Calcium Removal From Cardiac CT Images Using Deep Convolutional Neural Network
This addresses a specific medical imaging problem for cardiac CT diagnosis, though it appears incremental as it builds on existing deep learning methods for image processing.
The researchers tackled the problem of coronary calcium causing artifacts in cardiac CT images that reduce diagnostic accuracy by developing a deep convolutional neural network called Dense-Unet to remove calcification and restore arterial lumen. Their results showed this approach was feasible and validated against gold-standard X-ray angiography, potentially improving diagnostic accuracy.
Coronary calcium causes beam hardening and blooming artifacts on cardiac computed tomography angiography (CTA) images, which lead to overestimation of lumen stenosis and reduction of diagnostic specificity. To properly remove coronary calcification and restore arterial lumen precisely, we propose a machine learning-based method with a multi-step inpainting process. We developed a new network configuration, Dense-Unet, to achieve optimal performance with low computational cost. Results after the calcium removal process were validated by comparing with gold-standard X-ray angiography. Our results demonstrated that removing coronary calcification from images with the proposed approach was feasible, and may potentially improve the diagnostic accuracy of CTA.