ZFNet模型在胶质瘤MRI诊断中的应用

doi:10.3969/j.issn.1672-6731.2021.03.006

摘要/Abstract

摘要：

研究背景 胶质瘤疾病负担较重，尽早确诊和及时治疗可有效延长无进展生存期，临床实践中初诊疑似胶质瘤时首选头部MRI检查，人工阅片存在诊断结果不一致和阅片效率下降的缺陷，而通过深度学习算法进行医学影像识别与诊断成为可能。本研究采用人工神经网络相关机器学习算法，辅助影像科医师对胶质瘤患者头部MRI图像的人工阅片，以期改善人工阅片耗时、费力以及因主观判断导致阅片结果不同的缺陷。方法纳入TCIA数据库中130例成年胶质瘤患者计40 036张头部MRI图像，随机分为训练集（28 025张）和测试集（12 011张），再根据医学专家的标注定义为“肿瘤影像”和“正常影像”，采用ZFNet模型进行图像识别与分类模型的建立，绘制训练集的强化学习曲线，观察训练准确度随训练步数变化的趋势。将测试集导入模型，计算ZFNet模型预测“肿瘤影像”的分类准确率、阳性预测值、灵敏度、特异度和F1值。同时进行AlexNet模型对比建模，与ZFNet模型结果进行比较。结果 ZFNet模型在训练38 757步后训练准确度稳定为99.7%，AlexNet模型则在训练37 984步后稳定为98.23%；将测试集导入ZFNet模型，ZFNet模型预测“肿瘤影像”的准确度为84.42%（10 140/12 011）、阳性预测值为80.77%（4817/5964）、灵敏度为86.93%（4817/5541）、特异度为82.27%（5323/6470）、F1值为83.74%，AlexNet模型为80.74%（9698/12 011）、77.68%（4529/5830）、81.74%（4529/5541）、79.89%（5169/6470）和79.66%，ZFNet模型在各个维度的分类性能均优于AlexNet模型，效果满意。结论 ZFNet模型在胶质瘤患者头部MRI图像分类预测方面的效果尚佳，可为建立胶质瘤影像学辅助诊断模型提供良好的技术支持。

关键词: 神经胶质瘤, 人工智能, 磁共振成像

Abstract:

Background Glioma is a kind of intracranial space-occupying lesion, which causes heavy disease burden. For glioma patients, early accurate diagnosis and early treatment could effectively prolong the progression free survival. Suspected glioma patients would be firstly examined by MRI for diagnosis. The MRI images were mainly read by radiologists artificially. The diagnosis by different radiologists might be different, and the heavy workload might cause a decline in reading efficiency. In recent years, it is becoming possible to use deep learning technology for medical image recognition and diagnosis. This study used machine learning algorithms related to artificial neural network (ANN) to assist image practitioner in reading the head MRI images of patients with glioma, which may cause labor-saving, and could increase the efficiency of reading images and reduce the different results cause by different personal experiencse. Methods These images were from The Cancer Imaging Archive (TCIA) database. Format of these images was DICOM. And they were from 130 adult glioma cases, a total of 40 036 copies. These images were randomly split as training set (28 025 copies) and test set (12 011 copies). Then in each set, images were split as "tumor image" and "normal image" according to medical experts' annotation. ZFNet model, a type of convolutional neural network, was used to build image recognition and classification model. The reinforcement learning curve was draw to observe the trend of accuracy of training changed with the training steps. Put the test set into the model, the overall classification accuracy of all MRI images, the positive predictive value, sensitivity, specificity and F1-measure of the tumor images were calculated. At the same time, AlexNet was also used to build a same model to compare with the ZFNet model by the prediction indexes of the classification ability of MRI images (the prediction ability of diagnosis of glioma). Results The training accuracy of ZFNet model was 99.7% after 38 757 steps and of the AlexNet model was 98.23% after 37 984 steps. After testing, the image prediction accuracy of all MRI images of ZFNet model was 84.42% (10 140/12 011), the positive predictive value of prediction of "tumor image" was 80.77% (4817/5964), the sensitivity was 86.93% (4817/5541), the specificity was 82.27% (5323/6470), and the F1-measure was 83.74%. The above indexes of AlexNet model were 80.74% (9698/12 011), 77.68% (4529/5830), 81.74% (4529/5541), 79.89% (5169/6470) and 79.66%, respectively. The classification performances of ZFNet model were satisfied and were superior to AlexNet model in each dimension. Conclusions ZFNet model has a good prediction performance ability in glioma brain MRI image classification, and it is able to provide a good technical support for establishing a glioma image aided diagnosis model.

Key words: Glioma, Artificial intelligence, Magnetic resonance imaging

井奚月, 乔婕, 么秀华, 徐立霞, 闫华. ZFNet模型在胶质瘤MRI诊断中的应用[J]. 中国现代神经疾病杂志, 2021, 21(3): 156-161.

JING Xi-yue, QIAO Jie, YAO Xiu-hua, XU Li-xia, YAN Hua. The application of ZFNet model on the MRI diagnosis of glioma[J]. Chinese Journal of Contemporary Neurology and Neurosurgery, 2021, 21(3): 156-161.

http://journal11.magtechjournal.com/cjcnn/CN/Y2021/V21/I3/156

参考文献

[1] Ostrom QT, Gittleman H, Truitt G, Boscia A, Kruchko C, Barnholtz-Sloan JS. CBTRUS Statistical Report:primary brain and other central nervous system tumors diagnosed in the United States in 2011-2015[J]. Neuro Oncol, 2018, 20(Suppl_4):iv1-86.
[2] Louis DN, Perry A, Reifenberger G, von Deimling A, Figarella-Branger D, Cavenee WK, Ohgaki H, Wiestler OD, Kleihues P, Ellison DW. The 2016 World Health Organization Classification of Tumors of the Central Nervous System:a summary[J]. Acta Neuropathol, 2016, 131:803-820.
[3] Glioma Professional Committee of Neurosurgeon Branch for Chinese Medical Doctor Association. Chinese experts' consensus about multidisciplinary diagnosis and treatment of glioma (MDT)[J]. Zhonghua Shen Jing Wai Ke Za Zhi, 2018, 34:113-118.[中国医师协会神经外科医师分会脑胶质瘤专业委员会. 胶质瘤多学科诊治(MDT)中国专家共识[J]. 中华神经外科杂志, 2018, 34:113-118.]
[4] Dissaux G, Basse V, Schick U, El Kabbaj O, Auberger B, Magro E, Kassoul A, Abgral R, Salaun PY, Bourhis D, Querellou S. Prognostic value of 18F-FET PET/CT in newly diagnosed WHO 2016 high-grade glioma[J]. Medicine (Baltimore), 2020, 99:e19017.
[5] Editing group of Chinese guidelines for diagnosis and treatment of glioma of the central nervous system. Chinese guidelines for diagnosis and treatment of glioma of the central nervous system (2015)[J]. Zhonghua Yi Xue Za Zhi, 2016, 96:485-509.[《中国中枢神经系统胶质瘤诊断和治疗指南》编写组. 中国中枢神经系统胶质瘤诊断与治疗指南(2015)[J]. 中华医学杂志, 2016, 96:485-509.]
[6] Ricci PE, Dungan DH. Imaging of low-and intermediate-grade gliomas[J]. Semin Radiat Oncol, 2001, 11:103-112.
[7] Upadhyay N, Waldman AD. Conventional MRI evaluation of gliomas[J]. Br J Radiol, 2011, 84(Spec ISS2):107-111.
[8] Yang Y, He MZ, Li T, Yang X. MRI combined with PET-CT of different tracers to improve the accuracy of glioma diagnosis:a systematic review and meta-analysis[J]. Neurosurg Rev, 2019, 42:185-195.
[9] Vézina LG. Imaging of central nervous system tumors in children:advances and limitations[J]. J Child Neurol, 2008, 23:1128-1135.
[10] Chan HP, Samala RK, Hadjiiski LM, Zhou C. Deep learning in medical image analysis[J]. Adv Exp Med Biol, 2020, 1213:3-21.
[11] May M. Life science technologies:big biological impacts from big data[J]. Science, 2014, 344:1298-1300.
[12] Gao F, Yue Z, Wang J, Sun J, Yang E, Zhou H. A novel active semisupervised convolutional neural network algorithm for SAR image recognition[J]. Comput Intell Neurosci, 2017:ID3105053.
[13] Chen H, Zhang Y, Zhang W, Liao P, Li K, Zhou J, Wang G. Low-dose CT via convolutional neural network[J]. Biomed Opt Express, 2017, 8:679-694.
[14] Kang E, Min J, Ye JC. A deep convolutional neural network using directional wavelets for low-dose X-ray CT reconstruction[J]. Med Phys, 2017, 44:e360-375.
[15] Kang E, Ye JC. Wavelet Domain Residual Network (WavResNet) for low-dose X-ray CT reconstruction[EB/OL]. (2017-03-03)[2021-02-22]. https://arxiv.org/pdf/1703.01383.pdf.
[16] Diederich B, Wartmann R, Schadwinkel H, Heintzmann R. Using machine-learning to optimize phase contrast in a low-cost cellphone microscope[J]. PLoS One, 2018, 13:e0192937.
[17] Shin HC, Roth HR, Gao M, Lu L, Xu Z, Nogues I, Yao J, Mollura D, Summers RM. Deep convolutional neural networks for computer-aided detection:CNN architectures, dataset characteristics and transfer learning[J]. IEEE Trans Med Imaging, 2016, 35:1285-1298.
[18] Suzuki K. Overview of deep learning in medical imaging[J]. Radiol Phys Technol, 2017, 10:257-273.
[19] Li W, Cao P, Zhao D, Wang J. Pulmonary nodule classification with deep convolutional neural networks on computed tomography images[J]. Comput Math Methods Med, 2016:ID6215085.
[20] Krizhevsky A, Sutshever I, Hinton GE. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60:84-90.
[21] Matthew DZ, Rob F. Visualizing and Understanding Convolutional Networks[EB/OL]. (2013-11-28)[2021-02-22]. https://arxiv.org/pdf/1311.2901.pdf.
[22] Renganathan V. Text mining in biomedical domain with emphasis on document clustering[J]. Healthc Inform Res, 2017, 23:141-146.
[23] Erickson BJ, Korfiatis P, Akkus Z, Kline TL. Machine learning for medical imaging[J]. Radiographics, 2017, 37:505-515.
[24] Kunimatsu A, Kunimatsu N, Yasaka K, Akai H, Kamiya K, Watadani T, Mori H, Abe O. Machine learning-based texture analysis of contrast-enhanced MR imaging to differentiate between glioblastoma and primary central nervous system lymphoma[J]. Magn Reson Med Sci, 2019, 18:44-52.
[25] Li HM. Application of artificial intelligence based on radiomics in glioma MR imaging[J]. Guoji Yi Xue Fang She Xue Za Zhi, 2019, 42:531-534.[李惠明. 基于影像组学的人工智能在脑胶质瘤MRI诊断中的应用[J]. 国际医学放射学杂志, 2019, 42:531-534.]
[26] Yang Y, Yan LF, Zhang X, Nan HY, Hu YC, Han Y, Zhang J, Liu ZC, Sun YZ, Tian Q, Yu Y, Sun Q, Wang SY, Zhang X, Wang W, Cui GB. Optimizing texture retrieving model for multimodal MR image-based support vector machine for classifying glioma[J]. J Magn Reson Imaging, 2019, 49:1263-1274.
[27] Tian Q, Yan LF, Zhang X, Zhang X, Hu YC, Han Y, Liu ZC, Nan HY, Sun Q, Sun YZ, Yang Y, Yu Y, Zhang J, Hu B, Xiao G, Chen P, Tian S, Xu J, Wang W, Cui GB. Radiomics strategy for glioma grading using texture features from multiparametric MRI[J]. J Magn Reson Imaging, 2018, 48:1518-1528.
[28] Cho HH, Lee SH, Kim J, Park H. Classification of the glioma grading using radiomics analysis[J]. PeerJ, 2018, 6:e5982.
[29] Zhang X, Yan LF, Hu YC, Li G, Yang Y, Han Y, Sun YZ, Liu ZC, Tian Q, Han ZY, Liu LD, Hu BQ, Qiu ZY, Wang W, Cui GB. Optimizing a machine learning based glioma grading system using multi-parametric MRI histogram and texture features[J]. Oncotarget, 2017, 8:47816-47830.
[30] Hsieh KL, Lo CM, Hsiao CJ. Computer-aided grading of gliomas based on local and global MRI features[J]. Comput Methods Programs Biomed, 2017, 139:31-38.
[31] Sui Y, Xiong Y, Jiang J, Karaman MM, Xie KL, Zhu W, Zhou XJ. Differentiation of low- and high-grade gliomas using high b-value diffusion imaging with a non-gaussian diffusion model[J]. AJNR Am J Neuroradiol, 2016, 37:1643-1649.
[32] Hinton GE, Osindero S, Teh YW. A fast learning algorithm for deep belief nets[J]. Neural Comput, 2006, 18:1527-1554.

选择文件类型/文献管理软件名称

选择包含的内容

2 ZFNet模型在胶质瘤MRI诊断中的应用

The application of ZFNet model on the MRI diagnosis of glioma

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	潘希娟, 邢英琦, 刘玉梅. 脑小血管病影像学评分方法[J]. 中国现代神经疾病杂志, 2023, 23(7): 633-637.
[2]	邵暇荔, 张泳, 金莉蓉, 王剑. 震颤为主型帕金森病与特发性震颤患者脑灌注对比分析[J]. 中国现代神经疾病杂志, 2023, 23(6): 542-548.
[3]	刘智强, 陈金桃, 唐文龙, 翁超群, 张金锋, 林志雄. Dandy-Walker综合征患儿腹腔分流术前后临床与影像学特征分析[J]. 中国现代神经疾病杂志, 2023, 23(5): 433-438.
[4]	邹鹏, 罗鹏, 张昊阜子, 刘剑, 程光, 蒋晓帆. 脑胶质瘤清醒麻醉术中癫痫发作对术后情绪及肌力的影响[J]. 中国现代神经疾病杂志, 2023, 23(5): 454-459.
[5]	姚莉萍, 王睿晗, 包翌, 蔡瀚林, 陈芹. 后部皮质萎缩临床及影像学特征分析[J]. 中国现代神经疾病杂志, 2023, 23(4): 326-334.
[6]	葛懿, 陈聪, 王爽, 丁瑶. 7T超高场强MRI在耐药性癫痫诊断与治疗中的应用进展[J]. 中国现代神经疾病杂志, 2023, 23(3): 161-166.
[7]	耿雨梅, 李存, 李慧敏, 王梦莹, 康慧聪. 伴癫痫发作的自身免疫性脑炎临床特征分析[J]. 中国现代神经疾病杂志, 2023, 23(3): 205-213.
[8]	李建瑞, 刘宵雪, 许强, 骆仲强, 卢光明, 张志强. 扩散峰度成像直方图联合EphA2分级在胶质瘤分级诊断中的价值[J]. 中国现代神经疾病杂志, 2023, 23(3): 254-263.
[9]	巫嘉陵, 韩建达. 医疗人工智能：知识引导与数据挖掘联合驱动[J]. 中国现代神经疾病杂志, 2023, 23(1): 1-4.
[10]	孙梦凡, 姜季委, 王艳丽, 张源, 徐俊. 动脉自旋标记测定脑血流量在阿尔茨海默病研究中的进展[J]. 中国现代神经疾病杂志, 2023, 23(1): 29-34.
[11]	于宁波, 游煜根, 王鸿鹏, 秦岩丁, 梁思泉, 韩建达. 机器人辅助神经介入手术研究进展[J]. 中国现代神经疾病杂志, 2023, 23(1): 45-52.
[12]	李佩湖, 封浑, 李吕力, 姜炳坚. 以脑卒中样发病的可逆性胼胝体压部病变综合征Ⅱ型二例[J]. 中国现代神经疾病杂志, 2022, 22(9): 817-822.
[13]	姚盈盈, 王雷明, 滕梁红. 脑胶质瘤DNA甲基化临床意义及研究进展[J]. 中国现代神经疾病杂志, 2022, 22(9): 823-828.
[14]	姚庆宇, 马龙冰, 菅凤增. 脊髓空洞症大鼠模型长期稳定性评价[J]. 中国现代神经疾病杂志, 2022, 22(8): 662-668.
[15]	沈东超, 邹鹏, 侯波, 杨洵哲, 刘明生, 崔丽英. 肢体发病与延髓发病的肌萎缩侧索硬化症患者脑灰质结构成像对比研究[J]. 中国现代神经疾病杂志, 2022, 22(7): 601-607.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

2 ZFNet模型在胶质瘤MRI诊断中的应用

The application of ZFNet model on the MRI diagnosis of glioma

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价