Convolutional Neural Networks (CNNs), which are currently state-of-the-art for most image analysis tasks, are ill suited to leveraging the key benefits of ultrasound imaging - specifically, ultrasound's portability and real-time capabilities. CNNs have large memory footprints, which obstructs their implementation on mobile devices, and require numerous floating point operations, which results in slow CPU inference times. In this paper, we propose three approaches to training efficient CNNs that can operate in real-time on a CPU (catering to the clinical setting), with a low memory footprint, for minimal compromise in accuracy. We first demonstrate the power of 'thin' CNNs, with very few feature channels, for fast medical image segmentation. We then leverage separable convolutions to further speed up inference, reduce parameter count and facilitate mobile deployment. Lastly, we propose a novel knowledge distillation technique to boost the accuracy of light-weight models, while maintaining inference speed-up. For a negligible sacrifice in test set Dice performance on the challenging ultrasound analysis task of nerve segmentation, our final proposed model processes images at 30fps on a CPU, which is 9× faster than the standard U-Net, while requiring 420× less space in memory.

中文翻译：

---

低内存CNN支持对移动部署进行实时超声分割。

卷积神经网络（CNN）当前是大多数图像分析任务的最新技术，不适用于利用超声成像的关键优势-特别是超声的便携性和实时功能。CNN占用的内存很大，这妨碍了它们在移动设备上的实现，并且需要大量的浮点运算，这导致CPU推理时间变慢。在本文中，我们提出了三种方法来训练有效的CNN，这些CNN可以在CPU上实时运行（适应临床情况），且内存占用量少，以最大程度地降低准确性。我们首先展示具有很少特征通道的“薄” CNN的功能，以实现快速医学图像分割。然后，我们利用可分离的卷积来进一步加快推理速度，减少参数数量并促进移动部署。最后，我们提出了一种新颖的知识提炼技术，以提高轻量模型的准确性，同时保持推理速度。为了在神经分割的挑战性超声分析任务中牺牲测试设备Dice的性能而忽略不计，我们最终提出的模型在CPU上以30fps的速度处理图像，这比标准U-Net快9倍，同时所需的空间减少了420倍记忆。