3D B-mode ultrasound speckle reduction using deep learning for 3D registration applications
This addresses the slow processing of 3D ultrasound images for medical applications, though it is incremental as it extends deep learning from 2D to 3D speckle reduction.
The study tackled the problem of 3D B-mode ultrasound speckle reduction, which impedes image processing tasks like registration, by proposing a 3D dense U-Net model. The result showed similar suppression and mean preservation index (1.066 vs. 0.978) compared to conventional filters while reducing runtime by two orders of magnitude and halving the mean square error in 3D registration.
Ultrasound (US) speckles are granular patterns which can impede image post-processing tasks, such as image segmentation and registration. Conventional filtering approaches are commonly used to remove US speckles, while their main drawback is long run-time in a 3D scenario. Although a few studies were conducted to remove 2D US speckles using deep learning, to our knowledge, there is no study to perform speckle reduction of 3D B-mode US using deep learning. In this study, we propose a 3D dense U-Net model to process 3D US B-mode data from a clinical US system. The model's results were applied to 3D registration. We show that our deep learning framework can obtain similar suppression and mean preservation index (1.066) on speckle reduction when compared to conventional filtering approaches (0.978), while reducing the runtime by two orders of magnitude. Moreover, it is found that the speckle reduction using our deep learning model contributes to improving the 3D registration performance. The mean square error of 3D registration on 3D data using 3D U-Net speckle reduction is reduced by half compared to that with speckles.