A computer aided diagnosis system for metacarpal and phalangeal fracture based on GA_Faster R-CNN

doi:10.3969/j.issn.1000-1565.2021.04.011

Abstract

Abstract: In clinical diagnosis, metacarpal and phalangeal fractures are difficult to be detected and easy to be misdiagnosed. Therefore, an experimental study was carried out on accurate detection and location of the fractures in X-ray image. A computer-aided diagnosis(CAD)system based on deep learning was developed by improving the anchor frame, feature map generation and loss function of Faster R-CNN. The proposed CAD system greatly enhanced the network performance while saving computing power. 5195 X-ray images of hand were collected, of which 4675 were randomly selected as training set and 520 as test set. The experimental results show that the average precision of the proposed model in the test set is 69.3%. Compared with Faster R-CNN, the new framework can identify the location of metacarpal and phalangeal fractures more accurately, which verifies the potential clinical application value of the new model.- DOI:10.3969/j.issn.1000-1565.2021.04.011一种基于GA_Faster R-CNN的掌指骨骨折计算机辅助诊断系统杨昆^1,2,罗萍¹,吕一品¹,闫惟娜¹,吴海涛¹,刘爽^1,2,薛林雁^1,2(1.河北大学质量技术监督学院,河北保定 071002;2.河北大学计量仪器与系统国家地方联合工程研究中心,河北保定 071002)摘要:为解决临床掌指骨骨折诊断中难检测、易漏诊的问题,对如何更加准确地检测和定位X线图像中掌指骨骨折区域进行实验研究.开发了一种基于深度学习的计算机辅助诊断系统,改进了Faster R-CNN网络模型中的锚框、特征映射和损失函数,在节约算力的同时大大提高了网络性能.建立了包含5 195张手部的X线图像的数据集,其中随机抽取4 675张作为训练集,520张作为测试集.实验表明,提出的新模型在测试集上的平均精度达到了69.3%,与Faster R-CNN相比能够更加精准地识别出掌指骨骨折区域,验证了该模型具有潜在的临床应用价值.关键词:掌指骨骨折;Faster R-CNN;计算机辅助诊断;导向锚定中图分类号:TP391.7 文献标志码:A 文章编号:1000-1565(2021)04-0412-07A computer aided diagnosis system for metacarpal and phalangeal fracture based on GA_Faster R-CNNYANG Kun^1,2,LUO Ping¹,LYU Yipin¹,YAN Weina¹,WU Haitao¹,LIU Shuang^1,2,XUE Linyan^1,2(1.College of Quality and Technical Supervision, Hebei University, Baoding 071002, China; 2.National & Local Joint Engineering Research Center of Metrology Instrument and System, Hebei University, Baoding 071002, China)Abstract: In clinical diagnosis, metacarpal and phalangeal fractures are difficult to be detected and easy to be misdiagnosed. Therefore, an experimental study was carried out on accurate detection and location of the fractures in X-ray image. A computer-aided diagnosis(CAD)system based on deep learning was developed by improving the anchor frame, feature map generation and loss function of Faster R-CNN. The proposed CAD system greatly enhanced the network performance while saving computing power. 5195 X-ray images of hand were collected, of which 4675 were randomly selected as training set and 520 as test set. The experimental results show that the average precision of the proposed model in the test set is 69.3%. Compared with Faster R-CNN, the new framework can identify the location of metacarpal and phalangeal fractures more accurately, which verifies the potential clinical application value of the new model.- 收稿日期:2020-09-23 基金项目:河北省自然科学基金资助项目(H2019201378);河北大学自然科学多学科交叉研究计划资助项目(DXK201914);河北大学校长科研基金资助项目(XZJJ201914;XZJJ201917;XZJJ201918) 第一作者:杨昆(1976—),男,河北保定人,河北大学教授,博士生导师,博士,主要从事生物医学图像处理方向研究.E-mail:hbuyangkun@163.com 通信作者:薛林雁(1981—),女,河北广平人,河北大学副教授,博士,主要从事生物医学图像处理方向研究.E-mail:lineysnow@163.com第4期杨昆等:一种基于GA_Faster R-CNN的掌指骨骨折计算机辅助诊断系统

Key words: metacarpal and phalangeal fracture, Faster R-CNN, computer aided diagnosis(CAD), guided anchoring

CLC Number:

TP391.7

YANG Kun,LUO Ping,LYU Yipin,YAN Weina,WU Haitao,LIU Shuang,XUE Linyan. A computer aided diagnosis system for metacarpal and phalangeal fracture based on GA_Faster R-CNN[J]. Journal of Hebei University(Natural Science Edition), 2021, 41(4): 412-418.

References

[1] 王海洲,王冠.基于深度学习的人工智能医疗应用与存在的问题[J].沈阳大学学报(自然科学版), 2019, 31(2): 118-121. DOI: 10.16103/j.cnki.21-1583/n.2019.02.007.
[2] 邹文凯,陆慧娟,叶敏超,等.基于卷积神经网络的乳腺癌组织病理图像分类[J].计算机工程与设计, 2020, 41(6): 1749-1754. DOI: 10.16208/j.issn1000-7024.2020.06.040.
[3] I??塁IN A, DIREKOGˇLU C, ??塁AH M. Review of MRI-based brain tumor image segmentation using deep learning methods[J]. Procedia Computer Science, 2016, 102: 317-324. DOI: 10.1016/j.procs.2016.09.407.
[4] ZHU W T, LIU C C, FAN W, et al. DeepLung: Deep 3D dual path nets for automated pulmonary nodule detection and classification[Z]. IEEE Winter Conference on Applications of Computer Vision(WACV), Lake Tahoe, USA, 2018. DOI: 10.1109/WACV.2018.00079.
[5] 林秀娇,张栋,黄明毅,等.计算机辅助诊断根尖X线片图像中恒牙邻面龋初探[J].中华口腔医学杂志, 2020, 55(9): 654-660. DOI: 10.3760/cma.j.cn112144-20200209-00040.
[6] CHUNG S W, HAN S S, LEE J W, et al. Automated detection and classification of the proximal humerus fracture by using deep learning algorithm[J]. Acta Orthop, 2018, 89(4): 468-473 DOI: 10.1080/17453674.2018.1453714.
[7] 周清清,王佳硕,唐雯,等.基于卷积神经网络成人肋骨骨折CT自动检测和分类的应用研究[J].影像诊断与介入放射学, 2019, 29(1): 27-31. DOI: 10.3969/j.issn.1005-8001.2020.01.005.
[8] BADGELEY M A, ZECH J R, OAKDEN-RAYNER L, et al. Deep learning predicts hip fracture using confounding patient and healthcare variables[J]. NPJ Digit Med, 2019, 2: 31. DOI:10.1038/s41746-019-0105-1.
[9] GUAN B, YAO J K, ZHANG G S, et al. Thigh fracture detection using deep learning method based on new dilated convolutional feature pyramid network[J]. Pattern Recognition Letters, 2019, 125: 521-526. DOI: 10.1016/j.patrec.2019.06.015.
[10] KITAMURA G, CHUNG C Y, MOORE B E. Ankle fracture detection utilizing a convolutional neural network ensemble implemented with a small sample, de novo training, and multiview incorporation[J]. J Digit Imaging, 2019, 32(4): 672-677. DOI: 10.1007/s10278-018-0167-7.
[11] REN S Q, HE K M, GIRSHICK R, et al. Faster R-CNN: Towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137-1149. DOI: 10.1109/TPAMI.2016.2577031.
[12] 曹之君,张良.基于Faster-RCNN的快速目标检测算法[J].航天控制, 2020, 38(4): 49-55. DOI: 10.16804/j.cnki.issn1006-3242.2020.04.004.
[13] UIJLINGS J R R, SANDE K E A, GEVERS T, et al. Selective search for object recognition[J]. International Journal of Computer Vision, 2013, 104(2): 154-171. DOI: 10.1007/s11263-013-0620-5.
[14] GIRSHICK R. Fast R-CNN[Z]. IEEE International Conference on Computer Vision(ICCV), Santiago, Chile, 2015. DOI: 10.1109/ICCV.2015.169.
[15] WANG J Q, CHEN K, YANG S, et al. Region proposal by guided anchoring[Z]. IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR), Long Beach, USA, 2019. DOI: 10.1109/CVPR.2019.00308
[16] LIN T Y,DOLLAR P.GIRSHICK R,et al.Feature pyramid networks for object detection[Z].IEEE Conference on Computer Vision and Pattern Recognition(CVPR),Hawaii,USA,2017.DOI:10.1109/CVPR.2017.106.
[17] KENDALL A, GAL Y, CIPOLLA R. Multi-task learning using uncertainty to weigh losses for scene geometry and semantics[Z]. IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, 2018. DOI: 10.1109/CVPR.2018.00781.
[18] PANG J M, CHEN K, SHI J P, et al. Libra R-CNN: Towards balanced learning for object detection[Z]. IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR), Long Beach, USA, 2019. DOI: 10.1109/CVPR.2019.00091. (