浏览全部资源
扫码关注微信
纸质出版日期： 2021-07-16 ，
录用日期： 2021-02-15
移动端阅览
顾婷菲, 郝鹏翼, 白琮, 柳宁. 结合多通道注意力的糖尿病性视网膜病变分级[J]. 中国图象图形学报, 2021,26(7):1726-1736.
Tingfei Gu, Pengyi Hao, Cong Bai, Ning Liu. Diabetic retinopathy grading based on multi-channel attention[J]. Journal of Image and Graphics, 2021,26(7):1726-1736.
目的
2
糖尿病性视网膜病变（diabetic retinopathy，DR）是一种常见的致盲性视网膜疾病，需要患者在早期就能够被诊断并接受治疗，否则将会造成永久性的视力丧失。能否检测到视网膜图像中的微小病变如微血管瘤，是糖尿病性视网膜病变分级的关键。然而这些病变过于细小导致使用一般方法难以正确地辨别。为了解决这一问题，本文提出了一种基于多通道注意力选择机制的细粒度分级方法（fine-grained grading method based on multi-channel attention selection，FGMAS）用于糖尿病性视网膜病变的分级。
方法
2
该方法结合了细粒度分类方法和多通道注意力选择机制，通过获取局部特征提升分级的准确度。此外考虑到每一层通道特征信息量与分类置信度的关系，本文引入了排序损失以优化每一层通道的信息量，用于获取更加具有信息量的局部区域。
结果
2
使用两个公开的视网膜数据集（Kaggle和Messidor）来评估提出的细粒度分级方法和多通道注意力选择机制的有效性。实验结果表明：FGMAS在Kaggle数据集上进行的五级分类任务中相较于现有方法，在平均准确度（average of classification accuracy，ACA）上取得了3.4%10.4%的提升。尤其是对于病变点最小的1级病变，准确率提升了11%18.9%。此外，本文使用FGMAS在Messidor数据集上进行二分类任务。在推荐转诊/不推荐转诊分类上FGMAS得到的准确度（accuracy，Acc）为0.912，比现有方法提升了0.1%1.9%，同时AUC（area under the curve）为0.962，比现有方法提升了0.5%9.9%；在正常/不正常分类上FGMAS得到的准确度为0.909，比现有方法提升了2.9%8.8%，AUC为0.950，比现有方法提升了0.4%8.9%。实验结果表明，本文方法在五分类和二分类上均优于现有方法。
结论
2
本文所提细粒度分级模型，综合了细粒度提取局部区域的思路以及多通道注意力选择机制，可以获得较为准确的分级结果。
Objective
2
Diabetic retinopathy (DR) is a common blinding retinal disease that cannot be cured in the later stage
and requires patients to be diagnosed and treated at an early stage; otherwise
it causes permanent vision loss. The prevalence of diabetic retinopathy is extremely high in China
and is in the stage of rapid growth. At present
China has become the country that has the largest number of patients with diabetic retinopathy. Diagnosis of DR is usually performed by analyzing fundus medical images. Detection of microscopic lesions such as microaneurysms in retinal images is necessary in grading diabetic retinopathy with neural networks. This condition requires the attention mechanism to simulate the focus of the human eyes and focus on the local area with information. However
most of the present methods only consider the attention in the spatial domain and ignore the information in the channel attention
which cause difficulty in distinguishing the small lesions. To solve this problem
a fine-grained grading method based on multi-channel attention selection (FGMAS) mechanism is proposed for the grading of diabetic retinopathy in this paper.
Method
2
This method combines fine-grained classification with a multi-channel attention selection mechanism. First
the structure of fine-grained classification is adopted to improve the recognition accuracy of small differences between categories by obtaining local regional features. Then
the characteristics of different feature layers in the channel domain with different information content are used to select high-information channels. The model establishes the relationship between information content and classification confidence
and obtains the lesion area that is conducive to classification results. Finally
the local and global features are combined to improve the accuracy of classification. In addition
considering the relationship between the channel characteristic information of each layer and the classification confidence
this study also introduces Rank_loss to optimize the channel information of each layer. The loss function enables the regions with higher classification confidence to have higher information content and obtain better classification results.
Result
2
Two open retinal datasets (Kaggle and Messidor) are used to evaluate the effectiveness of the proposed fine-grained grading method and multi-channel attention selection mechanism. The experimental results show that FGMAS performs a five-level classification on the Kaggle dataset with better results than the existing method
with an average accuracy of 0.577
which is 3.4%10.4% higher than the accuracy of other methods. The first category shows small lesion points
which are difficult to distinguish in other methods. However
the accuracy rate of 0.301 can be obtained through FGMAS proposed in this paper
which is better than other methods with the improvement of 11%18.9%
such as 0.190 of VGGNet with Extra Kernel/LGI (VNXK/LGI). Meanwhile
FGMAS is used in the Messidor dataset to perform a dichotomous task
including recommended reference/non-reference and normal/abnormal classification. In the reference/non-reference classification task
the experimental results are 0.912 of accuracy and 0.962 of AUC(area under the curve)
which is superior to the existing methods by 0.1%1.9% and 0.5%9.9%
respectively. In the normal/abnormal classification task
the experimental results are 0.909 of accuracy and 0.950 of AUC
which are improved by 2.9%8.8% and 0.4%8.9% respectively
compared with existing methods. In addition
parameter experiments are set up in this study
and the function of each parameter and optimal parameter selection result are analyzed in detail.
Conclusion
2
This study proposes a fine-grained grading model that combines the fine-grained classification and multi-channel attention models. In addition
Rank_loss combines the ranking result and information of every layer. It is used to obtain the local feature area
which is beneficial to the classification result. According to the experimental results
the model can obtain good results in five-classification and two-classification tasks.
糖尿病性视网膜病变(DR)病变分级细粒度分级深度学习多通道注意力选择机制局部特征提取
diabetic retinopathy(DR)lesion gradingfine-grained gradingdeep learningmulti-channel attention selectionlocal feature extraction
Antal B and Hajdu A. 2012. An ensemble-based system for microaneurysm detection and diabetic retinopathy grading. IEEE Transactions on Biomedical Engineering, 59(6): 1720-1726[DOI: 10.1109/TBME.2012.2193126]
Bravo M A and Arbeláez Pablo A. 2017. Automatic diabetic retinopathy classification//13th International Symposium on Medical Information Processing and Analysis. San Andres Island, Colombia: SPIE: 105721E. 1-10[DOI: 10.1117/12.2285939http://dx.doi.org/10.1117/12.2285939]
Burges C, Shaked T, Renshaw E, Lazier A, Deeds M, Hamilton N and Hullender G. 2005. Learning to rank using gradient descent//Proceedings of the 22nd International Conference on Machine Learning. New York, USA: ACM: 89-96[DOI: 10.1145/1102351.1102363http://dx.doi.org/10.1145/1102351.1102363]
Chen L, Zhang H W, Xiao J, Nie L Q, Shao J, Liu W and Chua T S. 2017. SCA-CNN: spatial and channel-wise attention in convolutional networks for image captioning//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA: IEEE: 6298-6306[DOI: 10.1109/CVPR.2017.667http://dx.doi.org/10.1109/CVPR.2017.667]
Chen Y W, Wu T Y, Wong W H and Lee C Y. 2018. Diabetic retinopathy detection based on deep convolutional neural networks//Proceedings of 2018 IEEE International Conference on Acoustics, Speech and Signal Processing. Calgary, Canada: IEEE: 1030-1034[DOI: 10.1109/ICASSP.2018.8461427http://dx.doi.org/10.1109/ICASSP.2018.8461427]
Decencière E, Zhang X W, Cazuguel G, Lay B, Cochener B, Trone C, Gain P, Ordonez R, Massin P, Erginay A, Charton B and Klein J C. 2014. Feedback on a publicly distributed image database: the messidor database. Image Analysis and Stereology, 33(3): 231-234[DOI: 10.5566/ias.1155]
EyePACS. 2015. Diabetic-Retinopathy-Detection[EB/OL]. [2021-04-15].https://www.kaggle.com/c/diabetic-retinopathy-detectionhttps://www.kaggle.com/c/diabetic-retinopathy-detection
Fundus ophthalmology Group of Chinese Medical Association. 2014. Chinese guidelines for clinical diagnosis and treatment of diabetic retinopathy. Chinese Journal of Ophthalmology, 50(11): 851-865
中华医学会眼科学会眼底病学组. 2014. 我国糖尿病视网膜病变临床诊疗指南. 中华眼科杂志, 50(11): 851-865)[DOI: 10.3760/cma.j.issn.0412-4081.2014.11.014]
Huang G, Liu Z, van der Maaten L and Weinberger K Q. 2017. Densely connected convolutional networks//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA: IEEE: 2261-2269[DOI: 10.1109/CVPR.2017.243http://dx.doi.org/10.1109/CVPR.2017.243]
Pires R, Avila S, Jelinek H F, Wainer J, Valle E and Rocha A. 2017. Beyond lesion-based diabetic retinopathy: a direct approach for referral. IEEE Journal of Biomedical and Health Informatics, 21(1): 193-200[DOI: 10.1109/JBHI.2015.2498104]
Rocha A, Carvalho T, Jelinek H F, Goldenstein S and Wainer J. 2012. Points of interest and visual dictionaries for automatic retinal lesion detection. IEEE Transactions on Biomedical Engineering, 59(8): 2244-2253[DOI: 10.1109/TBME.2012.2201717]
Russakovsky O, Deng J, Su H, Krause J, Satheesh S, Ma S, Huang Z H, Karpathy A, Khosla A, Bernstein M, Berg A C and Li F F. 2015. ImageNet large scale visual recognition challenge. International Journal of Computer Vision, 115(3): 211-252[DOI: 10.1007/s11263-015-0816-y]
Ratchasima N. 2019. APTOS: Eye Preprocessing in Diabetic Retinopathy[CP/OL]. [2021-04-15].https://www.kaggle.com/ratthachat/aptos-eye-preprocessing-in-diabetic-retinopathyhttps://www.kaggle.com/ratthachat/aptos-eye-preprocessing-in-diabetic-retinopathy
Salami M J E, Khorshidtalab A, Baali A and Aibinu A M. 2014. Classification of retinal images based on statistical moments and principal component analysis//Proceedings of 2014 International Conference on Computer and Communication Engineering. Kuala Lumpur, Malaysia: IEEE: 92-95[DOI: 10.1109/ICCCE.2014.37http://dx.doi.org/10.1109/ICCCE.2014.37]
Sánchez C Niemeijer M, Dumitrescu A V, Suttorp-Schulten M S A, Abràmoff M D and van Ginneken B. 2011. Evaluation of a computer-aided diagnosis system for diabetic retinopathy screening on public data. Investigative Ophthalmology and Visual Science, 52(7): 4866-4871[DOI: 10.1167/iovs.10-6633]
Seoud L, Hurtut T, Chelbi J, Cheriet F and Langlois J M P. 2016. Red lesion detection using dynamic shape features for diabetic retinopathy screening. IEEE Transactions on Medical Imaging, 35(4): 1116-1126[DOI: 10.1109/TMI.2015.2509785]
Vo H H and Verma A. 2016. New deep neural nets for fine-grained diabetic retinopathy recognition on hybrid color space//2016 IEEE International Symposium on Multimedia. San Jose, USA: IEEE: 209-215[DOI: 10.1109/ism.2016.0049http://dx.doi.org/10.1109/ism.2016.0049]
Wang Z, Yin Y X, Shi J P, Fang W, Li H S and Wang X G. 2017. Zoom-in-Net: deep mining lesions for diabetic retinopathy detection//Proceedings of Medical Image Computing and Computer Assisted Intervention. Quebec City, Canada: Springer: 267-275[DOI: 10.1007/978-3-319-66179-7_31http://dx.doi.org/10.1007/978-3-319-66179-7_3]
Woo S, Park J, Lee J Y and Kweon I S. 2018. CBAM: convolutional block attention module//Proceedings of European Conference on Computer Vision. Munich, Germany: Springer: 3-19[DOI: 10.1007/978-3-030-01234-2_1http://dx.doi.org/10.1007/978-3-030-01234-2_1]
Zhao Z Y, Zhang K R, Hao X J, Tian J, Chua M C H, Chen L and Xu X. 2019. BiRA-Net: bilinear attention net for diabetic retinopathy grading//Proceedings of 2019 IEEE International Conference on Image Processing. Taipei, China: IEEE: 1385-1389[DOI: 10.1109/ICIP.2019.8803074http://dx.doi.org/10.1109/ICIP.2019.8803074]
Zhou K, Gu Z W, Liu W, Luo W X, Cheng J, Gao S H and Liu J. 2018. Multi-cell multi-task convolutional neural networks for diabetic retinopathy grading//Proceedings of the 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Honolulu, USA: IEEE: 2724-2727[DOI: 10.1109/EMBC.2018.8512828http://dx.doi.org/10.1109/EMBC.2018.8512828]
相关作者
相关机构
京公网安备11010802024621