Iterative fully convolutional neural networks for automatic vertebra segmentation and identification

• Published in: Medical image analysis Vol. 53; pp. 142 - 155
• Main Authors: Lessmann, Nikolas, van Ginneken, Bram, de Jong, Pim A., Išgum, Ivana
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01-04-2019; Elsevier BV
• Subjects: Abnormalities; Artificial neural networks; Chest; Compression; Computed tomography; Deep learning; Fractures; Fractures, Compression - diagnostic imaging; Humans; Identification; Image Interpretation, Computer-Assisted - methods; Image processing; Image segmentation; Iterative instance segmentation; Labels; Magnetic Resonance Imaging; Neural networks; Neural Networks, Computer; Predictions; Regression analysis; Spinal Fractures - diagnostic imaging; Spine; Tomography, X-Ray Computed; Vertebra identification; Vertebra segmentation; Vertebrae; Deep learning; Vertebra segmentation; Iterative instance segmentation; Vertebra identification
• ISSN: 1361-8415; 1361-8423
• DOI: 10.1016/j.media.2019.02.005

•Vertebrae are segmented with an iterative instance segmentation algorithm.•The method does not make assumptions about the number of visible vertebrae.•Detected vertebrae are anatomically labeled using a global probabilistic model.•A fully convolutional neural network performs both segmentation and identification.•Vertebra segmentations and identifications are evaluated on five CT and MR datasets.
[Display omitted]
Precise segmentation and anatomical identification of the vertebrae provides the basis for automatic analysis of the spine, such as detection of vertebral compression fractures or other abnormalities. Most dedicated spine CT and MR scans as well as scans of the chest, abdomen or neck cover only part of the spine. Segmentation and identification should therefore not rely on the visibility of certain vertebrae or a certain number of vertebrae. We propose an iterative instance segmentation approach that uses a fully convolutional neural network to segment and label vertebrae one after the other, independently of the number of visible vertebrae. This instance-by-instance segmentation is enabled by combining the network with a memory component that retains information about already segmented vertebrae. The network iteratively analyzes image patches, using information from both image and memory to search for the next vertebra. To efficiently traverse the image, we include the prior knowledge that the vertebrae are always located next to each other, which is used to follow the vertebral column. The network concurrently performs multiple tasks, which are segmentation of a vertebra, regression of its anatomical label and prediction whether the vertebra is completely visible in the image, which allows to exclude incompletely visible vertebrae from further analyses. The predicted anatomical labels of the individual vertebrae are additionally refined with a maximum likelihood approach, choosing the overall most likely labeling if all detected vertebrae are taken into account. This method was evaluated with five diverse datasets, including multiple modalities (CT and MR), various fields of view and coverages of different sections of the spine, and a particularly challenging set of low-dose chest CT scans. For vertebra segmentation, the average Dice score was 94.9 ± 2.1% with an average absolute symmetric surface distance of 0.2 ± 10.1mm. The anatomical identification had an accuracy of 93%, corresponding to a single case with mislabeled vertebrae. Vertebrae were classified as completely or incompletely visible with an accuracy of 97%. The proposed iterative segmentation method compares favorably with state-of-the-art methods and is fast, flexible and generalizable.