Skip to main content
Fig. 1 | The Egyptian Heart Journal

Fig. 1

From: Use of a deep-learning-based lumen extraction method to detect significant stenosis on coronary computed tomography angiography in patients with severe coronary calcification

Fig. 1

The architecture of 2D mapping functions that translate CCTA and non-contrast CT images. The network is designed to output the weighted average of single convolutions of 3 × 3 (“Single convolutional part”) by pixel-wise classification results by the U-net (“Classification part”). The pixel-wise classification results are softmax-activated so that all classification scores add up to 1. Conv convolution, ReL U rectified linear unit

Back to article page