VertDetect: Fully End-to-End 3D Vertebral Instance Segmentation Model

TL;DR

VertDetect is an end-to-end 3D CNN model that accurately detects and segments vertebrae in CT scans, improving spine surgery and radiation therapy planning, especially in challenging imaging scenarios.

Contribution

This paper introduces VertDetect, a novel fully automated 3D vertebral segmentation model combining CNN and GCN to enhance vertebral labeling and segmentation accuracy.

Findings

Achieved state-of-the-art Dice scores on VerSe datasets.

Effectively handles incomplete spine imaging and anatomical variations.

Utilizes shared CNN backbone for detection and segmentation.

Abstract

Vertebral detection and segmentation are critical steps for treatment planning in spine surgery and radiation therapy. Accurate identification and segmentation are complicated in imaging that does not include the full spine, in cases with variations in anatomy (T13 and/or L6 vertebrae), and in the presence of fracture or hardware. This paper proposes VertDetect, a fully automated end-to-end 3D vertebral instance segmentation Convolutional Neural Network (CNN) model to predict vertebral level labels and segmentations for all vertebrae present in a CT scan. The utilization of a shared CNN backbone provides the detection and segmentation branches of the network with feature maps containing both spinal and vertebral level information. A Graph Convolutional Network (GCN) layer is used to improve vertebral labelling by using the known structure of the spine. This model achieved a Dice Similarity Coefficient (DSC) of 0.883 (95% CI, 0.843-0.906) and 0.882 (95% CI, 0.835-0.909) in the VerSe 2019 and 0.868 (95\% CI, 0.834-0.890) and 0.869 (95\% CI, 0.832-0.891) in the VerSe 2020 public and hidden test sets, respectively. This model achieved state-of-the-art performance for an end-to-end architecture, whose design facilitates the extraction of features that can be subsequently used for downstream tasks.