Annu Int Conf IEEE Eng Med Biol Soc. 2022 Jul;2022:3882-3886. doi: 10.1109/EMBC48229.2022.9870991.
The L1-norm regularization is applied to ultrasonic reverberation characteristics imaging and vectoral Doppler measurement, of which performances are compared with those of L2-norm regularizations. The L1 regularization yields the sharper image than the L2 regularization. Alternatively, for the Doppler measurement, the L1 regularization yields less accuracy than the L2 regularization. Clinical Relevance- This study will permit us to perform quantitative ultrasonic reverberation characteristics and accurate vectoral Doppler observation.
L1-范数正则化被应用于超声混响特性成像和矢量型多普勒测量，对其性能与 L2-范数正则化进行了比较。L1 正则化比 L2 正则化产生更清晰的图像。相反，对于多普勒测量，L1 正则化的精度低于 L2 正则化。临床意义-本研究将使我们能够进行定量超声混响特性和准确的矢量型多普勒观察。
