基于多尺度融合卷积神经网络的熔解曲线有效性分类

doi:10.13229/j.cnki.jdxbgxb20200871

Abstract

Abstract:

Aiming at the issue of peak classification in melting curve images， a convolutional neural network （CNN） classification model based on multi-scale fusion is proposed. First， the multi-scale context information obtained through dilated convolution is integrated with the feature information extracted from the residual module， which makes up for the shortcoming of the deep network that loses the global information. Then， different from traditional convolution with constant convolution kernel， a dynamic filter that changes with the input is used to make the network learning more accurate. In addition， in order to train the model， a dataset of melting curves is created， which includes a balanced dataset and an unbalanced dataset. This paper uses six classification-based evaluation indicators to compare with six classification methods based on deep learning. Experimental results show that the proposed method is significantly better than other methods in objective evaluation indicators.

Key words: image classification, convolutional neural networks, multi-scale fusion, melting curve, dynamic filter

CLC Number:

TP391

Xiang-jun LI,Jie-ying TU,Zhi-bin ZHAO. Validity classification of melting curve based on multi⁃scale fusion convolutional neural network[J].Journal of Jilin University(Engineering and Technology Edition), 2022, 52(3): 633-639.

Figures/Tables 6

Fig.1

Fig.2

Table 1

Table 2

Table 3

Fig.3

References 22

1	Jiang Y, He L, Wu P, et al. Simultaneous identification of ten bacterial pathogens using the multiplex ligation reaction based on the probe melting curve analysis[J]. Scientific Reports, 2017, 7(1): 1-9.
2	Ahmed F E, Gouda M M, Hussein L A, et al. Role of melt curve analysis in interpretation of Nutrigenomics' MicroRNA expression data[J]. Cancer Genomics-Proteomics, 2017, 14(6): 469-481.
3	Min S, Lee B, Yoon S. Deep learning in bioinformatics[J]. Briefings in bioinformatics, 2017, 18(5): 851-869.
4	Tang B, Pan Z, Yin K, et al. Recent advances of deep learning in bioinformatics and computational biology[J]. Frontiers in Genetics, 2019, 10: 214.
5	赵宏伟, 刘晓涵, 张媛, 等. 基于关键点注意力和通道注意力的服装分类算法[J]. 吉林大学学报:工学版, 2020, 50(5): 1765-1770.
	Zhao Hong-wei, Liu Xiao-han, Zhang Yuan, et al. Clothing classification algorithm based on landmark attention and channel attention[J]. Journal of Jilin University (Engineering and Technology Edition), 2020, 50(5): 1765-1770.
6	Lakhani P, Sundaram B. Deep learning at chest radiography: automated classification of pulmonary tuberculosis by using convolutional neural networks[J]. Radiology, 2017, 284(2): 574-582.
7	陈绵书, 于录录, 苏越,等. 基于卷积神经网络的多标签图像分类[J]. 吉林大学学报:工学版, 2020, 50(3):1077-1084.
	Chen Mian-shu, Yu Lu-lu, Su Yue, et al. Multi-label images classification based on convolutional neural network [J]. Journal of Jilin University (Engineering and Technology Edition), 2020, 50(3): 1077-1084.
8	Wang F, Jiang M, Qian C, et al. Residual attention network for image classification[C]∥Proceedings of the 30th IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, USA, 2017: 3156-3164.
9	Phan H, Andreotti F, Cooray N, et al. Joint classification and prediction CNN framework for automatic sleep stage classification[J]. IEEE Transactions on Biomedical Engineering, 2018, 66(5): 1285-1296.
10	He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]∥Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition, Seattle, USA, 2016: 770-778.
11	Yu F, Koltun V, Funkhouser T. Dilated residual networks[C]∥Proceedings of the 30th IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, USA, 2017: 472-480.
12	Jia X, De Brabandere B, Tuytelaars T, et al. Dynamic filter networks[C]∥ Proceedings of the 30th International Conference on Neural Information Processing Systems, Barcelona, Spain, 2016: 667-675.
13	Landi F, Baraldi L, Corsini M, et al. Embodied vision-and-language navigation with dynamic convolutional filters[J]. arXiv Preprint arXiv:, 2019.
14	Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need[C]∥ Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, USA, 2017: 5998-6008.
15	Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[J]. arXiv Preprint arXiv:, 2014.
16	Iandola F N, Han S, Moskewicz M W, et al. SqueezeNet: alexNet-level accuracy with 50x fewer parameters and < 0.5 MB model size[J]. arXiv Preprint arXiv:, 2016.
17	Szegedy C, Liu W, Jia Y, et al. Going deeper with convolutions[C]∥Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition, Boston, USA, 2015: 1-9.
18	Tan M X, Chen B, Pang R M, et al. Mnasnet: platform-aware neural architecture search for mobile[C]∥Proceedings of the 32nd IEEE Conference on Computer Vision and Pattern Recognition, Long Beach, USA, 2019: 2820-2828.
19	Howard A, Sandler M, Chu G, et al. Searching for mobilenetv3[C]∥Proceedings of the 2019 IEEE International Conference on Computer Vision, Seoul, South Korea, 2019: 1314-1324.
20	Howard A G, Zhu M, Chen B, et al. Mobilenets: efficient convolutional neural networks for mobile vision applications[J]. arXiv Preprint arXiv:, 2017.
21	Sandler M, Howard A, Zhu M, et al. Mobilenetv2: Inverted residuals and linear bottlenecks[C]∥Proceedings of the 31st IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, 2018: 4510-4520.
22	Huang G, Liu Z, Van Der Maaten L, et al. Densely connected convolutional networks[C]∥Proceedings of the 30th IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, USA, 2017: 4700-4708.

Related Articles 10

[1]	Liang DUAN,Chun-yuan SONG,Chao LIU,Wei WEI,Cheng-ji LYU. State recognition in bearing temperature of high-speed train based on machine learning algorithms [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(1): 53-62.
[2]	Hui ZHONG,Heng KANG,Ying-da LYU,Zhen-jian LI,Hong LI,Ruo-chuan OUYANG. Image manipulation localization algorithm based on channel attention convolutional neural networks [J]. Journal of Jilin University(Engineering and Technology Edition), 2021, 51(5): 1838-1844.
[3]	Hou⁃jie LI,Fa⁃sheng WANG,Jian⁃jun HE,Yu ZHOU,Wei LI,Yu⁃xuan DOU. Pseudo sample regularization Faster R⁃CNN for traffic sign detection [J]. Journal of Jilin University(Engineering and Technology Edition), 2021, 51(4): 1251-1260.
[4]	Qian-yi XU,Gui-he QIN,Ming-hui SUN,Cheng-xun MENG. Classification of drivers' head status based on improved ResNeSt [J]. Journal of Jilin University(Engineering and Technology Edition), 2021, 51(2): 704-711.
[5]	Xiang-jiu CHE,Hua-luo LIU,Qing-bin SHAO. Fabric defect recognition algorithm based onimproved Fast RCNN [J]. Journal of Jilin University(Engineering and Technology Edition), 2019, 49(6): 2038-2044.
[6]	CHEN Mian-shu, SU Yue, SANG Ai-jun, LI Pei-peng. Image classification methods based on space vector model [J]. 吉林大学学报(工学版), 2018, 48(3): 943-951.
[7]	CHEN Zai-qing, SHI Jun-sheng, BAI Feng-xiang. Automatic image classification based on fuzzy-rough set [J]. 吉林大学学报(工学版), 2013, 43(增刊1): 209-212.
[8]	WANG Ying, GUO Lei, LIANG Nan. Classification algorithm of hyperspectral images based on kernel entropy analysis [J]. , 2012, (06): 1597-1601.
[9]	LIU Ping-ping, ZHAO Hong-wei, GENG Qing-tian, DAI Jin-bo. Image classification method based on local feature and visual cortex recognition mechanism [J]. 吉林大学学报(工学版), 2011, 41(05): 1401-1406.
[10]	Cao Chun-hong Zhang Bin,Li Xiao-lin . Medical image classification technology based on fuzzy support vector machine [J]. 吉林大学学报(工学版), 2007, 37(03): 630-0633.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

Comments

Recommended 0

No Suggested Reading articles found!

方法	Accuracy	Specificity	Sensitivity
ResNet-1	0.9231	0.8633	0.9425
ResNet-2	0.9253	0.8641	0.9434
ResNet-3	0.9272	0.8648	0.9421
ResNet-1_2	0.9292	0.8709	0.9503
ResNet-1_3	0.9301	0.8734	0.9527
ResNet-1_2_3	0.9325	0.8721	0.9532

方法	Accuracy	Specificity	Sensitivity
ResNet-18	0.9325	0.8721	0.9532
ResNet-34	0.9405	0.8793	0.9576
ResNet-50	0.9418	0.8837	0.9583
ResNet-101	0.9421	0.8843	0.9590
ResNet-152	0.9425	0.8848	0.9595

方法	Accuracy	Precision	Specificity	Sensitivity	F1	MCC
VGG	0.9071	0.9365	0.8464	0.9321	0.9343	0.7758
SqueezeNet	0.9142	0.9324	0.8459	0.9445	0.9384	0.7970
GoogLeNet	0.9179	0.9528	0.8776	0.9332	0.9429	0.7978
MnasNet	0.9075	0.9495	0.8720	0.9211	0.9350	0.7754
DenseNet	0.9101	0.9457	0.8619	0.9286	0.9371	0.7797
MobileNetV3	0.9120	0.9221	0.8371	0.9488	0.9352	0.7986
本文	0.9425	0.9623	0.8848	0.9595	0.9619	0.8454

Validity classification of melting curve based on multi⁃scale fusion convolutional neural network

RICH HTML

PDF (PC)