计算机科学

计算机科学 ›› 2023, Vol. 50 ›› Issue (11A): 221200107-6.doi: 10.11896/jsjkx.221200107

• 交叉&应用 • 上一篇    下一篇

基于语义注意力的医学图像超分辨率方法

林毅, 周芃, 陈彦明   

  1. 安徽大学计算机科学与技术学院安徽省医疗成像先进技术国际联合研究中心 合肥 230601
  • 发布日期:2023-11-09
  • 通讯作者: 周芃(zhoupeng@ahu.edu.cn)
  • 作者简介:(e20201049@stu.ahu.edu.cn)
  • 基金资助:
    国家自然科学基金(62176001,61806003);安徽省高校优秀青年科研项目(2023AH030004)

Medical Image Super-resolution Method Based on Semantic Attention

LIN Yi,ZHOU Peng, CHEN Yanming   

  1. Anhui Provincial Medical Imaging Advanced Technology National Joint Research Center, School of Computer Science, Technology, Anhui University, Hefei 230601, China
  • Published:2023-11-09
  • About author:LIN Yi,born in 1996,postgraduate.His main research interests include medical image processing and so on.
    ZHOU Peng,born in 1989,Ph.D,is a member of China Computer Federation.His main research interests include machine learning and data mining.
  • Supported by:
    National Natural Science Foundation of China(62176001,61806003) and Natural Science Project of Anhui Provincial Education Department(2023AH030004).

PDF (PC)

170

摘要： 在医学图像领域,清晰的医学图像能够帮助医生更好地诊断疾病。然而,由于受到成像设备的限制,生成的医学图像往往分辨率较低并可能影响后期诊断。因此,使用超分辨率方法提高图像的分辨率显得尤为重要。近些年来,随着深度学习的发展,基于深度学习的自然图像超分辨率方法被广泛研究,并取得了一定效果。然而,不同于自然图像超分辨率,医学图像超分辨率往往是为下游医学任务服务。许多下游医学任务,例如疾病诊断、语义分割等等,往往会对某些区域感兴趣。但是传统图像超分辨率方法往往平等地对待图像中所有区域,没有考虑到感兴趣区域对于下游医学任务的重要性。针对此问题,提出了一种基于语义注意力的医学图像超分辨率方法。该注意力机制通过加权方式对图像中感兴趣区域进行额外关注,从而使得超分辨率图像更有助于下游医学任务。该方法在新冠肺炎数据集COVID_19和胃肠息肉数据集Kvasir-SEG上都取得了领先于其他主流超分辨率方法的效果。

关键词: 医学图像, 超分辨率, 深度学习, 感兴趣区域, 语义注意力

Abstract: In the field of medical images processing,clear medical images can help doctors to diagnose diseases better.However,due to the limitations of imaging equipment,the generated medical images are often of low resolution and thus may be in appro-priate for diagnosis.Therefore,it is very important to use super-resolution method to improve the image resolution.In recent years,with the development of deep learning,natural image super-resolution methods based on deep learning have been widely studied and achieved promising performance.However,unlike natural image super-resolution,medical image super-resolution often serves downstream medical tasks.The downstream medical tasks,such as disease diagnosis and semantic segmentation,tend to be of interest to certain regions.However,traditional image super-resolution methods often tend to treat all regions in the image equally,without considering the importance of the regions of interest for downstream medical tasks.To tackle this problem,this paper proposes a medical image super-resolution method based on semantic attention.The semantic attention module pays extra attention to the regions of interest in the image by weighting,so that the super-resolution image is more helpful for downstream medical tasks.Experimental results show that the proposed method outperforms other mainstream super-resolution methods on COVID-19 dataset and gastrointestinal polyps dataset Kvasir-SEG.

Key words: Medical image, Super-resolution, Deep learning, Region of interest, Semantic attention

中图分类号: 

  • TP391
[1]ZHU H,HAN G,PENG Y,et al.Functional-Realistic CT Image Super Resolution for Early-Stage Pulmonary Nodule Detection[J].Future Generation Computer Systems,2021,115:475-485.
[2]FAN F,GAO Y,QIN P L,et al.Abdominal MRI Image Multi-ScaleSuper-Resolution Reconstruction Based on Parallel Channel-Spatial Attention Mech-anism[J].Journal of Computer Applications,2020,40(12):3624-3630.
[3]DONG C,LOY C C,HE K M,et al.Learning a Deep Convolutional Network for Image Super-Resolution[C]//European Conferenceon Computer Vision.Germany:Srpinger,2014:184-199.
[4]KIM J,LEE J K,MU LEE K M.Accurate Image Super-Resolution Using Very Deep Convolutional Networks[C]//IEEE Conference on Computer Vision and Pattern Recognition.USA:IEEE,2016:16 46-1654.
[5]LEDIG C,THEIS L,HUSZAR F,et al.Photo-Realistic SingleImage Super-Resolution Using a Generative Adversarial Network[C]//IEEE Conference on Computer Vision and Pattern Recognition.USA:IEEE,2017:105-114.
[6]ZHANG Y L,LI K P,LI K,et al.Image Super-Resolution Using Very Deep Residual Channel Attention Networks[C]//EuropeanConference on Computer Vision.USA:IEEE,2018:286-301.
[7]VASWANI A,NOAM S,PARMAR N,et al.Attention is AllYou Need[C]//Annual Conference on Neural Information Processing Systems.USA:NIPS,2017:5998-6008.
[8]LIANG J Y,GAO J Z,SUN G L,et al.SwinIR:Image Restoration Using Swin Transformer[C]//IEEE International Confe-rence on Computer Vision Workshops.USA:IEEE,2021:1833-1844.
[9]RONNEBERGER O,FISCHER P,BROX T.U-Net:Convolu-tional Networks for Biomedical Image Segmentation[C]//International Conference on Medical Image Computing and Computer Assisted Intervention.Germany:Springer,2015:234-241.
[10]LIM B,SON S,KIM H,et al.Enhanced Deep Residual Networks for Single Image Super-Resolution[C]//IEEE Conference on Computer Vision and Pattern Recognition Workshops.USA:IEEE,2017:1132-1140.
[11]DAI T,CAI J R,ZHANG Y B,et al.Second-Order Attention Network for Single Image Super-Resolution[C]//IEEE Confe-rence on Computer Vision and Pattern Recognition.USA:IEEE,2019:11065-11074.
[12]NIU V,WEN W L,REN W Q,et al.Single Image Super-Resolution via a Holistic Attention Network[C]//European Confe-rence on Computer Vision.Berlin:Springer,2020:191-207.
[13]MEI Y Q,FAN Y C,ZHOU Y Q.Image Super-Resolution With No-Local Sparse Attention[C]//IEEE Conference on Computer Vision and Pattern Recognition.USA:IEEE,2021:3517-3526.
[14]WANG X T,YU K,WU S X,et al.Esrgan:Enhanced Super-Resolution Generative Adversarial Networks[C]//European Conference on Computer Vision Workshops.Germany:Sprin-ger,2018:701-710.
[15]WANG X T,XIE L B,DONG C.Real-ESRGAN:Training Real-World Blind Super-Resolution with Pure Synthetic Data[C]//IEEE International Conference on Computer Vision Workshops.USA:IEEE,2021:1905-1914.
[16]YANG F Z,YANG H,FU J L,et al.Learning Texture Transformer Network for Image Super-Resolution[C]//IEEE Confe-rence on Computer Vision and Pattern Recognition.USA:IEEE,2020:5790-5799.
[17]YU H C,LIU D,SHI H H,et al.Computed Tomography Super-Resolution Using Convolutional Neural Networks[C]//IEEE International Conference on Image Processing.USA:IEEE,2017:3944-3948.
[18]DU X F,HE Y F.Gradient-Guided Convolutional Neural Network for MRI Image Super-Resolution[J].Applied Sciences,2019,9(22):4874.
[19]YOU C Y,LI G,ZHANG Y,et al.CT Super-Resolution GAN Constrained by the Identical,Residual,and Cycle Learning Ensemble(GAN-CIRCLE)[J].IEEE Transactions on Medical Imaging,2020,39(1):188-203.
[20]WANG J W,ZHOU P,HAN X J,et al.Medical Image Super-Resolution via Diagnosis-Guided Attention[C]//International Conference on Multimedia and Expo.USA:IEEE,2023:1-6.
[21]DAI Z X,LI J X,ZHANG X D,et al.Super-Resolution Reconstruction of MRI Based on DNGAN[J].Computer Science,2022,49(7):113-119.
[22]SHI W Z,CABALLERO J,HUSZAR F,et al.Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network[C]//IEEE Conference on Computer Vision and Pattern Recognition.USA:IEEE,2016:1874-1883.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发