Tracing 3D Anatomy in 2D Strokes: A Multi-Stage Projection Driven Approach to Cervical Spine Fracture Identification

2026-01-21

AI Summary

This study addresses the high computational complexity of detecting cervical spine fractures in 3D CT scans by proposing an efficient end-to-end framework based on multi-view 2D projections. The method reconstructs 3D anatomical structures using variance and energy projections, integrating YOLOv8 for localization, DenseNet121-Unet for segmentation, and a 2.5D temporal model for vertebra-level fracture identification. Interpretability is validated through saliency maps compared against expert annotations. The approach achieves strong performance with minimal computational overhead, yielding a 3D localization mIoU of 94.45%, a vertebra segmentation Dice score of 87.86%, and vertebra-level and patient-level F1 scores of 68.15% and 82.26%, respectively. Corresponding ROC-AUC values are 91.62% and 83.04%, demonstrating diagnostic accuracy comparable to that of radiology experts.

Technology Category

• Computer Vision: Segmentation
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Security and Privacy: Data transparency and provenance
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Web data provenance, reliability, and authenticity

Abstract

Cervical spine fractures are critical medical conditions requiring precise and efficient detection for effective clinical management. This study explores the viability of 2D projection-based vertebra segmentation for vertebra-level fracture detection in 3D CT volumes, presenting an end-to-end pipeline for automated analysis of cervical vertebrae (C1-C7). By approximating a 3D volume through optimized 2D axial, sagittal, and coronal projections, regions of interest are identified using the YOLOv8 model from all views and combined to approximate the 3D cervical spine area, achieving a 3D mIoU of 94.45 percent. This projection-based localization strategy reduces computational complexity compared to traditional 3D segmentation methods while maintaining high performance. It is followed by a DenseNet121-Unet-based multi-label segmentation leveraging variance- and energy-based projections, achieving a Dice score of 87.86 percent. Strategic approximation of 3D vertebral masks from these 2D segmentation masks enables the extraction of individual vertebra volumes. The volumes are analyzed for fractures using an ensemble of 2.5D Spatio-Sequential models incorporating both raw slices and projections per vertebra for complementary evaluation. This ensemble achieves vertebra-level and patient-level F1 scores of 68.15 and 82.26, and ROC-AUC scores of 91.62 and 83.04, respectively. We further validate our approach through an explainability study that provides saliency map visualizations highlighting anatomical regions relevant for diagnosis, and an interobserver variability analysis comparing our model's performance with expert radiologists, demonstrating competitive results.

Problem

• cervical spine fracture
• 3D CT
• vertebra segmentation
• fracture detection
• medical image analysis

Innovation

• 2D projection-driven
• multi-stage segmentation
• 2.5D spatio-sequential modeling
• vertebra-level fracture detection
• explainable AI