Deep learning for automated segmentation of pelvic muscles, fat, and bone from CT studies for body composition assessment.

Journal: Skeletal radiology
Published Date: Aug 8, 2019

Abstract

OBJECTIVE: To develop a deep convolutional neural network (CNN) to automatically segment an axial CT image of the pelvis for body composition measures. We hypothesized that a deep CNN approach would achieve high accuracy when compared to manual segmentations as the reference standard.

Authors

  • Robert Hemke
    Division of Musculoskeletal Imaging and Intervention, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, 55 Fruit Street, YAW 6048, Boston, MA, 02114, USA.
  • Colleen G Buckless
    Division of Musculoskeletal Imaging and Intervention, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, 55 Fruit Street, YAW 6048, Boston, MA, 02114, USA.
  • Andrew Tsao
    Division of Musculoskeletal Imaging and Intervention, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, 55 Fruit Street, YAW 6048, Boston, MA, 02114, USA.
  • Benjamin Wang
    Department of Radiology, Division of Musculoskeletal Radiology, NYU Langone Health, 301 E 17th St, 6th Floor, New York, NY, 10003 (B.W., C.B., R.S.A.); and Department of Musculoskeletal Imaging, Hôpital Lariboisière, Paris, France (L.P.).
  • Martin Torriani
    Division of Musculoskeletal Imaging and Intervention, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, 55 Fruit Street, YAW 6048, Boston, MA, 02114, USA. mtorriani@mgh.harvard.edu.

Keywords

External Resources

Popular Topics
Recent Journals