Ultrasound Scatterer Density Classification Using Convolutional Neural Networks by Exploiting Patch Statistics
This work provides a more robust and accurate method for classifying ultrasound scatterer density, which is crucial for quantitative ultrasound applications in medical diagnostics, particularly for scenarios with small patch sizes or varying imaging parameters.
This paper addresses the problem of classifying ultrasound scatterer density as either fully developed speckle (FDS) or low-density scatterers (LDS), which is conventionally done using statistical parameters that are inaccurate for small patch sizes and dependent on imaging settings. The authors propose a convolutional neural network (CNN) architecture, enhanced with patch statistics as additional input, which demonstrates superior performance in classifying tissues with different scatterer densities across simulation, experimental phantom, and in vivo data, while also being robust to varying imaging parameters without needing a reference phantom.
Quantitative ultrasound (QUS) can reveal crucial information on tissue properties such as scatterer density. If the scatterer density per resolution cell is above or below 10, the tissue is considered as fully developed speckle (FDS) or low-density scatterers (LDS), respectively. Conventionally, the scatterer density has been classified using estimated statistical parameters of the amplitude of backscattered echoes. However, if the patch size is small, the estimation is not accurate. These parameters are also highly dependent on imaging settings. In this paper, we propose a convolutional neural network (CNN) architecture for QUS, and train it using simulation data. We further improve the network performance by utilizing patch statistics as additional input channels. We evaluate the network using simulation data, experimental phantoms and in vivo data. We also compare our proposed network with different classic and deep learning models, and demonstrate its superior performance in classification of tissues with different scatterer density values. The results also show that the proposed network is able to work with different imaging parameters with no need for a reference phantom. This work demonstrates the potential of CNNs in classifying scatterer density in ultrasound images.