Deep learning algorithms for identifying developmental dysplasia of the hip based on sonographic images: a retrospective, prospective, multicenter study in China
Year: 2025
Type: article
Abstract: Background: Hip ultrasound is the first-line tool to identify developmental dysplasia of the hip (DDH) among suspected infants, yet it has limitations including poor reproducibility and high diagnostic error rates. This study aims to develop and validate a deep convolutional neural network algorithm, named HipSonoNeuNet model (HSNN), using multicenter hip ultrasound data. Methods: This multicenter cross-sectional study combined data from 22 Chinese hospitals (September 2022-January 2025), enrolling 3082 participants. A total of 7286 hip ultrasound images (1429 dynamic, 5857 static) were collected and were divided into three datasets. The study was conducted in three phases. Phase I trained the models using 2431 participants (Dataset 1). Phase II compared diagnostic performance between radiologists of varied experience and the model across 500 participants (Dataset 2). Phase III prospectively validated the model's generalizability with 151 participants (Dataset 3). Findings: In Phase I, the HSNN yielded AUC of 0.99 (95% CI: 0.99-1.00), sensitivity of 1.00 (95% CI: 0.99-1.00), specificity of 0.91 (95% CI: 0.88-1.00), F1 score of 0.90 (95% CI: 0.87-1.00) on internal test dataset. In Phase II, the HSNN achieved an accuracy of 0.94 (95% CI: 0.88-1.00), AUC of 0.99 (95% CI: 0.99-1.00), sensitivity of 1.00 (95% CI: 0.99-1.00), specificity of 0.94 (95% CI: 0.87-1.00), F1 score of 0.58 (95% CI: 0.50-0.66), and strong agreement with expert (κ = 0.77). AI assistance improved all 7 junior radiologists' diagnostic performance (accuracy from 0.90 to 0.93, AUC from 0.80 to 0.95, sensitivity from 0.69 to 0.97) and reduced examination time with enhanced interobserver agreement. In Phase III, the model maintained robust performance (accuracy = 0.92, AUC = 0.99, sensitivity = 1.00, κ with experts = 0.76). Interpretation: The HSNN demonstrates accurate, robust, and generalizable performance in DDH detection. Funding: 1. Guangdong High-level Hospital Construction Fund (SZGSP012), 2. Shenzhen Clinical Research Center (20220819113341005) "Shenzhen Clinical Research Center for Child Health and Disease (szcrc2024_005)", 3. Guangdong Medical Research Funded Project (A2024019), 4. Shenzhen Science and Technology Innovation Commission General Program for Basic Research (JCYJ20220530160000001), and National Natural Science Foundation of China (82271996).
Source: EClinicalMedicine
Authors: Na Xu, Tong Han, Bingxuan Huang, Wei Fan, Xiping Chen, Miaochan Zhu, Lili Miao, Yuhao Huang, Zhiyuan Zhu, Lili Tong, Liang Chen, Jinqiao Liu, Shangfu Lin, Lili Nie, Chuan Liu, Jun Gao, Xia Zhan, Lin Lin, Mudan Meng, ShanShan Xu, Yehong Wang, Honglan Peng, Xuanfan Hu, Ziyi Cao, Zongyu Zhang, Dejuan Kong, Tao Feng, Dong Ni, Xin Yang, Luyao Zhou
Institutions: Shenzhen University, Shantou University, Shantou University Medical College, Shenzhen Children's Hospital, Wuhan Union Hospital, Huazhong University of Science and Technology, Kunming Medical University, Kunming Children's Hospital, Children's Hospital of Chongqing Medical University, Chongqing Medical University, Shandong University, China Three Gorges University, Central South University, Hunan Children's Hospital, Wuhan Children's Hospital, Jinan University, Southern University of Science and Technology, ShenZhen People’s Hospital, Hainan Medical University, Zunyi Medical University, Minda Hospital, Jilin University, First Hospital of Jilin University, Binzhou People's Hospital
Language: English
Subfield: Health Informatics
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $5,780
Awards: G3444250074, G4087466556
OpenAlex ID: W4415017438