Login / Signup

The Time-Domain Integration Method of Digital Subtraction Angiography Images.

Shuo HuangLe ChengBin ZhuPing ZhouYu SunBing ZhangSuiren Wan
Published in: Computational and mathematical methods in medicine (2018)
The clarity improvement and the noise suppression of digital subtraction angiography (DSA) images are very important. However, the common methods are very complicated. An image time-domain integration method is proposed in this study, which is based on the blood flow periodicity. In this method, the images of the first cardiac cycle after the injection of the contrast agent are integrated to obtain the time-domain integration image. This method can be used independently or as a postprocessing method of the denoising method on the signal image. The experimental results on DSA data from an aortic dissection patient show that the image time-domain integration method is efficient in image denoising and enhancement, which also has a good real-time performance. This method can also be used to improve the denoising and image enhancement effect of some common models.
Keyphrases
  • deep learning
  • convolutional neural network
  • optical coherence tomography
  • computed tomography
  • heart failure
  • magnetic resonance imaging
  • case report
  • big data
  • ultrasound guided