Preoperative differentiation of spinal tuberculosis, pyogenic, and brucellar spondylitis: a multimodal machine learning study across five centers.
Authors
Affiliations (7)
Affiliations (7)
- Department of Orthopedics, General Hospital of Ningxia Medical University, Yinchuan, China.
- First Clinical Medical College, Ningxia Medical University, Yinchuan, China.
- Department of Orthopedics, Zhoukou Central Hospital, Zhoukou, China.
- Department of Tuberculosis, Henan Infectious Disease Hospital (The Sixth People's Hospital of Zhengzhou), Zhengzhou, China.
- Department of General Practice, The First People's Hospital of Zhengzhou, Zhengzhou, China.
- Department of Spinal and Spinal Cord Surgery, Henan Provincial People's Hospital, Zhengzhou, China.
- Ningxia Key Laboratory of Clinical Pathogenic Microorganisms, General Hospital of Ningxia Medical University, Yinchuan, China.
Abstract
Infectious spondylitis, encompassing spinal tuberculosis (STB), pyogenic spondylitis (PS), and brucellar spondylitis (BS), is difficult to diagnose preoperatively because the three entities present with overlapping clinical features. Differentiation is often delayed by prolonged culture turnaround times and the unavailability of serological testing. In this multicenter retrospective diagnostic study at five tertiary hospitals in China (2015-2023), patients with STB (n = 1,000), PS (n = 579), and BS (n = 255) were enrolled and assigned to derivation (n = 1,439) and external validation (n = 395) cohorts. We developed and validated a multimodal machine learning model that integrates routine preoperative clinical and laboratory variables with radiomic features extracted from non-contrast MRI. Unimodal, early fusion, and late fusion architectures were compared. Pathogen-specific serological tests were excluded from the primary model to avoid data leakage. We evaluated discrimination, calibration, clinical utility, and incremental value; SHAP analysis provided case-level interpretability. The late fusion LightGBM model achieved the highest external validation performance (macro-averaged AUC 0.852, 95% CI 0.822-0.882). Radiomic integration provided significant incremental value over the clinical-only model (ΔAUC 0.080; NRI 0.398; IDI 0.072). Calibration was good (Brier score 0.158), and decision curve analysis demonstrated positive net benefit across clinically relevant thresholds. SHAP analysis revealed class-specific, biologically plausible predictors: low serum albumin and elevated T1W GLCM contrast for STB; high CRP and NLR for PS; and residence in an endemic region for BS. Secondary analysis showed that adding serological tests yielded only modest incremental improvement (ΔAUC 0.042). This preoperative multimodal explainable AI model achieves accurate and interpretable differentiation of STB, PS, and BS using only routine clinical data and non-contrast MRI, although its sensitivity for BS remains limited and prospective validation is warranted. It may serve as a clinically actionable decision-support tool for preoperative triage, particularly when serological confirmation is pending or unavailable.