食道病灶检测的多尺度细节增强金字塔网络

doi:10.3778/j.issn.1002-8331.2209-0162

摘要/Abstract

摘要： 针对食道卢戈染色内窥镜（Lugol’s chromoendoscopy, LCE）检查图像中病变类间相似度高和类内尺度变化大等问题，提出了一种以Sparse R-CNN作为基底网络并搭载多尺度细节增强金字塔网络（multi-scale detail enhancement pyramid network, MDEPN）结构的多类食道病灶检测方法。为了改善Sparse R-CNN中的特征金字塔网络（feature pyramid network, FPN）结构在对多尺度特征进行融合时存在的信息丢失和语义差异问题，MDEPN结构首先使用Gabor调制卷积模块对不同尺度特征进行增强，利用Gabor对方向和尺度的强关注度提高了特征图中纹理信息的表达能力；其次使用方向通道池化模块提取不同尺度特征的局部特征方向相似性和局部与全局特征的相关性，减轻了不同尺度特征融合时的语义差异。在自建的多类食道LCE病灶数据集上进行测试，结果表明该方法的mAP0.50检测精度达到了65.0%，相同条件下，比基准模型Sparse R-CNN提升了2.4个百分点，并超过了对比的其余主流检测方法。所设计的MDEPN模块可作为一种独立的结构融入其他检测模型中以提升性能，具有一定的通用性。

关键词: 食道病变检测, 金字塔网络, Gabor调制卷积, 方向通道池化

Abstract: Aiming at problems such as high interclass similarity and large intraclass scale changes in Lugol??s chromoendoscopy (LCE) images, this paper proposes a method for the detection of multiple esophageal diseases, which is based on Sparse R-CNN and equipped with multi-scale detail enhancement pyramid network (MDEPN) structure. In order to solve the problems of information loss and semantic difference in feature pyramid network (FPN) structure of Sparse R-CNN, the MDEPN structure firstly uses Gabor modulated convolution module to enhance the features of different scales, and uses Gabor??s strong attention to direction and scale to improve the expression ability of texture information in the feature map. Secondly, the directional channel pooling module is used to extract the local directional similarity and the correlation between local and global features of different scale features, so as to reduce the semantic differences in the fusion of different scale features. After testing on a self-built dataset of multiple esophageal LCE lesions, the accuracy of mAP0.50 is 65.0%, 2.4 percentage points higher than that of the benchmark model Sparse R-CNN, and higher than other major detection methods. In addition, the designed MDEPN module can be integrated into other detection models as an independent structure to improve performance, and has certain versatility.

Key words: detection of esophageal lesions, pyramid network, Gabor modulated convolution, directional channel pooling

李驰, 周颖玥, 姚韩敏, 李小霞, 秦佳敏, 庄鸣, 文黎明. 食道病灶检测的多尺度细节增强金字塔网络[J]. 计算机工程与应用, 2024, 60(4): 229-236.

LI Chi, ZHOU Yingyue, YAO Hanmin, LI Xiaoxia, QIN Jiamin, ZHUANG Ming, WEN Liming. Multi-Scale Detail Enhanced Pyramid Network for Esophageal Lesion Detection[J]. Computer Engineering and Applications, 2024, 60(4): 229-236.

参考文献

[1] FAN J, LIU Z, Mao X, et al. Global trends in the incidence and mortality of esophageal cancer from 1990 to 2017[J]. Cancer Medicine, 2020, 9(18): 6875-6887.
[2] UHLENHOPP D J, THEN E O, SUNKARA T, et al. Epidemiology of esophageal cancer: update in global trends, etiology and risk factors[J]. Clinical Journal of Gastroenterology, 2020, 13(6): 1010-1021.
[3] FITZMAURICE C. Global, regional, and national cancer incidence, mortality, years of life lost, years lived with disability, and disability-adjusted life-years for 29 cancer groups, 2006 to 2016: a systematic analysis for the global burden of disease study[J]. Journal of Clinical Oncology, 2018, 36: 1568.
[4] ODA I, SHIMIZU Y, YOSHIO T, et al. Long-term outcome of endoscopic resection for intramucosal esophageal squamous cell cancer: a secondary analysis of the Japan esophageal cohort study[J]. Endoscopy, 2020, 52(11): 967-975.
[5] KAREEM O S, ABDULAZEEZ A M, ZEEBAREE D Q. Skin lesions classification using deep learning techniques: review[J]. Asian Journal of Research in Computer Science,2021, 9(1): 1-22.
[6] CHEN L B, WAN J J, CHEN T Y, et al. A self-attention based Faster R-CNN for polyp detection from colonoscopy images[J]. Biomedical Signal Processing and Control, 2021, 70: 103019.
[7] LU S Y, WANG S H, ZHANG Y D, SAFNet: a deep spatial attention network with classifier fusion for breast cancer detection[J]. Computers in Biology and Medicine, 2022, 148: 105812.
[8] GHATWARY N, ZOLGHARNI M, YE X. Early esophageal adenocarcinoma detection using deep learning methods[J]. International Journal of Computer Assisted Radiology and Surgery, 2019, 14(4): 611-621.
[9] GHATWARY N, YE X, ZOLGHARNI M. Esophageal abnormality detection using DenseNet based faster R-CNN with Gabor features[J]. IEEE Access, 2019, 7: 84374-84385.
[10] HORIE Y, YOSHIO T, AOYAMA K, et al. Diagnostic outcomes of esophageal cancer by artificial intelligence using convolutional neural networks[J]. Gastrointestinal Endoscopy, 2019, 89(1): 25-32.
[11] 王士旭, 柯岩, 刘雨蒙, 等. 内镜下早期食管癌及癌前病变识别人工智能YOLOv5l模型的建立及临床验证[J]. 中华肿瘤杂志, 2022, 44(5): 395-401.
WANG S X, KE Y, LIU Y M, et al. Establishment and clinical validation of an artificial intelligence YOLOv51 model for the detection of precancerous lesions and superficial esophageal cancer in endoscopic procedure[J]. Chinese Journal of Cancer, 2021, 44(5): 395-401.
[12] WU Z, GE R, WEN M, et al. ELNet: automatic classification and segmentation for esophageal lesions using convolutional neural network[J]. Medical Image Analysis, 2021, 67: 101838.
[13] SUN P, ZHANG R, JIANG Y, et al. Sparse R-CNN: end-to-end object detection with learnable proposals[C]//Proceedings of the 2021 IEEE Conference on Computer Vision and Pattern Recognition, 2021: 14454-14463.
[14] CARION N, MASSA F, SYNNAEVE G, et al. End-to-end object detection with transformers[C]//Proceedings of the 16th European Conference on Computer Vision, 2020: 213-229.
[15] LIN T-Y, DOLLAR P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition, 2017: 936-944.
[16] CAI Z W, NUNO V. Cascade R-CNN: delving into high quality object detection[C]//Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition, 2018: 6154-6162.
[17] LUAN S, CHEN C, ZHANG B, et al. Gabor convolutional networks[J]. IEEE Transactions on Image Processing, 2018, 27(9): 4357-4366.
[18] SARWAR S S, PANDA P, ROY K. Gabor filter assisted energy efficient fast learning convolutional neural networks[C]//Proceedings of the 2017 IEEE ACM International Symposium on Low Power Electronics and Design, 2017: 1-6.
[19] REDMON J, FARHADI A. YOLOv3: an incremental improvement[J]. arXiv:1804.02767, 2018.
[20] LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot multibox detector[C]//Proceedings of the 14th European Conference on Computer Vision. Cham: Springer, 2016: 21-37.
[21] LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(2): 318-327.
[22] CHENG J F, YU J Z, YU G, et al. TOOD: task-aligned one-stage object detection[C]//Proceedings of the 2021 International Conference on Computer Vision, 2021: 3490-3499.
[23] 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, 2017, 39(6): 1137-1149.