Automatic Cobb Angle Detection using Vertebra Detector and Vertebra Corners Regression

TL;DR

This paper introduces an automatic method for estimating Cobb Angles in spinal X-ray images by detecting vertebrae and their corners, aiming to improve accuracy and reduce variability in scoliosis assessment.

Contribution

A novel framework combining vertebra detection and landmark regression for automatic Cobb Angle measurement in spinal X-rays.

Findings

Achieved a SMAPE score of 25.69 in MICCAI 2019 challenge

Demonstrated improved consistency over manual estimation

Effective in reducing inter-rater variability

Abstract

Correct evaluation and treatment of Scoliosis require accurate estimation of spinal curvature. Current gold standard is to manually estimate Cobb Angles in spinal X-ray images which is time consuming and has high inter-rater variability. We propose an automatic method with a novel framework that first detects vertebrae as objects followed by a landmark detector that estimates the 4 landmark corners of each vertebra separately. Cobb Angles are calculated using the slope of each vertebra obtained from the predicted landmarks. For inference on test data, we perform pre and post processings that include cropping, outlier rejection and smoothing of the predicted landmarks. The results were assessed in AASCE MICCAI challenge 2019 which showed a promise with a SMAPE score of 25.69 on the challenge test set.