Detection of Pneumonia, Covid-19 and Tuberculosis Using Deep Learning
Giddaluru Vennela, Bandi Sivasai, Koyyana Anuhya, Mamatha Samsom
Abstract
Chest radiographs, often known as X-rays of the chest, are regarded to be useful in observing and analysing a range of respiratory illnesses, including as pneumonia, tuberculosis (TB), and Covid-19. Due to recent advancements in computer-assisted diagnostic technologies, many disorders can now be classified with greater accuracy. Without losing feature extraction or classification accuracy, the suggested deep learning (DL) classification method is utilized to forecast four different categories. Convolutional neural network practise datasets of chest X-ray (CXR) images are used to test and verify the ensemble model. Based on methods like Inception V3, Mobile Net, Dense Net, and Mobile Net +Dense Net, this study develops a contemporary deep learning channel to identify patients with COVID-19, pneumonia, TB, and healthy conditions using CXR pictures. This work could help radiologists to make clinical decisions accurately
MeSH terms
- Pneumonia
- Deep learning
- Convolutional neural network
- Coronavirus disease 2019 (COVID-19)
- Artificial intelligence
- Computer science
- Tuberculosis
- Feature extraction
- Radiography
- Medicine
- Machine learning