Investigate and implement state-of-the-art deep learning-based image enhancement techniques for improving the quality of chest X-ray (CXR) images.

DOI: https://doie.org/10.10399/JBSE.2025909936

Santosh Mugali, Dr.S.Kotresh

Keywords:

Chest X-ray enhancement, Deep learning, Drago tone mapping, supervised contrastive learning.

Abstract:

Chest radiography or chest X-ray (CXR), is one of the most widely used imaging modalities for diagnosing a variety of pulmonary and cardiovascular diseases. Despite of its critical role in clinical decision-making, CXR imaging faces several challenges such as poor image quality, inherent biases in datasets.  It requires automated and efficient diagnostic models. Traditional image interpretation relies on expert radiologists, making the process time-consuming, subjective and prone to inter-personnel variability. To overcome these challenges, this research work helps to revolutionize the chest radiography by integrating state-of-the-art deep learning techniques to enhance image quality. Also, to develop an advanced diagnostic framework and mitigate biases to improve fairness in automated CXR interpretation

PDF