ABSTRACT
3D ultrasound (US) is widely used for its rich diagnostic information, but it suffers from a limited field of view. 3D freehand US reconstruction is promising in addressing the problem by providing broad range and freeform scan. The existing deep learning-based methods only focus on the basic cases of scan strategies, and the model rely on the training data heavily. The sequences in clinical practice involve a mix of diverse strategies and have complex scanning paths. Furthermore, deep learning models should adapt themselves to the testing cases with prior knowledge for better robustness, rather than only fit to the training cases. In this chapter, we describe a novel approach to freehand 3D US reconstruction considering the complex scan strategies. Our contribution is three-fold. First, we advance a novel online learning framework by designing a differentiable reconstruction algorithm. It realizes an end-to-end optimization from section sequences to the reconstructed volume. Second, a self-supervised learning method was developed to explore the context information that is reconstructed by the testing data itself, promoting the perception of the model. Third, inspired by the effectiveness of a shape prior, we also introduced adversarial training to strengthen the learning of anatomical the shape prior in the reconstructed 3D image. By mining the context and structural cues of the testing data, our online learning methods can drive the model to handle complex scan strategies. Experimental results on developmental dysplasia of the hip US and fetal US datasets show that our proposed method can outperform the start-of-the-art methods regarding the shift errors and path similarities.
