Research on Classification of Pulmonary Nodules by Three-Dimensional Multi-Scale Cross Fusion Network

doi:10.3778/j.issn.1002-8331.2102-0209

Abstract

Abstract: Accurate classification of benign and malignant pulmonary nodules on chest CT images is vital to the prevention and treatment of lung cancer. However, it is a challenging task due to the complex background of chest CT images and the unclear delineation of benign and malignant pulmonary nodules. A deep three-dimensional multi-scale cross fusion convolutional neural network is proposed to realize the accurate classification of benign and malignant pulmonary nodules. It uses dense connection structure to automatically extract multi-scale features of lung nodules. In order to reduce the loss of lung nodule-related information in the feature extraction process, a cross-fusion strategy is introduced for multi-scale features to obtain multi-scale feature groups, which enhances high and low-level semantic information expressing ability, and at the same time enhance the transmission and transfer of features in the network. This article uses the lung image federation data set for verification, the classification accuracy rate reaches 90.96%, and the AUC is 94.95%.

Key words: pulmonary nodule classification, dense connection network, multiscale cross fusion

摘要： 计算机断层扫描影像中良、恶性肺结节的准确分类对肺癌的预防和治疗至关重要。然而，由于计算机断层扫描影像中肺结节背景的复杂性，以及良、恶性肺结节判定之间存在的不确定性，使得良恶性肺结节的准确分类成为了一项极具挑战性的工作。提出了一种深度三维多尺度交叉融合卷积神经网络实现了良恶性肺结节的精确分类。使用密集连接结构自动提取肺结节多尺度特征，为了减少特征提取过程中肺结节相关信息的丢失，对多尺度特征引入了交叉融合策略得到多尺度特征组，增强了高、低层次语义信息的表达能力，同时增强特征在网络中的传递和转移。将提取的特征组分别连接至多个softmax分类器，模拟多位经验不同医生共同决策，实现了良、恶性肺结节的精确识别。使用肺图像联合会数据集进行验证，分类准确率达到了90.96%，AUC为94.95%。

关键词: 肺结节分类, 密集连接网络, 多尺度交叉融合

YANG Jianli, ZHU Dejiang, SHAO Jiajun, LIU Xiuling. Research on Classification of Pulmonary Nodules by Three-Dimensional Multi-Scale Cross Fusion Network[J]. Computer Engineering and Applications, 2022, 58(14): 121-125.

杨建利, 朱德江, 邵嘉俊, 刘秀玲. 三维多尺度交叉融合网络肺结节分类研究[J]. 计算机工程与应用, 2022, 58(14): 121-125.

References

[1] LV S，XUE J，WU C，et al.Identification of a panel of serum microRNAs as biomarkers for early detection of lung adenocarcinoma[J].Journal of Cancer，2017，8（1）：48.
[2] LI C，LV Y，SHAO C，et al.Tumor-derived exosomal lncRNA GAS5 as a biomarker for early-stage non-small-cell lung cancer diagnosis[J].Journal of Cellular Physiology，2019，234（11）：20721-20727.
[3] YANG X，ZHANG Q，ZHANG M，et al.Serum microRNA signature is capable of early diagnosis for non-small cell lung cancer[J].International Journal of Biological Sciences，2019，15（8）：1712-1722.
[4] DEGENS J，SANDERS K J C，DR JONG E E C，et al.The prognostic value of early onset，CT derived loss of muscle and adipose tissue during chemotherapy in metastatic non-small cell lung cancer[J].Lung Cancer，2019，133：130-135.
[5] TOGACAR M，ERGEN B，COMERT Z，et al.Detection of lung cancer on chest CT images using minimum redundancy maximum relevance feature selection method with convolutional neural networks[J].Biocybernetics and Biomedical Engineering，2020，40（1）：23-39.
[6] HELENO B，SIERSMA V，BRODERSEN J，et al.Estimation of overdiagnosis of lung cancer in low-dose computed tomography screening：a secondary analysis of the Danish lung cancer screening trial[J].JAMA Internal Medicine，2018，178（10）：1420-1422.
[7] XIE Y，ZHANG J，XIA Y，et al.Fusing texture，shape and deep model-learned information at decision level for automated classification of pulmonary nodules on chest CT[J].Information Fusion，2018，42：102-110.
[8] FILHO A O，SILVA A C，DE PAIVA A C，et al.Classification of patterns of benignity and malignancy based on CT using topology-based phylogenetic diversity index and convolutional neural network[J].Pattern Recognition，2018，81：200-212.
[9] 徐久强，洪丽萍，朱宏博，等.一种用于肺结节恶性度分类的生成对抗网络[J].东北大学学报（自然科学版），2018，39（11）：39-44.
XU J Q，HONG L P，ZHU H B，et al.Generative adversarial networks for the classification of lung nodules malignant[J].Journal of Northeastern University（Natural Science），2018，39（11）：39-44.
[10] 吴世洋，任劲松，张冉，等.基于卷积神经网络的肺结节良恶性分类[J].中国医学工程，2020，28（1）：1-3.
WU S Y，REN J S，ZHANG R，et al.Classification of benign and malignant pulmonary nodules based on convolutional neural networks[J].China Medical Engineering，2020，28（1）：1-3.
[11] SHEN W，ZHOU M，YANG F，et al.Multi-scale convolutional neural networks for lung nodule classification[M].[S.l.]：Springer International Publishing，2015：588-599.
[12] SHEN W，ZHOU M，YANG F，et al.Multi-crop convolutional neural networks for lung nodule malignancy suspiciousness classification[J].Pattern Recognition，2017，61：663-673.
[13] LEI Y，TIAN Y，SHAN H，et al.Shape and margin-aware lung nodule classification in low-dose CT images via soft activation mapping[J].Medical Image Analysis，2019，60：101628.
[14] XIE Y，XIA Y，ZHANG J，et al.Knowledge-based collaborative deep learning for benign-malignant pulmonary nodule classification on chest CT[J].IEEE Transactions on Medical Imaging，2019，38（4）：991-1004.
[15] SHEN S，HAN S X，ABERLE D R，et al.An interpretable deep hierarchical semantic convolutional neural network for pulmonary nodule malignancy classification[J].Expert Systems With Applications，2019，15（128）：84-95.
[16] TAFTI A P，BASHIRI F S，LAROSE E，et al.Diagnostic classification of lung CT images using deep 3d multi-scale convolutional neural network[C]//2018 IEEE International Conference on Healthcare Informatics（ICHI）.New York：IEEE，2018：412-414.
[17] BONAVITA I，RAFAEL-PALOU X，CERESA M，et al.Integration of convolutional neural networks for pulmonary nodule malignancy assessment in a lung cancer classification pipeline[J].Computer Methods and Programs in Biomedicine，2020，185：105172.
[18] KANG G，LIU K，HOU B，et al.3D multi-view convolutional neural networks for pulmonary nodule classification[J].PloS One，2017，12（11）：e0188290.
[19] JIANG H，GAO F，XU X，et al.Attentive and ensemble 3D dual path networks for pulmonary nodules classification[J].Neurocomputing，2019，398：103.
[20] HUSSEIN S，CAO K，SONG Q，et al.Risk stratification of pulmonary nodules using 3D CNN-based multi-task learning[C]//International Conference on Information Processing in Medical Imaging，2017：249-260.
[21] ARMATO III S G，MCLENNAN G，BIDAUT L，et al.The lung image database consortium（LIDC） and image database resource initiative（IDRI）：a completed reference database of pulmonary nodules on CT scans[J].Medical Physics，2011，38（2）：915-931.
[22] HUANG G，LIU Z，LAURENS V D M，et al.Densely connected convolutional networks[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition（CVPR），2017，243：2261-2269.
[23] HE K，ZHANG X，REN S，et al.Identity mappings in deep residual networks[J].Lecture Notes in Computer Science，2016，9908：630-645.
[24] The Cancer Imaging Archive.Lung image database consortium-reader annotation and markup-annotation and markup issues/comments[EB/OL].[2021-01-05].https：//wiki.cancerimagingarchive.net/display/Public/LIDC-IDRI.