基于卷积神经网络与显微高光谱的胃癌组织分类方法研究
Gastric Carcinoma Classification Based on Convolutional Neural Network and Micro-Hyperspectral Imaging
查看参考文献19篇
|
文摘
|
为了探究高光谱技术在胃癌组织病理诊断中的应用,将高光谱成像与显微系统结合,采集胃部切片组织的高光谱图像。针对胃癌组织与胃部正常组织在410~910nm波段的光谱特性差异,提出了一种基于卷积神经网络模型的胃癌组织分类方法,对原始光谱进行S-G平滑和一阶导数等预处理,通过分析光谱数据的特点和模型的分类效率,确定了最佳的网络结构及参数。实验结果表明:该模型对胃部癌变和正常组织的分类准确率为96.53%,鉴别胃癌组织的灵敏度和特异性分别达到94.29%和97.14%;相比于浅层学习方法,卷积神经网络模型能够充分提取癌变组织的深层光谱特征,同时能有效避免过拟合现象。将深度学习理论与显微高光谱结合的方法为医学病理研究提供了新思路。 |
|
其他语种文摘
|
In order to explore the application of hyperspectral technology in the pathological diagnosis of gastric cancer,we combine hyperspectral imaging and microscopy to acquire hyperspectral images of gastric slices. According to the difference of spectral characteristics between gastric cancer tissue and normal gastric tissue in the wavelength of 410-910nm,we propose a classification method based on convolutional neural network(CNN).The original spectrum is preprocessed by S-G smoothing and the first order derivative.We establish the optimal network structure and parameters by analyzing the spectral data characteristics and the classification efficiency.Experimental results show that the classification accuracy of cancerous and normal gastric tissues is 96.53%,the sensitivity and specificity of distinguishing gastric carcinoma reach 94.29% and 97.14%,respectively.Compared with shallow learning methods,the CNN model can fully extract the deep spectral characteristics of cancerous tissues and effectively prevent over-fitting.The method of deep learning combined with micro-hyperspectral imaging can also provide a new idea for the medical pathology research. |
|
来源
|
光学学报
,2018,38(6):0617001-1-0617001-7 【核心库】
|
|
DOI
|
10.3788/AOS201838.0617001
|
|
关键词
|
光谱学
;
胃癌组织分类
;
卷积神经网络
;
显微高光谱成像
;
深度学习
|
|
地址
|
1.
中国科学院西安光学精密机械研究所, 中国科学院光谱成像技术重点实验室, 陕西, 西安, 710119
2.
中国科学院大学, 北京, 100049
|
|
语种
|
中文 |
|
文献类型
|
研究性论文 |
|
ISSN
|
0253-2239 |
|
学科
|
物理学 |
|
基金
|
国家重点研发计划
;
国家自然科学基金
;
中国科学院光谱成像技术重点实验室开发基金
|
|
文献收藏号
|
CSCD:6268513
|
参考文献 共
19
共1页
|
|
1.
Ferlay J. Cancer incidence and mortality worldwide:sources,methods and major patterns in GLOBOCAN 2012.
International Journal of Cancer,2015,136(5):359-386
|
被引
1098
次
|
|
|
|
|
2.
Vodinh T. A hyperspectral imaging system for in vivo optical diagnostics.
IEEE Engineering in Medicine and Biology Magazine,2004,23(5):40-49
|
被引
5
次
|
|
|
|
|
3.
Luo B. Robust autodial morphological profiles for the classification of high-resolution satellite images.
IEEE Transactions on Geoscience and Remote Sensing,2013,52(2):1451-1462
|
被引
2
次
|
|
|
|
|
4.
侯榜焕. 面向高光谱图像分类的空谱半监督局部判别分析.
光学学报,2017,37(7):0728002
|
被引
8
次
|
|
|
|
|
5.
Suzuki Y. Mapping the spatial distribution of botanical composition and herbage mass in pastures using hyperspectral imaging.
Grassland Science,2012,58(1):1-7
|
被引
2
次
|
|
|
|
|
6.
候华毅. 皮肤胆固醇无创光谱检测模拟和在体实验研究.
中国激光,2016,43(9):0907001
|
被引
4
次
|
|
|
|
|
7.
董安国. 基于谱聚类和稀疏表示的高光谱图像分类算法.
光学学报,2017,37(8):0828005
|
被引
15
次
|
|
|
|
|
8.
Akbari H. Detection of cancer metastasis using a novel macroscopic hyperspectral method.
SPIE. 8317,2012:831711
|
被引
1
次
|
|
|
|
|
9.
刘洪英. 超光谱成像技术应用于神经分类的研究.
光谱学与光谱分析,2015,35(1):38-43
|
被引
2
次
|
|
|
|
|
10.
Zhu S. Identification of cancerous gastric cells based on common features extracted from hyperspectral microscopic images.
Biomedical Optics Express,2015,6(4):1135-1145
|
被引
4
次
|
|
|
|
|
11.
Akbari H. Hyperspectral imaging and quantitative analysis for prostate cancer detection.
Journal of Biomedical Optics,2012,17(7):076005
|
被引
9
次
|
|
|
|
|
12.
Gerstner A O. Hyperspectral imaging of mucosal surfaces in patients.
Journal of Biophotonics,2012,5(3):255-262
|
被引
4
次
|
|
|
|
|
13.
Liu Z. Tongue tumor detection in medical hyperspectral images.
Sensors,2012,12(1):162-174
|
被引
8
次
|
|
|
|
|
14.
原江伟. 基于声光可调滤波器的肺癌组织快速显微光谱成像.
中国激光,2018,45(4):0407003
|
被引
3
次
|
|
|
|
|
15.
Hinton G E. Reducing the dimensionality of data with neural networks.
Science,2006,313(5786):504-507
|
被引
1674
次
|
|
|
|
|
16.
Lecun Y. Gradient-based learning applied to document recognition.
Proceedings of the IEEE,1998,86(11):2278-2324
|
被引
2031
次
|
|
|
|
|
17.
Karen S. Very deep convolutional networks for large-scale image recognition.
Proceedings of ICLR,2015:1-14
|
被引
1
次
|
|
|
|
|
18.
Hinton G E. Improving neural networks by preventing co-adaptation of feature detectors.
Computer Science,2012,3(4):212-223
|
被引
168
次
|
|
|
|
|
19.
Zuo Z. Learning contextual dependence with convolutional hierarchical recurrent neural networks.
IEEE Transactions on Image Processing,2016,25(7):2983-2996
|
被引
1
次
|
|
|
|
|
了解气象卫星更多信息，请访问