Identifying and Segmenting COVID-19-Infected Lung Images Using a Novel Hybrid-Integrated Gradient-Based Self-Refined Parallel Model

In the aftermath of COVID-19, patients often experience lingering pulmonary complications, which can be effectively detected and analyzed through chest X-rays. This study uses the potent artificial intelligence technique, U-Net structure, to automatically identify and segment lung anomalies caused by COVID-19 in chest X-ray pictures. This work aims to develop an efficient and interpretable AI system that can accurately detect and monitor entities. The proposed model leverages the strength of explainable AI and applies the self-refined segment editing technique, allowing the model to better discern and track infected regions in X-ray im-ages, which enhances the model performance through iterative learning. The integration of Explainable AI enhances the transparency and interpretability of the segmentation process, providing clinicians and researchers with valuable insights into the post-COVID-19 pulmonary complications. The model's performance is evaluated using evaluation measures such as the Dice coefficient and Intersection over Union (IoU).