VERIDAH: Solving Enumeration Anomaly Aware Vertebra Labeling across Imaging Sequences

2026-01-20

AI Summary

This study addresses the clinical challenge of automatically identifying abnormal vertebral counts in the thoracolumbar region, which critically impacts diagnosis and surgical planning. The authors propose a novel deep learning approach that integrates a multi-classification head neural network with weighted vertebral sequence prediction, enabling, for the first time, automated detection of variations in thoracic and lumbar vertebrae counts across arbitrary field-of-view T2-weighted MRI and CT images. The method achieves remarkable performance, with full vertebral labeling accuracies of 98.30% on MRI and 99.18% on CT in cross-modality evaluations. It also demonstrates high accuracy in detecting thoracic anomalies (87.80%–96.30%) and lumbar anomalies (94.48%–97.22%), substantially outperforming existing techniques.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

The human spine commonly consists of seven cervical, twelve thoracic, and five lumbar vertebrae. However, enumeration anomalies may result in individuals having eleven or thirteen thoracic vertebrae and four or six lumbar vertebrae. Although the identification of enumeration anomalies has potential clinical implications for chronic back pain and operation planning, the thoracolumbar junction is often poorly assessed and rarely described in clinical reports. Additionally, even though multiple deep-learning-based vertebra labeling algorithms exist, there is a lack of methods to automatically label enumeration anomalies. Our work closes that gap by introducing"Vertebra Identification with Anomaly Handling"(VERIDAH), a novel vertebra labeling algorithm based on multiple classification heads combined with a weighted vertebra sequence prediction algorithm. We show that our approach surpasses existing models on T2w TSE sagittal (98.30% vs. 94.24% of subjects with all vertebrae correctly labeled, p<0.001) and CT imaging (99.18% vs. 77.26% of subjects with all vertebrae correctly labeled, p<0.001) and works in arbitrary field-of-view images. VERIDAH correctly labeled the presence 2 M\"oller et al. of thoracic enumeration anomalies in 87.80% and 96.30% of T2w and CT images, respectively, and lumbar enumeration anomalies in 94.48% and 97.22% for T2w and CT, respectively. Our code and models are available at: https://github.com/Hendrik-code/spineps.

Problem

• vertebra labeling
• enumeration anomaly
• thoracolumbar junction
• spine imaging
• anomaly detection

Innovation

• vertebra labeling
• enumeration anomaly
• deep learning
• multi-sequence imaging
• spinal segmentation