基于ConA-FPN的肝脏肿瘤检测算法

doi:10.3778/j.issn.1002-8331.2107-0235

摘要/Abstract

摘要： 深度学习方法在病灶检测任务中被广泛应用，但因肝脏肿瘤较小和样本较少的问题，导致无法达到辅助诊断的准确率要求。针对以上问题，提出基于ConA-FPN的肝脏肿瘤检测算法，具体过程为：使用融合ResNet和注意力机制的特征金字塔替换Faster R-CNN中的特征提取网络；使用融合特征解决特征金字塔中的高层模块通道信息损失问题，通过添加CAG注意力机制解决了特征融合带来的特征混叠问题，增强模型对小肿瘤的检测能力；使用迁移学习和数据增强提升模型在小数据集上的检测能力和泛化能力。实验结果表明，ConA-FPN在LITS2017和3D-IRCADB上的平均精度达到87.43%，明显优于主流检测模型。

关键词: ConA-FPN, 肝脏肿瘤, 特征融合, 小目标, 小数据集

Abstract: Deep learning methods are widely used in lesion detection tasks, but the problems of small liver tumors and small samples lead to the inability to achieve the accuracy requirements for assisted diagnosis. To address the above problems, this paper proposes a liver tumor detection algorithm based on ConA-FPN, and the specific process is as follows：first, the feature extraction network in Faster R-CNN is replaced by using a feature pyramid fused with ResNet and attention mechanism; then, the fused features are used to solve the problem of information loss of higher-level module channels in the feature pyramid, the CAG attention mechanism is added to solve the problem that feature fusion brings the feature conflation, and enhances the detection ability of the model for small tumors; finally, transfer learning and data augmentation are used to enhance the detection ability and generalization ability of the model on small data sets. The experimental results show that the average accuracy of ConA-FPN on LITS2017 and 3D-IRCADB reaches 87.43%, which is significantly better than the mainstream detection models.

Key words: ConA-FPN, liver tumor, feature fusion, small targets, small datasets

马金林, 毛凯绩, 马自萍, 邓媛媛, 欧阳轲, 陈勇. 基于ConA-FPN的肝脏肿瘤检测算法[J]. 计算机工程与应用, 2023, 59(2): 161-169.

MA Jinlin, MAO Kaiji, MA Ziping, DENG Yuanyuan, OUYANG Ke, CHEN Yong. ConA-FPN Based Algorithm for Liver Tumor Detection[J]. Computer Engineering and Applications, 2023, 59(2): 161-169.

参考文献

[1] 石羽，曹晓光，张弘，等.肺结节自动检测算法研究[J].仪器仪表学报，2006（S3）：2265-2267.
SHI Y，CAO X G，ZHANG H，et al.Study of automatic detection algorithm for lung nodules[J].Chinese Journal of Scientific Instrument，2006（S3）：2265-2267.
[2] TODOROKI Y，IWAMOTO Y，LIN L，et al.Automatic detection of focal liver lesions in multi-phase CT images using a multi-channel & multi-scale CNN[C]//2019 41st Annual International Conference of the IEEE Engineering in Medicine & Biology Society，2019.
[3] KAUSAR T，WANG M J，ASHRAF M A，et al.SmallMitosis：small size mitotic cells detection in breast histopathology images[J].IEEE Access，2020，9：905-922.
[4] LIN T Y，DOLLáR P，GIRSHICK R，et al.Feature pyramid networks for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2017：2117-2125.
[5] ZHANG H，CHEN Y，SONG Y，et al.Automatic kidney lesion detection for CT images using morphological cascade convolutional neural networks[J].IEEE Access，2019，7：83001-83011.
[6] VESAL S，PATIL S M，RAVIKUMAR N，et al.A multi-task framework for skin lesion detection and segmentation[C]//Workshop on Skin Image Analysis，2018.
[7] 姚红革，沈新霞，李宇，等.多模态融合的深度学习脑肿瘤检测方法[J].光子学报，2019，48（7）：165-176.
YAO H G，SHEN X X，LI Y，et al.Multi-modal fusion brain tumor detection method based on deep learning[J].Acta Photonica Sinica，2019，48（7）：165-176.
[8] LIANG D，LIN L，CHEN X，et al.Multi-stream scale-insensitive convolutional and recurrent neural networks for liver tumor detection in dynamic CT images[C]//2019 IEEE International Conference on Image Processing（ICIP），2019.
[9] ZHANG Z，ZHANG X，LIN X，et al.Ultrasonic diagnosis of breast nodules using modified Faster R-CNN[J].Ultrasonic Imaging，2019，41（6）：353-367.
[10] REN S，HE K，GIRSHICK R，et al.Faster R-CNN：towards real-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2016，39（6）：1137-1149.
[11] CAO C，WANG B，ZHANG W，et al.An improved faster R-CNN for small object detection[J].IEEE Access，2019，7：106838-106846.
[12] MO N，YAN L.Improved Faster RCNN based on feature amplification and oversampling data augmentation for oriented vehicle detection in aerial images[J].Remote Sensing，2020，12（16）：2558.
[13] CHEN S，LI Z，TANG Z.Relation R-CNN：a graph based relation-aware network for object detection[J].IEEE Signal Processing Letters，2020，27：1680-1684.
[14] XIA C，NIU K，HE Z，et al.SVM-based bone tumor detection by using the texture features of X-Ray image[C]//2018 International Conference on Network Infrastructure and Digital Content（IC-NIDC），2018：130-134.
[15] MESSAY T，HARDIE R C，TUINSTRA T R.Segmentation of pulmonary nodules in computed tomography using a regression neural network approach and its application to the lung image database consortium and image database resource initiative dataset[J].Medical Image Analysis，2015，22（1）：48-62.
[16] 孙雨琛，刘宇红，张达峰，等.基于深度学习的糖尿病视网膜病变诊断方法[J].激光与光电子学进展，2020，57（24）：359-366.
SUN Y C，LIU Y H，ZHANG D F，et al.Diagnosis method of diabetic retinopathy based on deep learning[J].Laser & Optoelectronics Progress，2020，57（24）：359-366.
[17] 卢逸君，滕少华.小样本数据生成及其在异常检测中的应用[J].江西师范大学学报（自然科学版），2020，44（4）：385-393.
LU Y J，TENG S H.The generation of minority sample data and its application in abnormal detection[J].Journal of Jiangxi Normal University（Natural Science Edition），2020，44（4）：385-393.
[18] RUSSAKOVSKY O，DENG J，SU H，et al.ImageNet large scale visual recognition challenge[J].International Journal of Computer Vision，2015，115（3）：211-252.
[19] HE K，ZHANG X，REN S，et al.Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2016：770-778.
[20] HU J，SHEN L，SUN G.Squeeze-and-excitation networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2018：7132-7141.
[21] GIRSHICK R.Fast R-CNN[C]//Proceedings of the IEEE International Conference on Computer Vision，2015：1440-1448.
[22] 杨淑莹，邓东升，郑清春.改进Faster R-CNN模型的CT图磨玻璃密度影目标检测[J].中国图象图形学报，2021，26（9）：2171-2180.
YANG S Y，DENG D S，ZHENG Q C.Ground-glass opacity target detection in CT scans based on improved Faster R-CNN model[J].Journal of Image and Graphics，2021，26（9）：2171-2180.
[23] LUO Y，CAO X，ZHANG J，et al.CE-FPN：enhancing channel information for object detection[J].arXiv：2103.10643，2021.
[24] BILIC P，CHRIST P F，VORONTSOV E，et al.The liver tumor segmentation benchmark（lits）[J].arXiv：1901.04056，2019.
[25] Ircad France.3Dircadb[EB/OL].[2021-05-10].https：//www.icad.fr/research/3dircadb.
[26] EVERINGHAM M，VAN GOOL L，WILLIAMS C K I，et al.The pascal visual object classes（voc） challenge[J].International Journal of Computer Vision，2010，88（2）：303-338.
[27] HOWARD A G，ZHU M，CHEN B，et al.Mobilenets：efficient convolutional neural networks for mobile vision applications[J].arXiv：1704.04861，2017.
[28] HOU Q，ZHOU D，FENG J.Coordinate attention for efficient mobile network design[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition，2021：13713-13722.

[29] DATTA S K，SHAIKH M A，SRIHARI S N，et al.Soft-attention improves skin cancer classification performance[J].arXiv：2105.03358，2021.

[30] BOCHKOVSKIY A，WANG C Y，LIAO H Y M.Yolov4：optimal speed and accuracy of object detection[J].arXiv：2004.10934，2020.