Computed tomography super-resolution using convolutional neural networks | IEEE Conference Publication | IEEE Xplore

Computed tomography super-resolution using convolutional neural networks


Abstract:

The practical application of Computed Tomography (CT) faces the dilemma between higher image resolution and less X-ray exposure for patients, motivating the research on C...Show More

Abstract:

The practical application of Computed Tomography (CT) faces the dilemma between higher image resolution and less X-ray exposure for patients, motivating the research on CT super-resolution (SR). In this paper, we apply state-of-the-art SR techniques to reconstruct CT images using two proposed advanced CT SR models based on Convolutional Neural Networks (CNNs) and residual learning: a single-slice CT SR network (S-CTSRN), and a multi-slice CT SR network (M-CTSRN). S-CTSRN improves the high-frequency feature extraction by incorporating the residual learning strategy, while M-CTSRN further utilizes the coherence between neighboring CT slices for better SR reconstruction. We evaluate both models on a large-scale CT dataset1, and obtain competitive results both quantitatively and qualitatively.
Date of Conference: 17-20 September 2017
Date Added to IEEE Xplore: 22 February 2018
ISBN Information:
Electronic ISSN: 2381-8549
Conference Location: Beijing, China

Contact IEEE to Subscribe

References

References is not available for this document.