Automatic Cobb Angle Detection using Vertebra Detector and Vertebra Corners Regression
Published in Computational Methods and Clinical Applications for Spine Imaging, 2020
Abstract: Correct evaluation and treatment of Scoliosis require accu- rate 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 as- sessed in AASCE MICCAI challenge 2019 which showed a promise with a SMAPE score of 25.69 on the challenge test set.
Bidur Khanal, Lavsen Dahal, Prashant Adhikari, Bishesh Khanal.Automatic Cobb Angle Detection using Vertebra Detector and Vertebra Corners Regression. In MICCAI 2019 Challenge on Accurate Automated Spinal Curvature Estimation & Workshop on Computational Methods and Clinical Applications for Spine Imaging.