Speckle reduction of OCT via super resolution reconstruction and its application on retinal layer segmentation

作者：

Highlights：

• A new super-resolution based method is proposed to despeckle OCT scans.

• Our method uses a multi-frame fusion strategy that merges scans of the same scene.

• Our method utilizes the movements of sub-pixels to recover missing signals.

• The results show the superiority of our method in terms of MSE, PSNR and MSSIM.

• Our method enables retinal layer segmentation methods to get more accurate results.

摘要

•A new super-resolution based method is proposed to despeckle OCT scans.•Our method uses a multi-frame fusion strategy that merges scans of the same scene.•Our method utilizes the movements of sub-pixels to recover missing signals.•The results show the superiority of our method in terms of MSE, PSNR and MSSIM.•Our method enables retinal layer segmentation methods to get more accurate results.

论文关键词：Speckle reduction,OCT,Retinal layer segmentation

论文评审过程：Received 1 September 2019, Revised 17 February 2020, Accepted 2 May 2020, Available online 15 May 2020, Version of Record 5 June 2020.

论文官网地址：https://doi.org/10.1016/j.artmed.2020.101871