Pulmonary Angiography
Pulmonary angiography (PA), primarily using computed tomography pulmonary angiography (CTPA), is a crucial diagnostic tool for detecting pulmonary embolism (PE) and assessing pulmonary vascular anatomy. Current research heavily utilizes deep learning, employing convolutional neural networks (CNNs), transformer-based architectures (like Swin Transformers), and generative adversarial networks (GANs) to improve PE detection accuracy, automate segmentation of pulmonary vessels, and even generate synthetic CTPA scans from less expensive X-rays. These advancements aim to enhance diagnostic efficiency, reduce reliance on contrast agents, and improve risk stratification for PE patients, ultimately leading to better patient outcomes and more efficient healthcare resource allocation.