Super-resolution ultrasound (SR-US) imaging is a new technique that breaks the diffraction limit and allows visualization of microvascular structures down to tens of micrometers. The image processing methods for the spatiotemporal filtering needed in SR-US, such as singular value decomposition (SVD), are computationally burdensome and performed offline. Deep learning has been applied to many biomedical imaging problems, and trained neural networks have been shown to process an image in milliseconds. The goal of this study was to evaluate the effectiveness of deep learning to realize a spatiotemporal filter in the context of SR-US processing. A 3-D convolutional neural network (3DCNN) was trained on in vitro and in vivo data sets using SVD as ground truth in tissue clutter reduction. In vitro data were obtained from a tissue-mimicking flow phantom, and in vivo data were collected from murine tumors of breast cancer. Three training techniques were studied: training with in vitro data sets, training with in vivo data sets, and transfer learning with initial training on in vitro data sets followed by fine-tuning with in vivo data sets. The neural network trained with in vitro data sets followed by fine-tuning with in vivo data sets had the highest accuracy at 88.0%. The SR-US images produced with deep learning allowed visualization of vessels as small as $25~\mu \text{m}$ in diameter, which is below the diffraction limit (wavelength of $110~\mu \text{m}$ at 14 MHz). The performance of the 3DCNN was encouraging for real-time SR-US imaging with an average processing frame rate for in vivo data of 51 Hz with GPU acceleration.

中文翻译：

---

时空滤波的深度学习，用于快速超高分辨率超声成像

超分辨率超声（SR-US）成像是一项新技术，它打破了衍射极限，并允许可视化到数十微米的微血管结构。SR-US中所需的用于时空滤波的图像处理方法（例如奇异值分解（SVD））在计算上很繁琐，并且可以离线执行。深度学习已应用于许多生物医学成像问题，并且训练有素的神经网络已显示可在毫秒内处理图像。这项研究的目的是评估深度学习在SR-US处理背景下实现时空滤波器的有效性。对3D卷积神经网络（3DCNN）进行了训练体外 和 体内 SVD作为减少组织杂波的基本事实的数据集。 体外 数据是从模仿组织的流动模型获得的，并且 体内从乳腺癌的鼠类肿瘤中收集数据。研究了三种训练技巧：体外 数据集，通过 体内 数据集，并通过初步培训对学习进行转移 体外 数据集，然后进行微调 体内数据集。经过训练的神经网络体外 数据集，然后进行微调 体内数据集的最高准确性为88.0％。通过深度学习生成的SR-US图像可以可视化小至的血管 $ 25〜\ mu \ text {m} $ 直径，低于衍射极限（ $ 110〜\ mu \ text {m} $ 在14 MHz）。3DCNN的性能对于实时SR-US成像令人鼓舞，其平均处理帧速率为体内 GPU加速的51 Hz数据。