FHUSP-NET: A Multi-task model for fetal heart ultrasound standard plane recognition and key anatomical structures detection

Furong Li, Ping Li, Xiuming Wu, Pan Zeng, Guorong Lyu, Yuling Fan, Peizhong Liu, Haisheng Song, Zhonghua Liu

Abstract: Highlights•An artificial intelligence method for real-time fetal heart ultrasound image analysis was studied.•A deep learning multitasking framework FHUSP-NET was validated for five fetal heart ultrasound standard planes recognition.•It can also accurately detect ten key anatomical structures of the fetal heart.