Bipartite Distance for Shape-Aware Landmark Detection in Spinal X-Ray Images
This addresses the need for reliable, automated scoliosis diagnosis to reduce manual effort and variability in medical imaging.
The paper tackled the problem of automating scoliosis assessment by detecting spinal landmarks in X-ray images, proposing a bipartite distance loss that consistently improved detection performance.
Scoliosis is a congenital disease that causes lateral curvature in the spine. Its assessment relies on the identification and localization of vertebrae in spinal X-ray images, conventionally via tedious and time-consuming manual radiographic procedures that are prone to subjectivity and observational variability. Reliability can be improved through the automatic detection and localization of spinal landmarks. To guide a CNN in the learning of spinal shape while detecting landmarks in X-ray images, we propose a novel loss based on a bipartite distance (BPD) measure, and show that it consistently improves landmark detection performance.