Lin Zhouqin, Zhang Haoming, Duan Xingxing, Bai Yan, Wang Jian, Liang Qianhong, Zhou Jingran, Xie Fusui, Shentu Zhen, Huang Ruobing, Chen Yayan, Yu Hongkui, Weng Zongjie, Ni Dong, Liu Lei, Zhou Luyao
Department of Medical Ultrasonics, Shenzhen Children's Hospital, Shenzhen, PR China.
Shenzhen Pediatrics Institute of Shantou University Medical College, Shenzhen, PR China.
Nat Commun. 2025 Aug 20;16(1):7778. doi: 10.1038/s41467-025-63096-9.
Timely and accurate diagnosis of severe neonatal cerebral lesions is critical for preventing long-term neurological damage and addressing life-threatening conditions. Cranial ultrasound is the primary screening tool, but the process is time-consuming and reliant on operator's proficiency. In this study, a deep-learning powered neonatal cerebral lesions screening system capable of automatically extracting standard views from cranial ultrasound videos and identifying cases with severe cerebral lesions is developed based on 8,757 neonatal cranial ultrasound images. The system demonstrates an area under the curve of 0.982 and 0.944, with sensitivities of 0.875 and 0.962 on internal and external video datasets, respectively. Furthermore, the system outperforms junior radiologists and performs on par with mid-level radiologists, with 55.11% faster examination efficiency. In conclusion, the developed system can automatically extract standard views and make correct diagnosis with efficiency from cranial ultrasound videos and might be useful to deploy in multiple application scenarios.
及时、准确地诊断严重的新生儿脑部病变对于预防长期神经损伤和处理危及生命的状况至关重要。头颅超声是主要的筛查工具，但该过程耗时且依赖操作者的熟练程度。在本研究中，基于8757张新生儿头颅超声图像，开发了一种深度学习驱动的新生儿脑部病变筛查系统，该系统能够从头颅超声视频中自动提取标准视图并识别患有严重脑部病变的病例。该系统在内部和外部视频数据集上的曲线下面积分别为0.982和0.944，灵敏度分别为0.875和0.962。此外，该系统的表现优于初级放射科医生，与中级放射科医生相当，检查效率提高了55.11%。总之，所开发的系统能够从头颅超声视频中自动提取标准视图并高效地做出正确诊断，可能有助于在多种应用场景中部署。
