Orthopaedic Surgery · Published 2026-06-16 · DOI 10.1111/os.70335
Kejia Zhu, Hang Li, Hui Zhang, Biao Wang, Bin Shen, Yong Nie
ABSTRACT Background Deep vein thrombosis (DVT) following total knee arthroplasty (TKA) remains a major postoperative complication. Conventional risk assessment tools such as the Caprini score show limited predictive performance (AUC 0.65–0.72). We developed a multimodal deep learning framework that integrates early postoperative radiographs with clinical data to improve DVT risk stratification. Methods This single‐center retrospective cohort study analyzed 1200 patients undergoing primary unilateral TKA (DVT incidence: 18.0%) from January 2018 to June 2023. A dual‐branch architecture was implemented: ViT‐B/16 or ResNet50 processed standardized radiographs (metal artifact suppression; 512 × 512 resolution), while Clinical‐BERT encoded structured electronic medical record (EMR) representations. A dynamic attention fusion module generated sample‐specific modality weights (α) for feature fusion (α·v_img + (1 − α)·v_txt). Performance was evaluated on a stratified test set (n = 180) against the Caprini score, unimodal models, and a nonattention concatenation fusion baseline. AUCs were compared using the DeLong test, and sensitivity/specificity differences were assessed using the McNemar test. Results The multimodal model achieved an AUC of 0.89 (95% CI: 0.85–0.92; DeLong p < 0.001 vs. all comparators), sensitivity 83% (76%–89%), specificity 88% (82%–93%), and F1‐score 0.79. It identified 91.7% (11/12) of DVT cases labeled low‐risk by Caprini. The learned attention coefficient α suggested modality reliance patterns associated with specific findings (e.g., higher α in cases with prosthesis–bone lucency > 2 mm; higher text contribution in cases with D‐dimer > 1.0 mg/L), but these associations do not establish causality. Anatomical heatmaps consistently localized high‐risk regions, particularly the popliteal vein projection area and tibial interface. Conclusion This attention‐based multimodal model integrates biomechanical (radiographic) and biochemical/clinical (EMR‐derived) signals to improve early postoperative DVT risk prediction while providing interpretable anatomical cues.
Abstract from DOAJ. Public domain (CC0 1.0).
Read the article at the publisher →
Zhu, K., Li, H., Zhang, H., et al. (2026). Multimodal Deep Learning for Predicting Deep Vein Thrombosis Risk After Total Knee Arthroplasty: A Clinical Study. Orthopaedic Surgery. https://doi.org/10.1111/os.70335