Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Topology-Guided Biomechanical Profiling: A White-Box Framework for Opportunistic Screening of Spinal Instability on Routine CT

Zanting Ye, Xuanbin Wu, Guoqing Zhong, Shengyuan Liu, Jiashuai Liu, Ge Song, Zhisong Wang, Jing Hao, Xiaolong Niu, Yefeng Zheng, Yu Zhang, Lijun Lu

Abstract

Routine oncologic computed tomography (CT) presents an ideal opportunity for screening spinal instability, yet prophylactic stabilization windows are frequently missed due to the complex geometric reasoning required by the Spinal Instability Neoplastic Score (SINS). Automating SINS is fundamentally hindered by metastatic osteolysis, which induces topological ambiguity that confounds standard segmentation and black-box AI. We propose Topology-Guided Biomechanical Profiling (TGBP), an auditable white-box framework decoupling anatomical perception from structural reasoning. TGBP anchors SINS assessment on two deterministic geometric innovations: (i) canal-referenced partitioning to resolve posterolateral boundary ambiguity, and (ii) context-aware morphometric normalization via covariance-based oriented bounding boxes (OBB) to quantify vertebral collapse. Integrated with auxiliary radiomic and large language model (LLM) modules, TGBP provides an end-to-end, interpretable SINS evaluation. Validated on a multi-center, multi-cancer cohort ($N=482$), TGBP achieved 90.2\% accuracy in 3-tier stability triage. In a blinded reader study ($N=30$), TGBP significantly outperformed medical oncologists on complex structural features ($κ=0.857$ vs.\ $0.570$) and prevented compounding errors in Total Score estimation ($κ=0.625$ vs.\ $0.207$), democratizing expert-level opportunistic screening.

Topology-Guided Biomechanical Profiling: A White-Box Framework for Opportunistic Screening of Spinal Instability on Routine CT

Abstract

Routine oncologic computed tomography (CT) presents an ideal opportunity for screening spinal instability, yet prophylactic stabilization windows are frequently missed due to the complex geometric reasoning required by the Spinal Instability Neoplastic Score (SINS). Automating SINS is fundamentally hindered by metastatic osteolysis, which induces topological ambiguity that confounds standard segmentation and black-box AI. We propose Topology-Guided Biomechanical Profiling (TGBP), an auditable white-box framework decoupling anatomical perception from structural reasoning. TGBP anchors SINS assessment on two deterministic geometric innovations: (i) canal-referenced partitioning to resolve posterolateral boundary ambiguity, and (ii) context-aware morphometric normalization via covariance-based oriented bounding boxes (OBB) to quantify vertebral collapse. Integrated with auxiliary radiomic and large language model (LLM) modules, TGBP provides an end-to-end, interpretable SINS evaluation. Validated on a multi-center, multi-cancer cohort (), TGBP achieved 90.2\% accuracy in 3-tier stability triage. In a blinded reader study (), TGBP significantly outperformed medical oncologists on complex structural features ( vs.\ ) and prevented compounding errors in Total Score estimation ( vs.\ ), democratizing expert-level opportunistic screening.
Paper Structure (13 sections, 3 equations, 2 figures, 3 tables, 1 algorithm)

This paper contains 13 sections, 3 equations, 2 figures, 3 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Methodology
  3. Preprocessing and Segmentation Backbone
  4. Canal-Referenced Topological Partitioning
  5. Context-Aware Morphometric Normalization
  6. Kinematic Alignment Profiling
  7. Integration of Standardized Clinical Modules
  8. End-to-End Triage and Trustworthy AI
  9. Experiments and Results
  10. Experimental Setup and Baseline Localization
  11. System-Level Triage and Component Evaluation
  12. Bridging the Diagnostic Gap: The Compounding Error
  13. Discussion and Conclusion

Figures (2)

  • Figure 1: Motivation of TGBP. A reader study ($N=30$) exposes a diagnostic gap where medical oncologists struggle with complex SINS criteria, leading to missed surgical windows.
  • Figure 2: Overview of the TGBP framework. Left (Segmentation): A segmentation backbone trained via HITL strategy is frozen for downstream deployment. Middle (Explainable Modeling): The white-box engine translates anatomical masks and clinical text into explicit SINS criteria via transparent geometry and LLMs. Right (Triage & VR Integration): Profiles are aggregated for 3-tier stability classification. Notably, TGBP supports both fully automated screening and a semi-automated VR mode for immersive expert verification and correction.