School of Medical Information Engineering, Guangzhou University of Chinese Medicine, Guangzhou, China, 510006.
Department of Radiology, Guangdong General Hospital, Guangzhou, China, 510080.
Med Phys. 2019 Sep;46(9):3951-3960. doi: 10.1002/mp.13642. Epub 2019 Jul 20.
To effectively grade hepatocellular carcinoma (HCC) based on deep features derived from diffusion weighted images (DWI) with multiple b-values using convolutional neural networks (CNN).
Ninety-eight subjects with 100 pathologically confirmed HCC lesions from July 2012 to October 2018 were included in this retrospective study, including 47 low-grade and 53 high-grade HCCs. DWI was performed for each subject with a 3.0T MR scanner in a breath-hold routine with three b-values (0,100, and 600 s/mm ). First, logarithmic transformation was performed on original DWI images to generate log maps (logb0, logb100, and logb600). Then, a resampling method was performed to extract multiple 2D axial planes of HCCs from the log map to increase the dataset for training. Subsequently, 2D CNN was used to extract deep features of the log map for HCCs. Finally, fusion of deep features derived from three b-value log maps was conducted for HCC malignancy classification. Specifically, a deeply supervised loss function was devised to further improve the performance of lesion characterization. The data set was split into two parts: the training and validation set (60 HCCs) and the fixed test set (40 HCCs). Four-fold cross validation with 10 repetitions was performed to assess the performance of deep features extracted from single b-value images for HCC grading using the training and validation set. Receiver operating characteristic curve (ROC) and area under the curve (AUC) values were used to assess the characterization performance of the proposed deep feature fusion method to differentiate low-grade and high-grade in the fixed test set.
The proposed fusion of deep features derived from logb0, logb100, and logb600 with deeply supervised loss function generated the highest accuracy for HCC grading (80%), thus outperforming the method of deep feature derived from the ADC map directly (72.5%), the original b0 (65%), b100 (68%), and b600 (70%) images. Furthermore, AUC values of the deep features of the ADC map, the deep feature fusion with concatenation, and the proposed deep feature fusion with deeply supervised loss function were 0.73, 0.78, and 0.83, respectively.
The proposed fusion of deep features derived from the logarithm of the three b-value images yields high performance for HCC grading, thus providing a promising approach for the assessment of DWI in lesion characterization.
利用卷积神经网络（CNN）从多 b 值扩散加权图像（DWI）中提取的深度特征，有效地对肝细胞癌（HCC）进行分级。
本回顾性研究纳入了 2012 年 7 月至 2018 年 10 月期间 98 例经病理证实的 HCC 病变患者，包括 47 例低级别 HCC 和 53 例高级别 HCC。对每位患者使用 3.0T MR 扫描仪在屏气常规下进行 DWI 检查，采用 3 个 b 值（0、100 和 600 s/mm ）。首先，对原始 DWI 图像进行对数变换，生成对数图（logb0、logb100 和 logb600）。然后，采用重采样方法从 log 图中提取 HCC 的多个 2D 轴位平面，以增加训练数据集。随后，使用 2D CNN 从 log 图中提取 HCC 的深度特征。最后，融合来自三个 b 值 log 图的深度特征进行 HCC 恶性程度分类。具体来说，设计了一种深度监督损失函数，以进一步提高病变特征描述的性能。数据集分为两部分：训练集和验证集（60 例 HCC）和固定测试集（40 例 HCC）。使用训练集和验证集进行四折交叉验证，重复 10 次，评估从单 b 值图像提取的深度特征进行 HCC 分级的性能。使用接收器工作特征曲线（ROC）和曲线下面积（AUC）值评估所提出的深度特征融合方法在固定测试集区分低级别和高级别的能力。
提出的融合 logb0、logb100 和 logb600 衍生的深度特征并采用深度监督损失函数，用于 HCC 分级的准确率最高（80%），优于直接从 ADC 图提取深度特征（72.5%）、原始 b0（65%）、b100（68%）和 b600（70%）图像的方法。此外，ADC 图的深度特征、串联融合的深度特征和所提出的深度特征融合的深度特征的 AUC 值分别为 0.73、0.78 和 0.83。
从三个 b 值图像的对数衍生的深度特征融合，可为 HCC 分级提供高性能，从而为 DWI 在病变特征描述中的评估提供了一种有前景的方法。
