Detecting Bacterial and Viral Pneumonia in Chest X-Ray Images Using Deep Learning
Abstract (300 words maximum)
This study uses a Chest X-ray dataset to apply deep learning models for pneumonia detection. The study assesses four well-known deep learning networks on the provided dataset: AlexNet, ResNet18, GoogleNet, and VGG16. We also investigate the utility of transfer learning for these models. Our results show that transfer learning significantly improves model accuracy, with ResNet18 achieving the highest test accuracy of 87%. We also consider classifying the X-ray pictures using Generative Adversarial Networks (GANs), which perform better than traditional classification algorithms. According to our research, deep learning and GANs have the ability to identify pneumonia rapidly and accurately, which could lead to more successful treatment and diagnosis.
Academic department under which the project should be listed
CCSE - Computer Science
Primary Investigator (PI) Name
Md Abdullah Al Hafiz Khan
Detecting Bacterial and Viral Pneumonia in Chest X-Ray Images Using Deep Learning
This study uses a Chest X-ray dataset to apply deep learning models for pneumonia detection. The study assesses four well-known deep learning networks on the provided dataset: AlexNet, ResNet18, GoogleNet, and VGG16. We also investigate the utility of transfer learning for these models. Our results show that transfer learning significantly improves model accuracy, with ResNet18 achieving the highest test accuracy of 87%. We also consider classifying the X-ray pictures using Generative Adversarial Networks (GANs), which perform better than traditional classification algorithms. According to our research, deep learning and GANs have the ability to identify pneumonia rapidly and accurately, which could lead to more successful treatment and diagnosis.