A Localisation-Segmentation Approach for Multi-label Annotation of Lumbar Vertebrae using Deep Nets
This work addresses the problem of accurate vertebrae segmentation for medical imaging applications, particularly in cases with abnormalities like fractures and scoliosis, but it is incremental as it builds on existing deep learning methods.
The authors tackled multi-class segmentation of lumbar vertebrae in CT scans by proposing a two-stage deep learning approach that first localizes the lumbar region and then segments and labels the vertebrae, achieving an average Dice coefficient of over 90% on a challenging public dataset with severe deformities.
Multi-class segmentation of vertebrae is a non-trivial task mainly due to the high correlation in the appearance of adjacent vertebrae. Hence, such a task calls for the consideration of both global and local context. Based on this motivation, we propose a two-staged approach that, given a computed tomography dataset of the spine, segments the five lumbar vertebrae and simultaneously labels them. The first stage employs a multi-layered perceptron performing non-linear regression for locating the lumbar region using the global context. The second stage, comprised of a fully-convolutional deep network, exploits the local context in the localised lumbar region to segment and label the lumbar vertebrae in one go. Aided with practical data augmentation for training, our approach is highly generalisable, capable of successfully segmenting both healthy and abnormal vertebrae (fractured and scoliotic spines). We consistently achieve an average Dice coefficient of over 90 percent on a publicly available dataset of the xVertSeg segmentation challenge of MICCAI 2016. This is particularly noteworthy because the xVertSeg dataset is beset with severe deformities in the form of vertebral fractures and scoliosis.