Faculty Mechanical and Medical Engineering, Furtwangen University of Applied Science, Villingen-Schwenningen, Germany.
Faculty Informatik, Institute for Data Science, Cloud Computing and IT Security (IDACUS), Furtwangen University of Applied Science, Furtwangen, Germany.
Int J Comput Assist Radiol Surg. 2022 Dec;17(12):2231-2237. doi: 10.1007/s11548-022-02737-6. Epub 2022 Aug 26.
Ultrasound (US) and Shear Wave Elastography (SWE) imaging are non-invasive methods used for breast lesion characterization. While US and SWE images provide both morphological information, SWE visualizes in addition the elasticity of tissue. In this study a Discriminative Convolutional Neural Network (DCNN) model is applied to US and SWE images and their combination to classify the breast lesions into malignant or benign cases. Furthermore, it is identified whether analysing only the region of the elastogram or including the surrounding B-mode image gives a superior performance.
The dataset used in this study consists of 746 images obtained from 207 patients comprising 486 malignant and 260 benign breast lesions. From each image the US and SWE image was extracted, once including only the region of the elastogram and once including also the surrounding B-mode image. These four datasets were applied individually to a DCNN to determine their predictive capability. Each the best US and SWE dataset were used to examine different combination methods with DCNN. The results were compared to the manual assessment by an expert radiologist.
The combination of US and SWE images with the surrounding B-mode image using two ensembled DCNN models achieved best results with an accuracy of 93.53 %, sensitivity of 94.42 %, specificity of 90.75 % and area under the curve (AUC) of 96.55 %.
This study showed that using the whole US and SWE images through DCNN was superior to methods, in which only the region of elastogram was used. Combining breast cancer US and SWE images with two ensembled DCNN models in parallel improved the results. The accuracy, sensitivity and AUC of the best combination method were significantly superior to the results of using a single dataset through DCNN and to the results of the expert radiologist.
超声（US）和剪切波弹性成像（SWE）是用于乳房病变特征描述的非侵入性方法。虽然 US 和 SWE 图像提供了形态学信息，但 SWE 还可以可视化组织的弹性。在这项研究中，应用判别卷积神经网络（DCNN）模型对 US 和 SWE 图像及其组合进行分析，以将乳房病变分为恶性或良性病例。此外，还确定了仅分析弹性图区域或包括周围 B 模式图像是否会提高性能。
本研究使用的数据集包含 207 名患者的 746 张图像，其中包括 486 例恶性和 260 例良性乳腺病变。从每张图像中提取 US 和 SWE 图像，一次仅包括弹性图区域，另一次包括周围 B 模式图像。将这四个数据集分别应用于 DCNN 以确定其预测能力。使用最佳的 US 和 SWE 数据集来检查 DCNN 的不同组合方法。将结果与专家放射科医生的手动评估进行比较。
使用周围 B 模式图像的 US 和 SWE 图像与两个集成 DCNN 模型的组合获得了最佳结果，准确率为 93.53%，灵敏度为 94.42%，特异性为 90.75%，曲线下面积（AUC）为 96.55%。
本研究表明，使用 DCNN 对整个 US 和 SWE 图像进行分析优于仅使用弹性图区域的方法。通过并行使用两个集成的 DCNN 模型结合乳腺癌 US 和 SWE 图像可改善结果。最佳组合方法的准确性、灵敏度和 AUC 均显著优于通过 DCNN 使用单个数据集的结果，也显著优于专家放射科医生的结果。
