当前位置: X-MOL 学术Int. J. Imaging Syst. Technol. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Enhancing breast pectoral muscle segmentation performance by using skip connections in fully convolutional network
International Journal of Imaging Systems and Technology ( IF 3.3 ) Pub Date : 2020-02-27 , DOI: 10.1002/ima.22410
Muhammad Junaid Ali 1, 2 , Basit Raza 1, 2 , Ahmad Raza Shahid 1, 2 , Fahad Mahmood 1, 2 , Muhammad Adil Yousuf 1, 2 , Amir Hanif Dar 1, 2 , Uzair Iqbal 1, 2
Affiliation  

The precise detection and segmentation of pectoral muscle areas in mediolateral oblique (MLO) views is an essential step in the development of a computer‐aided diagnosis system to access breast malignant lesions or parenchyma. The goal of this article is to develop a robust and fully automatic algorithm for pectoral muscle segmentation from mammography images. This paper presents an image enhancement approach that improves the quality of mammogram scans and a convolutional neural network‐based fully convolutional network architecture enhanced with residual connections for automatic segmentation of the pectoral muscle from the MLO views of a digital mammogram. For this purpose, the model is tested and trained on three different mammogram datasets named MIAS, INBREAST, and DDSM. The ground truth labels of the pectoral muscle were identified under the supervision of experienced radiologists. For training and testing, 10‐fold cross‐validation was used. The proposed model was compared with baseline U‐Net‐based architecture. Finally, we used a postprocessing step to find the actual boundary of the pectoral muscle. Our presented architecture generated a mean Intersection over Union (IoU) of 97%, dice similarity coefficient (DSC) of 96% and 98% accuracy on testing data. The proposed architecture for pectoral muscle segmentation from the MLO views of mammogram images with high accuracy and dice score can be quickly merged with the breast tumor segmentation problem.

中文翻译:

通过在全卷积网络中使用跳过连接来提高乳房胸肌分割性能

在内侧斜位 (MLO) 视图中精确检测和分割胸肌区域是开发计算机辅助诊断系统以访问乳腺恶性病变或实质的重要步骤。本文的目标是开发一种强大且全自动的算法,用于从乳房 X 光检查图像中分割胸肌。本文提出了一种提高乳房 X 线照片扫描质量的图像增强方法,以及一种基于卷积神经网络的全卷积网络架构,增强了残差连接,用于从数字乳房 X 线照片的 MLO 视图中自动分割胸肌。为此,该模型在名为 MIAS、INBREAST 和 DDSM 的三个不同乳房 X 光照片数据集上进行了测试和训练。在经验丰富的放射科医生的监督下,确定了胸肌的真实标签。对于训练和测试,使用了 10 折交叉验证。将所提出的模型与基于 U-Net 的基线架构进行了比较。最后,我们使用后处理步骤来找到胸肌的实际边界。我们提出的架构在测试数据上产生了 97% 的平均交集 (IoU)、96% 的骰子相似系数 (DSC) 和 98% 的准确率。从具有高精度和骰子分数的乳房 X 光照片图像的 MLO 视图中提出的胸肌分割架构可以快速与乳腺肿瘤分割问题合并。将所提出的模型与基于 U-Net 的基线架构进行了比较。最后,我们使用后处理步骤来找到胸肌的实际边界。我们提出的架构在测试数据上产生了 97% 的平均交集 (IoU)、96% 的骰子相似系数 (DSC) 和 98% 的准确率。从具有高精度和骰子分数的乳房 X 光照片图像的 MLO 视图中提出的胸肌分割架构可以快速与乳腺肿瘤分割问题合并。将所提出的模型与基于 U-Net 的基线架构进行了比较。最后,我们使用后处理步骤来找到胸肌的实际边界。我们提出的架构在测试数据上产生了 97% 的平均交集 (IoU)、96% 的骰子相似系数 (DSC) 和 98% 的准确率。从具有高精度和骰子分数的乳房 X 光照片图像的 MLO 视图中提出的胸肌分割架构可以快速与乳腺肿瘤分割问题合并。
更新日期:2020-02-27
down
wechat
bug