Experimental results demonstrate that each anatomical and protocol variances are enabling aspects for DNN-based mostly US reconstruction; and learning tips on how to discriminate various subjects and predefined varieties of scanning paths equally significantly increase frame prediction accuracy, quantity reconstruction overlap, accrued tracking err