Grayscale self-adjusting network with weak feature enhancement for 3D lumbar anatomy segmentation |
| |
| Institution: | 1. Department of Computer Science, University of North Carolina, Chapel Hill, NC 27599, USA;2. Department of Biomedical Engineering, University of North Carolina, Chapel Hill, NC 27599, USA;3. Department of Biomedical Engineering, University of Virginia, Charlottesville, VA 22904, USA;4. Department of Radiology, University of North Carolina, Chapel Hill, NC 27599, USA;5. Biomedical Research Imaging Center (BRIC), University of North Carolina, Chapel Hill, NC 27599, USA;1. School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai 200030, China;2. Department of Functional Neurosurgery, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai 200025, China;1. School of Software, Shandong University, China;2. Department of Computer Science, The University of Hong Kong, China;3. Texas A&M University, USA;1. College of Information and Computer, Taiyuan University of Technology, Taiyuan, Shanxi, China;2. Digital Imaging Group of London, London, ON, Canada;3. Department of Medical Imaging, Western University, London, ON, Canada;1. Medical Image Processing Group, 602 Goddard building, 3710 Hamilton Walk, Department of Radiology, University of Pennsylvania, Philadelphia, PA 19104, United States;2. Quantitative Radiology Solutions, LLC, 3675 Market Street, Suite 200, Philadelphia, PA 19104, United States |
| |
| Abstract: | The automatic segmentation of lumbar anatomy is a fundamental problem for the diagnosis and treatment of lumbar disease. The recent development of deep learning techniques has led to remarkable progress in this task, including the possible segmentation of nerve roots, intervertebral discs, and dural sac in a single step. Despite these advances, lumbar anatomy segmentation remains a challenging problem due to the weak contrast and noise of input images, as well as the variability of intensities and size in lumbar structures across different subjects. To overcome these challenges, we propose a coarse-to-fine deep neural network framework for lumbar anatomy segmentation, which obtains a more accurate segmentation using two strategies. First, a progressive refinement process is employed to correct low-confidence regions by enhancing the feature representation in these regions. Second, a grayscale self-adjusting network (GSA-Net) is proposed to optimize the distribution of intensities dynamically. Experiments on datasets comprised of 3D computed tomography (CT) and magnetic resonance (MR) images show the advantage of our method over current segmentation approaches and its potential for diagnosing and lumbar disease treatment. |
| |
| Keywords: | Medical image segmentation Lumbar anatomy segmentation Weak feature enhancement Grayscale self-adjusting |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
