Convolutional Neural Networks based automated segmentation and labelling of the lumbar spine X-ray

TL;DR

This study compares instance and semantic segmentation neural networks for lumbar spine X-ray analysis, finding that instance segmentation models offer slightly better accuracy and easier integration into clinical workflows.

Contribution

It provides a comparative analysis of segmentation network types on a large dataset of lumbar spine X-rays, highlighting the advantages of instance segmentation for clinical applications.

Findings

Instance segmentation achieved up to 3% higher mean accuracy.

Semantic segmentation had slightly better pixel accuracy and IoU.

Instance segmentation models are easier to implement in clinical pipelines.

Abstract

The aim of this study is to investigate the segmentation accuracies of different segmentation networks trained on 730 manually annotated lateral lumbar spine X-rays. Instance segmentation networks were compared to semantic segmentation networks. The study cohort comprised diseased spines and postoperative images with metallic implants. The average mean accuracy and mean intersection over union (IoU) was up to 3 percent better for the best performing instance segmentation model, the average pixel accuracy and weighted IoU were slightly better for the best performing semantic segmentation model. Moreover, the inferences of the instance segmentation models are easier to implement for further processing pipelines in clinical decision support.