Suppr
超能文献

基于病灶信息增强补丁和合成数据增强的卷积神经网络对 CT 图像中的肝脏局灶性病变进行分类。

Classification of focal liver lesions in CT images using convolutional neural networks with lesion information augmented patches and synthetic data augmentation.

机构信息

School of Electrical Engineering, Korea Advanced Institute of Science and Technology, Daejeon, Republic of Korea.

Department of Software Convergence, College of Interdisciplinary Studies for Emerging Industries, Seoul Women's University, Seoul, Republic of Korea.

出版信息

Med Phys. 2021 Sep;48(9):5029-5046. doi: 10.1002/mp.15118. Epub 2021 Aug 4.

DOI:10.1002/mp.15118

PMID:34287951

Abstract

PURPOSE

We propose a deep learning method that classifies focal liver lesions (FLLs) into cysts, hemangiomas, and metastases from portal phase abdominal CT images. We propose a synthetic data augmentation process to alleviate the class imbalance and the Lesion INformation Augmented (LINA) patch to improve the learning efficiency.

METHODS

A dataset of 502 portal phase CT scans of 1,290 FLLs was used. First, to alleviate the class imbalance and to diversify the training data patterns, we suggest synthetic training data augmentation using DCGAN-based lesion mask synthesis and pix2pix-based mask-to-image translation. Second, to improve the learning efficiency of convolutional neural networks (CNNs) for the small lesions, we propose a novel type of input patch termed the LINA patch to emphasize the lesion texture information while also maintaining the lesion boundary information in the patches. Third, we construct a multi-scale CNN through a model ensemble of ResNet-18 CNNs trained on LINA patches of various mini-patch sizes.

RESULTS

The experiments demonstrate that (a) synthetic data augmentation method shows characteristics different but complementary to those in conventional real data augmentation in augmenting data distributions, (b) the proposed LINA patches improve classification performance compared to those by existing types of CNN input patches due to the enhanced texture and boundary information in the small lesions, and (c) through an ensemble of LINA patch-trained CNNs with different mini-patch sizes, the multi-scale CNN further improves overall classification performance. As a result, the proposed method achieved an accuracy of 87.30%, showing improvements of 10.81%p and 15.0%p compared to the conventional image patch-trained CNN and texture feature-trained SVM, respectively.

CONCLUSIONS

The proposed synthetic data augmentation method shows promising results in improving the data diversity and class imbalance, and the proposed LINA patches enhance the learning efficiency compared to the existing input image patches.

摘要

目的

我们提出了一种深度学习方法，能够从门静脉期腹部 CT 图像中将肝脏局灶性病变（FLL）分为囊肿、血管瘤和转移瘤。我们提出了一种合成数据增强过程，以减轻类别不平衡问题，并使用病变信息增强（LINA）补丁来提高学习效率。

方法

使用了一个包含 502 个门静脉期 CT 扫描和 1290 个 FLL 的数据集。首先，为了减轻类别不平衡问题并使训练数据模式多样化，我们建议使用基于 DCGAN 的病变掩模合成和基于 pix2pix 的掩模到图像转换来进行合成训练数据增强。其次，为了提高卷积神经网络（CNN）对小病变的学习效率，我们提出了一种新型输入补丁，称为 LINA 补丁，它在保持补丁中病变边界信息的同时，强调病变纹理信息。第三，我们通过在不同 mini-patch 大小的 LINA 补丁上训练的 ResNet-18 CNN 的模型集成构建了一个多尺度 CNN。

结果

实验表明：（a）合成数据增强方法在增强数据分布方面表现出与传统真实数据增强不同但互补的特征；（b）与现有类型的 CNN 输入补丁相比，所提出的 LINA 补丁由于小病变中的增强纹理和边界信息，提高了分类性能；（c）通过具有不同 mini-patch 大小的 LINA 补丁训练的 CNN 的集成，多尺度 CNN 进一步提高了整体分类性能。因此，所提出的方法实现了 87.30%的准确率，与传统的图像补丁训练的 CNN 和纹理特征训练的 SVM 相比，分别提高了 10.81%和 15.0%。