重加权在多类别不平衡医学图像检测中的应用

doi:10.3778/j.issn.1002-8331.2011-0359

摘要/Abstract

摘要： 在医学图像检测中，由于数据集经常存在每类样本数目不均衡的情况，使数据集样本出现长尾分布的问题，严重影检测模型的性能。针对网络在训练多类别不均衡数据集中训练时出现的过拟合现象，采用重加权的方式改进原有损失函数，并用CLAHE算法对X光图像进行预处理，以突出图像的内部细节，选用ResNext50网络作为特征提取网络。以covid-chestxray数据集作为实验用数据集，通过实验评估了模型的准确度、精确率、召回率和F1值，证实了该方法的有效性。

关键词: 类别不平衡, 医学图像, 胸部X光图像, 重加权

Abstract: In medical image detection, the dataset often has a situation od class-imbalanced, which causes the long-tail distribution problem in the dataset samples and the performance of the detection model becomes seriously. This paper aims to solve the over-fitting problem of the network when training with multiple class-imbalance dataset, improving the original loss function by re-weight method, and preprocessing the X-ray image by CLAHE algorithm to highlight the internal details of the image. Finally, ResNext50 is selected as the feature extraction network. By using covid-chestxray dataset as the dataset, this paper evaluates the Accuracy, Precision, Recall rate and F1 value of the model. The effectiveness of this method is verified in experiments.

Key words: class-imbalance, medical image, chest X-ray image, re-weight

柴文光, 李嘉怡. 重加权在多类别不平衡医学图像检测中的应用[J]. 计算机工程与应用, 2022, 58(8): 237-242.

CHAI Wenguang, LI Jiayi. Application of Re-Weight Method in Multiple Class-Imbalance Medical Images Detection[J]. Computer Engineering and Applications, 2022, 58(8): 237-242.

参考文献

[1] 徐文涛，陈胜.低对比度胸片中管线检测方法[J].电子科技，2015，28（12）：84-87.
XU Wentao，CHEN Sheng.Tube detection in low-contrast chest[J].Electronic Science and Technology，2015，28（12）：84-87.
[2] 刘林，黄英，贺振华.基于可拓检测的医学图像分割[J].广东工业大学学报，2013，30（3）：18-22.
LIU Lin，HUANG Ying，HE Zhenhua.Segmentation of medical images based on extension detecting technology[J].Journal of Guangdong University of Technology，2013，30（3）：18-22.
[3] 陈建立.基于深度学习的眼底微动脉瘤检测与识别的研究[D].成都：电子科技大学，2018.
CHEN Jianli.Detection and recognition of microaneurysms in fundus image based on deep learning[D].Chengdu：University of Electronic Science and Technology of China，2018.
[4] 陈寿宏，柳馨雨，马峻，等.深度卷积神经网络胸片肺结节分类识别研究[J].计算机工程与应用，2018，54（24）：176-181.
CHEN Shouhong，LIU Xinyu，MA Jun，et al.Research of deep convolution neural network in classification of chest radiographs and pulmonary nodules[J].Computer Engineering and Applications，2018，54（24）：176-181.
[5] 张驰名.基于深度学习方法的胸部常见病变X线诊断研究[D].绵阳：西南科技大学，2020.
ZHANG Chiming.X-ray diagnosis of common chest lesions based on deep learning[D].Mianyang：Southwest University of Science and Technology，2020.
[6] 张凯，陈亚军，张俊.生成对抗网络在医学小样本数据中的应用[J].内江师范学院学报，2020，35（4）：57-60.
ZHANG Kai，CHEN Yajun，ZHANG Jun.Applications of generative adversarial nets in medical small sample data[J].Journal of Neijiang Normal University，2020，35（4）：57-60.
[7] 冯毅博，仇大伟，曹慧，等.基于深度可分离稠密网络的新型冠状病毒肺炎X线图像检测方法研究[J].生物医学工程学杂志，2020，37（4）：557-565.
FENG Yibo，QIU Dawei，CAO Hui，et al.Research on coronavirus disease 2019（COVID-19） detection method based on depthwise separable DenseNet in chest X-ray images[J].Journal of Biomedical Engineering，2020，37（4）：557-565.
[8] ZUIDERVELD K.Contrast limited adaptive histogram equali-
zation[J].Graphics Gems，1994：474-485.
[9] 张宏钊，吕启深，党晓婧，等.基于加权损失函数的多尺度对抗网络图像语义分割算法[J].计算机应用与软件，2020，37（1）：284-291.
ZHANG Hongzhao，LV Qishen，DANG Xiaojing，et al.Multi-scale adversarial network semantic segmentation algorithm based on weighted loss function[J].Computer Applications and Software，2020，37（1）：284-291.
[10] 钟映春，孙思语，吕帅，等.铁塔航拍图像中鸟巢的YOLOv3识别研究[J].广东工业大学学报，2020，37（3）：42-48.
ZHONG?Yingchun，SUN?Siyu，LYU?Shuai，et al.Recognition of bird’s nest on transmission tower in aerial image of high-volage power line by YOLOv3 algorithm[J].Journal of Guangdong University of Technology，2020，37（3）：42-48.
[11] LIN T Y，GOYAL P，GIRSHICK R，et al.Focal loss for dense object detection[J].IEEE Transactions on Pattern Analysis & Machine Intelligence，2017：2999-3007.
[12] CHEN C，SONG X，JIANG S.Focal loss for region proposal network[C]//Chinese Conference on Pattern Recognition & Computer Vision，2018.
[13] CUI Y，JIA M，LIN T Y，et al.Class-balanced loss based on effective number of samples[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition（CVPR），Long Beach，CA，USA，2019：9260-9269.
[14] XIE S，GIRSHICK R，DOLLAR P，et al.Aggregated residual transformations for deep neural networks[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition（CVPR），Honolulu，HI，2017：5987-5995.
[15] HE K，ZHANG X，REN S，et al.Deep residual learning for image recognition[C]//IEEE Conference on Computer Vision and Pattern Recognition（CVPR），2016：770-778.
[16] SZEGEDY C，LIU W，JIA Y，et al.Going deeper with convolutions[C]//IEEE Conference on Computer Vision and Pattern Recognition（CVPR），Boston，MA，2015：1-9.
[17] 赵丽萍，袁霄，祝承，等.面向图像分类的残差网络进展研究[J].计算机工程与应用，2020，56（20）：9-19.
ZHAO Liping，YUAN Xiao，ZHU Cheng，et al.Research on residual networks for image classification[J].Computer Engineering and Applications，2020，56（20）：9-19.
[18] WANG L，WONG A.Covid-net：a tailored deep convolutional neural network design for detection of covid-19 cases from chest radiography images[J].arXiv：2003.09871，
2020.
[19] 毛本清，金小梅.自适应直方图均衡化算法在图像增强处理的应用[J].河北北方学院学报（自然科学版），2010（5）：64-68.
MAO Benqing，JIN Xiaomei.Application of self-adaptive histogram equalization algorithm to image enhancement processing[J].Journal of Hebei North University（Natural Science Edition），2010（5）：64-68.
[20] CAO K，WEI C，GAIDON A，et al.Imbalanced datasets with label-distribution-aware margin loss[J].arXiv：1906.07413，
2019.
[21] WANG Jiaqi，ZHANG Wenwei，ZANG Yuhang.Seesaw loss for long-tailed instance segmentation[J].arXiv：2008.
10032，2020.