch size optimized for each model. The dataset was split into training, validation, and test sets for robust evaluation.
