EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Development of a Multimodal Model Combining Imaging and Clinical Data to Predict Bone Cement Leakage During Vertebroplasty for Fresh Thoracolumbar Compression Fractures

Wangim Choi, Jianan Chen, Jincheng Qiu, Tianyu Xie, Yong Li, Zhiwei Pan, Zaoqiang Zhang, Guoxi Huang, Nianchun Liao, Youxi Lin iD, Zhihuai Deng, Xianjian Qiu und 5 weitere

JOR SPINE · 2026-12 · 9 (4)

DOI: 10.1002/jsp2.70227 ↗

Kurz gefasst

Fachartikel in der Zeitschrift JOR SPINE, online erschienen am 08.11.2026. Verfasst von Wangim Choi und 16 weiteren Autor:innen. Beteiligte Einrichtung: Department of Orthopedics Sun Yat-Sen Memorial Hospital Guangzhou China. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Guangzhou Municipal Science and Technology Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Bone cement leakage is a common complication of percutaneous vertebroplasty (PVP), but perioperative risk stratification remains challenging. Existing models based on single‐source clinical or imaging features have limited performance. This study aimed to develop and validate multimodal artificial intelligence (AI) models integrating clinical, imaging, and radiomics features to predict bone cement leakage during PVP.

Methods

This multicenter study included 486 treated vertebrae from 378 patients at the North, South, and Shenshan campuses of our hospital. The treated vertebra was the prediction unit, while all vertebrae from the same patient were retained in the same dataset partition and cross‐validation fold. Clinical variables, conventional CT imaging features, and CT‐based radiomics features were extracted. Logistic regression identified independent risk factors. Six machine‐learning algorithms were used to construct clinical, imaging, radiomics, and multimodal models. Performance was evaluated using the area under the receiver operating characteristic curve (AUC), accuracy, F1 score, and decision curve analysis. Independent external and prospective validations were performed, and SHapley Additive exPlanations (SHAP) were used for model interpretation.

Results

Among the 486 treated vertebrae, 215 (44.2%) had cement leakage. Higher body mass index (BMI), greater cement injection volume, cortical destruction, and posterior wall involvement were independently associated with leakage. Multimodal models generally achieved higher AUCs than the corresponding single‐modality models, with AUCs of 0.902–0.939 in the internal test cohort, 0.722–0.929 in the external validation cohort, and 0.834–0.925 in the prospective validation cohort. Patient‐level aggregation and one‐vertebra‐per‐patient sensitivity analyses yielded broadly comparable discrimination. SHAP identified cortical destruction, cement injection volume, and posterior wall involvement as the most influential predictors.

Conclusions

The multimodal models demonstrated good and generally consistent discrimination across the internal test, external validation, and prospective validation cohorts and may support perioperative risk assessment and individualized cement injection planning during PVP.

Schlagwörter

Stichwörter der Autor:innen

artificial intelligencebone cement leakagemultimodal modelpercutaneous vertebroplastyradiomics

Autor:innen und Institutionen

  • Department of Orthopedics Sun Yat-Sen Memorial Hospital Guangzhou China.
  • Department of Orthopedics The Third Affiliated Hospital of Sun Yat-Sen University Guangzhou China.
  • Department of Minimally Invasive Spine Surgery Panyu Hospital of Chinese Medicine Guangzhou Guangdong Province China.
  • Shenshan Medical Center Sun Yat-Sen Memorial Hospital, Sun Yat-sen University Shanwei Guangdong P. R. China.
  • Department of Radiology Sun Yat-Sen Memorial Hospital Guangzhou China.
  • Guangzhou University of Chinese Medicine Guangzhou China.

ORCID-Profile: Youxi Lin, Bo Gao, Zizhao Wu, Wenjie Gao

Förderung

  • National Natural Science Foundation of China
  • Guangzhou Municipal Science and Technology Project

Zitieren

APA

Choi, W., Chen, J., Qiu, J., Xie, T., Li, Y., Pan, Z., Zhang, Z., Huang, G., Liao, N., Lin, Y., Deng, Z., Qiu, X., Gao, B., Huang, D., Zhou, H., Wu, Z., & Gao, W. (2026). Development of a Multimodal Model Combining Imaging and Clinical Data to Predict Bone Cement Leakage During Vertebroplasty for Fresh Thoracolumbar Compression Fractures. JOR SPINE, 9(4). https://doi.org/10.1002/jsp2.70227

Vancouver

Choi W, Chen J, Qiu J, Xie T, Li Y, Pan Z, et al. Development of a Multimodal Model Combining Imaging and Clinical Data to Predict Bone Cement Leakage During Vertebroplasty for Fresh Thoracolumbar Compression Fractures. JOR SPINE. 2026;9(4). doi:10.1002/jsp2.70227

BibTeX
@article{Choi2026,
  author = {Choi, Wangim and Chen, Jianan and Qiu, Jincheng and Xie, Tianyu and Li, Yong and Pan, Zhiwei and Zhang, Zaoqiang and Huang, Guoxi and Liao, Nianchun and Lin, Youxi and Deng, Zhihuai and Qiu, Xianjian and Gao, Bo and Huang, Dongsheng and Zhou, Hang and Wu, Zizhao and Gao, Wenjie},
  title = {Development of a Multimodal Model Combining Imaging and Clinical Data to Predict Bone Cement Leakage During Vertebroplasty for Fresh Thoracolumbar Compression Fractures},
  journal = {JOR SPINE},
  year = {2026},
  volume = {9},
  number = {4},
  doi = {10.1002/jsp2.70227}
}
Kostenlos starten →