Deep learning approach for screening neonatal cerebral lesions on ultrasound in China.
Authors
Affiliations (22)
Affiliations (22)
- Department of Medical Ultrasonics, Shenzhen Children's Hospital, Shenzhen, PR China.
- Shenzhen Pediatrics Institute of Shantou University Medical College, Shenzhen, PR China.
- School of Biomedical Engineering, Shenzhen University, Shenzhen, PR China.
- Medical Ultrasound Image Computing (MUSIC) Lab, School of Biomedical Engineering, Medical School, Shenzhen University, Shenzhen, PR China.
- Department of Ultrasonography, Changsha Hospital for Maternal and Child Health Care, Changsha, PR China.
- Sichuan Provincial Women's and Children's Hospital/The Affiliated Women's and Children's Hospital of Chengdu Medical College, Chengdu, PR China.
- College of Computer Science and Software Engineering, Shenzhen University, Shenzhen, PR China.
- School of Artificial Intelligence, Shenzhen University, Shenzhen, PR China.
- National Engineering Laboratory for Big Data System Computing Technology, Shenzhen University, Shenzhen, PR China.
- Panyu Maternal and Child Care Service Centre of Guangzhou, Guangzhou, PR China.
- Ultrasound Department of Longhua District Maternal and Child Healthcare Hospital, Shenzhen, PR China.
- Department of ultrasound, Shenzhen Baoan Women's and Children's Hospital, Shenzhen, PR China.
- Department of Medical Ultrasonics, Fujian Maternity and Child Health Hospital, College of Clinical Medicine for Obstetrics & Gynecology and Pediatrics, Fujian Medical University, Fuzhou, PR China.
- Medical Ultrasound Image Computing (MUSIC) Lab, School of Biomedical Engineering, Medical School, Shenzhen University, Shenzhen, PR China. [email protected].
- School of Artificial Intelligence, Shenzhen University, Shenzhen, PR China. [email protected].
- National Engineering Laboratory for Big Data System Computing Technology, Shenzhen University, Shenzhen, PR China. [email protected].
- School of Biomedical Engineering and Informatics, Nanjing Medical University, Nanjing, PR China. [email protected].
- Department of Medical Ultrasonics, Shenzhen Children's Hospital, Shenzhen, PR China. [email protected].
- Shenzhen Pediatrics Institute of Shantou University Medical College, Shenzhen, PR China. [email protected].
- Department of Medical Ultrasonics, Shenzhen Children's Hospital, Shenzhen, PR China. [email protected].
- Shenzhen Pediatrics Institute of Shantou University Medical College, Shenzhen, PR China. [email protected].
- Medical School, Shenzhen University, Shenzhen, PR China. [email protected].
Abstract
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.