Deep Learning-Enabled Ultrasound for Advancing Anterior Talofibular Ligament Injuries Classification: A Multicenter Model Development and Validation Study.
Authors
Affiliations (17)
Affiliations (17)
- Department of Trauma and Orthopedics, Peking University People's Hospital, No.11, Xizhimen South Street, Xicheng District, Beijing 100044, PR China (X.S., H.X.).
- CAS Key Laboratory of Molecular Imaging, Institute of Automation, Chinese Academy of Sciences, No. 95, Zhongguancun East Road, Haidian District, Beijing 100190, PR China (H.Z., K.W.); School of Artificial Intelligence, University of Chinese Academy of Sciences, No.19, Yuquan Road, Shijingshan District, Beijing 100049, PR China (H.Z., K.W.).
- Department of Ultrasound, Tianjin Hospital, Tianjin University, No. 406, Jiefang South Road, Hexi District, Tianjin 300211, PR China (Y.Y.).
- Aerospace Information Research Institute (AIR), Chinese Academy of Sciences (CAS), No. 9, Dengzhuang South Road, Haidian District, Beijing 100094, PR China (Z.X.); The College of Resources and Environment, University of Chinese Academy of Sciences, No.19, Yuquan Road, Shijingshan District, Beijing 100049, PR China (Z.X.).
- Department of Ultrasound, Honghui Hospital, Xi'an Jiaotong University College of Medicine, No.76, Nanguo Road, Beilin District, Xi'an, Shaanxi 710000, PR China (L.M.).
- Department of Ultrasound, Orthopedic Hospital of Henan Province (Zhengzhou Campus), No.100, Yongping Road, Zhengdong District, Zhengzhou, Henan 450000, PR China (Z.X., Y.Q.).
- Department of Ultrasound, Affiliated Dongguan Hospital, Southern Medical University, No.3, Wandao Road, Wanjiang District, Dongguan, Guangdong 523000, PR China (S.L.).
- Department of Ultrasound, Postgraduate Yuncheng Central Hospital, No. 1, Renmin Road, Yanhu District, Yuncheng, Shanxi 044000, PR China (J.S.).
- Department of Ultrasound, Beijing Daxing District Hospital of Integrated Chinese and Western Medicine, No. 3, Zhongxing South Road, Yinghai town, Daxing District, Beijing 100076, PR China (J.C.).
- Department of Ultrasound, General Hospital of Central Theater Command, No. 627, Wuluo Road, Wuchang District, Wuhan, Hubei 430070, PR China (R.D.).
- Department of Ultrasound, People's Hospital Affiliated to Chongqing Three Gorges Medical College, No. 38, Gaosuntang Street, Wanzhou District, Chongqing 404100, PR China (Q.Y.).
- Department of Ultrasound, The First Affiliated Hospital of Guangxi University of Traditional Chinese Medicine, No. 89, Dongge Road, Qingxiu District, Nanning, Guangxi 530023, PR China (D.W.).
- Department of Functional Examination, Orthopedic Hospital of Henan Province (Luoyang Campus), No. 82, Qiming South Road, Chanhe District, Luoyang, Henan 471000, PR China (S.S.).
- Department of Ultrasound, Yunnan Provincial Hospital of Traditional Chinese Medicine, No. 1, Huachen Road, Xishan District, Kunming, Yunnan 650103, PR China (C.G.).
- Department of Ultrasound Medical Center, The Second Affiliated Hospital of Inner Mongolia University of Inner Mongolia Medical University, No.59, Keerqin South Road, Saihan District, Hohhot, Inner Mongolia 010000, PR China (P.L.).
- Department of Ultrasound, Shijiazhuang People's Hospital, No.30, Fanxi Road, Changan District, Shijiazhuang, Hebei 50011, PR China (L.B.).
- CAS Key Laboratory of Molecular Imaging, Institute of Automation, Chinese Academy of Sciences, No. 95, Zhongguancun East Road, Haidian District, Beijing 100190, PR China (H.Z., K.W.); School of Artificial Intelligence, University of Chinese Academy of Sciences, No.19, Yuquan Road, Shijingshan District, Beijing 100049, PR China (H.Z., K.W.). Electronic address: [email protected].
Abstract
Ultrasound (US) is the preferred modality for assessing anterior talofibular ligament (ATFL) injuries. We aimed to advance ATFL injuries classification by developing a US-based deep learning (DL) model, and explore how artificial intelligence (AI) could help radiologists improve diagnostic performance. Consecutive healthy controls and patients with acute ATFL injuries (mild strain, partial tear, complete tear, and avulsion fracture) at 10 hospitals were retrospectively included. A US-based DL model (ATFLNet) was trained (n=2566), internally validated (n=642), and externally validated (n=717 and 493). Surgical or radiological findings based on the majority consensus of three experts served as the reference standard. Prospective validation was conducted at three additional hospitals (n=472). The performance was compared to that of 12 radiologists at different levels (external validation sets 1 and 2); an ATFLNet-aided strategy was developed, comparing with the radiologists when reviewing B-mode images (external validation set 2); the strategy was then tested in a simulated scenario (reviewing images alongside dynamic clips; prospective validation set). Statistical comparisons were performed using the McNemar's test, while inter-reader agreement was evaluated with the Multireader Fleiss κ statistic. ATFLNet obtained macro-average area under the curve ≥0.970 across all five classes in each dataset, indicating robust overall performance. Additionally, it consistently outperformed senior radiologists in external validation sets (all p<.05). ATFLNet-aided strategy improved radiologists' average accuracy (0.707 vs. 0.811, p<.001) for image review. In the simulated scenario, it led to enhanced accuracy (0.794 to 0.864, p=.003), and a reduction in diagnostic variability, particularly for junior radiologists. Our US-based model outperformed human experts for ATFL injury evaluation. AI-aided strategies hold the potential to enhance diagnostic performance in real-world clinical scenarios.