Fine-Tuning the Deep Learning Models Using Transfer Learning for the Classification of Lung Diseases from Chest Radiographs

Lung diseases are one of the main sources of death across the globe which might prompt lung cancer when left unattended for an extensive stretch of time. X-ray imaging is the fundamental stage in clinical imaging for patients associated with lung oddities. However, because of the intricate morphology of the chest, radiologists have a difficult time visually interpreting the chest radiographs. The purpose of this study is to develop a medical image interpretation model for diagnosing multiple lung diseases by identifying abnormalities in chest X-ray images using transfer learning. The suggested approach has experimented with the four classes of the COVID-19 radiography dataset. The MobileNet V2 architecture performed effectively with the preprocessed dataset.