Ground-glass opacity target detection in CT scans based on improved Faster R-CNN model

Shuying Yang; Dongsheng Deng; Qingchun Zheng

doi:10.11834/jig.200544

Computerd Tomography Image | Views : 0 下载量: 12 CSCD: 2

PDF
Export
Share
Collection
Album

Ground-glass opacity target detection in CT scans based on improved Faster R-CNN model
Vol. 26, Issue 9, Pages: 2171-2180(2021)
Received：15 September 2020，

Revised：17 December 2020，

Accepted：24 December 2020，

Published：16 September 2021
DOI： 10.11834/jig.200544
稿件说明：

移动端阅览

DOI：

Shuying Yang, Dongsheng Deng, Qingchun Zheng. Ground-glass opacity target detection in CT scans based on improved Faster R-CNN model[J]. Journal of image and graphics, 2021, 26(9): 2171-2180. DOI： 10.11834/jig.200544.

摘要

目的

针对Faster R-CNN（faster region convolutional neural network）模型在肺部计算机断层扫描（computed tomography，CT）图磨玻璃密度影目标检测中小尺寸目标无法有效检测与模型检测速度慢等问题，对Faster R-CNN模型特征提取网络与区域候选网络（region proposal network，RPN）提出了改进方法。

方法

使用特征金字塔网络替换Faster R-CNN的特征提取网络，生成特征金字塔；使用基于位置映射的RPN产生锚框，并计算每个锚框的中心到真实物体中心的远近程度（用参数“中心度”表示），对RPN判定为前景的锚框进一步修正位置作为候选区域（region proposal），并将RPN预测的前景/背景分类置信度与中心度结合作为候选区域的排序依据，候选区域经过非极大抑制筛选出感兴趣区域（region of interest，RoI）。将RoI对应的特征区域送入分类回归网络得到检测结果。

结果

实验结果表明，在新冠肺炎患者肺部CT图数据集上，本文改进的模型相比于Faster R-CNN模型，召回率（recall）增加了7%，平均精度均值（mean average precision，mAP）增加了3.9%，传输率（frames per second，FPS）由5帧/s提升至9帧/s。特征金字塔网络的引入明显提升了模型的召回率与mAP指标，基于位置映射的RPN显著提升了模型的检测速度。与其他最新改进的目标检测模型相比，本文改进的模型保持了双阶段目标检测模型的高精度，并拉近了与单阶段目标检测模型在检测速度指标上的距离。

结论

本文改进的模型能够有效检测到患者肺部CT图的磨玻璃密度影目标区域，对小尺寸目标同样适用，可以快速有效地为医生提供辅助诊断。

Abstract

Objective

The outbreak of corona virus disease 2019 (COVID-19) has become a serious public health event of concern worldwide. The key to controlling the spread of this disease is early detection. Computed tomography (CT) is highly sensitive to the early diagnosis of patients with COVID-19

and the changes in clinical symptoms are time-consistent with the changes in lung CT lesions

which is a simpler

faster indicator for judging changes in the condition. Faint ground-glass opacity is common in the early stage of COVID-19 lesions

and the ground-glass opacity gradually increases as the lesion progresses. Manual detection methods are time consuming

and manual detection inevitably has subjective diagnostic errors. In recent years

deep learning has made great progress in computer vision and achieved outstanding performance in the detection of lung CT scans. In the target detection task

the two-stage target detection method easily achieves a higher precision. The most representative model is faster region convolutional neural network (Faster R-CNN). However

with the increasing diversification and complexity of target detection tasks

the shortcomings of the Faster R-CNN model have also been exposed. In the detection of the ground-glass opacity target

the target size range is large

and Faster R-CNN only uses the highest layer feature map to obtain the region proposal

which has the problem of low recognition rate for small targets. When the region proposal network of Faster R-CNN model supervises the foreground/background classification

most of the overlap calculations between the anchor boxes and the background area are redundant calculations. in the task of detecting ground-glass opacity targets in CT scans of the lung and given the problems of the Faster R-CNN model

an improved method for the feature extraction network and region proposal network of the Faster R-CNN model is proposed.

Method

First

the feature pyramid network replaces the feature extraction network of Faster R-CNN to generate a feature pyramid. Then

the region proposal network based on location mapping generates anchor boxes and calculates the distance from the center of each anchor boxes to the center of the real object

which is represented by the parameter "centrality". The anchor box judged as the foreground by the region proposal network is further modified as a region proposal

and the foreground/background classification confidence predicted by the region proposal network and centrality are combined as the sorting basis for the region proposal. The interest regions are filtered out from region proposals through non-maximal suppression. Finally

the characteristic regions corresponding to regions of interest are sent to the classification regression network to obtain the detection results. Content of main experiments

the experiment uses recall

mean average precision (mAP)

and frames per second (FPS) as evaluation indicators to compare the performance of the standard Faster R-CNN

Faster R-CNN + FPS

and the proposed model

and the effects of different backbone networks on the model in this paper.

Result

On the dataset of COVID-19

the experimental results show that compared with the Faster R-CNN model

the improved model increases recall by 7%

mAP by 3.9%

and FPS from 5 to 9.

Conclusion

The improved model can effectively detect the ground-glass opacity target of the patient's lung CT scans and is suitable for small targets. The improved region proposal network reduces network output parameters

saves calculation time

and increases model running speed. Meanings using the feature pyramid network to replace the feature extraction network of Faster R-CNN can be a general method to solve the problem of a large size range of target objects. The method of using the location mapping-based region proposal network to replace the traditional multianchor box mapping-based region proposal network can also provide a reference for accelerating the running speed of the model.

关键词

Keywords

references

Bottou L. 2010. Large-scale machine learning with stochastic gradient descent//Proceedings of the 19th International Conference on Computational Statistics. Paris, France: Physica-Verlag HD: 177-186[ DOI:10.1007/978-3-7908-2604-3_16 http://dx.doi.org/10.1007/978-3-7908-2604-3_16 ]

Committee of the Infectious Diseases Radiology Group of Chinese Society of Radiology, Committee of the Infectious Diseases Radiology Section of Chinese Medical Doctor Association, Infectious Disease Imaging Group, Infectious Disease Branch, Chinese Research Hospital Association, Imaging Committee of Chinese Association for the Prevention and Treatment of STD AIDS Infection (Infectious Disease), Infectious Diseases Group, General Radiological Equipment Committee, China Association of Medical Equipment and Beijing Imaging Diagnosis and Treatment Technology Innovation Alliance.2020. Guideline for imaging diagnosis of novel coronavirus (2019-nCoV) infected pneumonia (1st edition 2020). New Medicine, 30(1): 22-34

中华医学会放射学分会传染病学组, 中国医师协会放射医师分会感染影像专委会, 中国研究型医院学会感染与炎症放射学分会, 中国性病艾滋病防治协会感染(传染病)影像工作委员会, 中国医院协会传染病分会传染病影像学组, 中国装备协会普通放射装备专业委员会传染病学组, 北京影像诊疗技术创新联盟. 2020. 新型冠状病毒感染的肺炎影像学诊断指南(2020第一版). 医学新知, 30(1): 22-34[DOI:10.12173/j.issn.1004-5511.2020.01.07]

Fu C Y, Liu W, Ranga A, Tyagi A and Berg A C. 2017. DSSD: deconvolutional single shot detector[EB/OL]. [2020-07-15] . https://arxiv.org/pdf/1701.06659.pdf https://arxiv.org/pdf/1701.06659.pdf

Girshick R. 2015. Fast R-CNN//Proceedings of 2015 IEEE International Conference on Computer Vision. Santiago, Chile: IEEE: 1440-1448[ DOI:10.1109/ICCV.2015.169 http://dx.doi.org/10.1109/ICCV.2015.169 ]

Girshick R, Donahue J, Darrell T and Malik J. 2014. Rich feature hierarchies for accurate object detection and semantic segmentation//Proceedings of 2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus, USA: IEEE: 580-587[ DOI:10.1109/CVPR.2014.81 http://dx.doi.org/10.1109/CVPR.2014.81 ]

He K, Zhang X Y, Ren S Q and Sun J. 2014. Spatial pyramid poolingin deep convolutional networks for visual recognition. IEEE Transactions on Pattern Analysis&Machine Intelligence, 37(9): 1904-1916[DOI:10.1007/978-3-319-10578-9_23]

He K M, Zhang X Y, Ren S Q and Sun J. 2016a. Deep residual learning for image recognition//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, USA: IEEE: 770-778[ DOI:10.1109/CVPR.2016.90 http://dx.doi.org/10.1109/CVPR.2016.90 ]

Hwang S, Kim H E, Jihoon J M D and Kim H J. 2016. A novel approach for tuberculosis screening based on deep convolutional neural networks//Proceedings of SPIE, Medical imaging 2016: Computer-Aided Diagnosis. San Diego, USA: SPIE: 97852 W[ DOI:10.1117/12.2216198 http://dx.doi.org/10.1117/12.2216198 ]

Islam M T, Aowal A, Minhaz A T and Ashraf K. 2017. Abnormality detection and localization in chest x-rays using deep convolutional neural networks[EB/OL]. [2019-12-15] . http://arxiv.org/pdf/1705.09850.pdf http://arxiv.org/pdf/1705.09850.pdf

Li H J. 2020. Guideline for medical imaging in auxiliary diagnosis of coronavirus disease 2019. Chinese Journal of Medical Imaging Technology, 36(3): 321-331

李宏军. 2020. 新型冠状病毒肺炎影像学辅助诊断指南. 中国医学影像技术, 36(3): 321-331[DOI:10.13929/j.issn.1003-3289.2020.03.001]

Lin T Y, Dollár P, Girshick R, He K M, Hariharan B and Belongie S. 2017. Feature pyramid networks for object detection//Proceedings of the 30th IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA: IEEE: 936-944[ DOI:10.1109/CVPR.2017.106 http://dx.doi.org/10.1109/CVPR.2017.106 ]

Liu W, Anguelov D, Erhan D, Szegedy C, Reed S, Fu C Y and Berg A C. 2016. SSD: single shot MultiBox detector//Proceedings of the 14th European Conference on Computer Vision-ECCV 2016. Amsterdam, the Netherlands: Springer: 21-37[ DOI:10.1007/978-3-319-46448-0_2 http://dx.doi.org/10.1007/978-3-319-46448-0_2 ]

Redmon J, Divvala S, Girshick R and Farhadi A. 2016. You only look once: unified, real-time object detection//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas,USA: IEEE: 779-788[ DOI:10.1109/CVPR.2016.91 http://dx.doi.org/10.1109/CVPR.2016.91 ]

Redmon J and Farhadi A. 2018. YOLOv3: an Incremental Improvement[EB/OL]. [2020-07-10] . https://arxiv.org/pdf/1804.02767.pdf https://arxiv.org/pdf/1804.02767.pdf

Ren S Q, He K M, Girshick R and Sun J. 2017. Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(6): 1137-1149[DOI:10.1109/TPAMI.2016.2577031]

Simonyan K and Zisserman A. 2015. Very deep convolutional networks for large-scale image recognition[EB/OL]. [2019-12-12] . https://arxiv.org/pdf/1409.1556.pdf https://arxiv.org/pdf/1409.1556.pdf

Tong G F, Li Y, Chen H R, Zhang Q C and Jiang H Y. 2018. Improved U-NET network for pulmonary nodules segmentation. Optik, 174: 460-469[DOI:10.1016/j.ijleo.2018.08.086]

Uijlings J R R, van de Sande K E A, Gevers T and Smeulders A W M. 2013. Selective search for object recognition. International Journal of Computer Vision, 104(2): 154-171[DOI:10.1007/s11263-013-0620-5]

Wang Y L, Qiao Z W, Liu W Y, Xu C J, Li P, Wang L Y, Zhang X C, Lin J Z, Xu Q Z, Lu Z Y, Cheng G X, Liu Q, Liu X J, Li L, Yang Q, Zhang L N, Zhang H J, Hou D L, Yin X P and Liao M Y. 2020. Guideline for imaging diagnosis of novel coronavirus infected pneumonia. Journal of Capital Medical University, 41(2): 168-173

王亚丽, 乔中伟, 刘文亚, 许传军, 李萍, 汪丽娅, 张笑春, 林吉征, 徐秋贞, 鲁植艳, 成官迅, 刘强, 刘新疆, 李莉, 杨旗, 张立娜, 张惠娟, 侯代伦, 殷小平, 廖美焱. 2020. 新型冠状病毒肺炎影像诊断指南(2020年第二版简版). 首都医科大学学报, 41(2): 168-173[DOI:10.3969/j.issn.1006-7795.2020.02.004]

Zhao K C, Suo T, Wang Y N, Huang Y, Xing B Z, Chen G Z and Zou J J. 2020. Application of chest CT scans in the diagnosis and treatment of COVID-19. Medical Journal of Wuhan University, 42(05): 704-708

赵考昌, 索涛, 王妍妮, 黄毅, 邢变枝, 陈国忠, 邹进晶. 2020. 胸部CT扫描在新型冠状病毒肺炎诊治中的应用价值. 武汉大学学报(医学版), 42(05): 704-708[DOI:10.14188/j.1671-8852.2020.0344]

Alert me when the article has been cited

提交

Multi-scale context information fusion for instance segmentation

Automatically measuring the coordinates of streetlights in vehicle-borne spherical images