eduzhai > Applied Sciences > Engineering >

Unsupervised MRI Super-Resolution Using Deep External Learning and Guided Residual Dense Network with Multimodal Image Priors

  • king
  • (0) Download
  • 20210507
  • Save

... pages left unread,continue reading

Document pages: 10 pages

Abstract: Deep learning techniques have led to state-of-the-art single imagesuper-resolution (SISR) with natural images. Pairs of high-resolution (HR) andlow-resolution (LR) images are used to train the deep learning model (mappingfunction). These techniques have also been applied to medical imagesuper-resolution (SR). Compared with natural images, medical images haveseveral unique characteristics. First, there are no HR images for training inreal clinical applications because of the limitations of imaging systems andclinical requirements. Second, other modal HR images are available (e.g., HRT1-weighted images are available for enhancing LR T2-weighted images). In thispaper, we propose an unsupervised SISR technique based on simple priorknowledge of the human anatomy; this technique does not require HR images fortraining. Furthermore, we present a guided residual dense network, whichincorporates a residual dense network with a guided deep convolutional neuralnetwork for enhancing the resolution of LR images by referring to different HRimages of the same subject. Experiments on a publicly available brain MRIdatabase showed that our proposed method achieves better performance than thestate-of-the-art methods.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×