An ultrasound standard plane detection model of fetal head based on multi-task learning and hybrid knowledge graph

Lei Zhao, Kenli Li, Bin Pu, Jianguo Chen, Shengli Li, Xiangke Liao

Abstract: Highlights•A multi-task learning model that jointly optimizes key structure recognition and plane classification tasks is proposed for fetal head ultrasound standard plane detection.•Prior clinical knowledge is integrated into our method to reduce the missed and error detection rates.•Unlike most “end-to-end” automatic detection modes, our method provides explanations for its decisions.•Compared with several competitive benchmarks, our method displays a promising performance.