People's Health Press
ISSN 2096-2738 CN 11-9370/R

Source Journal for Chinese Scientific and Technical Papers and Citations
Source Journal for Annual Report for Chinese Academic Journal Impact Factors(2022)
Indexed Journals in the Database of the Chemical Abstracts Service (CAS), USA
Indexed Journals in the Database of the Japan Science and Technology Agency (JST)

  • Official WeChat

  • Official Weibo

  • Electronic Journal of Emerging Infectious Diseases ›› 2026, Vol. 11 ›› Issue (4): 30-35.doi: 10.19871/j.cnki.xfcrbzz.2026.04.005

    • Original Articles • Previous Articles     Next Articles

    Development and evaluation of a musculoskeletal ultrasound-based prediction model for knee joint involvement in brucellosis

    Qi Xiu1, Ma Lingfeng2, Li Ping3   

    1. 1. Department of Ultrasound, Hohhot Second Hospital,Hohhot 010020, China;
      2. Department of Ultrasound, The Eighth Affiliated Hospital of Southern Medical University, The First People's Hospital of Shunde, Guangdong Foshan 528308, China;
      3. Department of Ultrasound, The Second Affiliated Hospital of Inner Mongolia Medical University, Inner Mongolia Hohhot 010030, China
    • Received:2026-05-28 Online:2026-08-31 Published:2026-09-14

    Abstract: Objective To explore the high risk factors of knee joint involvement in patients with brucellosis based on musculoskeletal ultrasound, to construct and verify the prediction model of knee joint involvement, and to provide theoretical support for early clinical screening, risk stratification and individualized intervention. Method A total of 132 patients with brucellosis who were admitted to the Second Hospital of Hohhot from October 2023 to October 2025 were enrolled, including 60 patients with knee joint involvement. General data and musculoskeletal ultrasound images were collected. Independent risk factors were screened by χ2 test and Logistic regression analysis, and a prediction model was constructed. Bootstrap method was used for internal verification. The accuracy and clinical practicability of the prediction model were evaluated by receiver operating characteristic curve (ROC), Hosmer-Lemeshow test, calibration curve, decision curve analysis (DCA), Shapley Additive Explanations (SHAP) and nomogram. Result Blood flow signal of infrapatellar fat pad and joint cavity / suprapatellar bursa effusion were independent risk factors (all P<0.05). The model was well distinguished and the calibration degree was high. The clinical net income was good when the threshold probability of DCA was 0.07-0.65. SHAP showed that the blood flow signal of infrapatellar fat pad contributed the most, followed by joint cavity / suprapatellar bursa effusion, both of which positively increased the risk. The prediction probability of the nomogram visualization model is 64.89%. Conclusion Blood flow signal of infrapatellar fat pad and joint cavity / suprapatellar bursa effusion is an independent risk factor for knee joint involvement in patients with brucellosis. The prediction model has the ability to predict the risk of knee joint involvement.

    Key words: musculoskeletal ultrasound, Brucellosis, Knee joint involvement, Shapley additive explanations

    CLC Number: