An image processing approach to analyze morphological features of microscopic images of muscle fibers

Comput Med Imaging Graph. 2014 Dec;38(8):803-14. doi: 10.1016/j.compmedimag.2014.07.003. Epub 2014 Jul 31.

Abstract

We present an image processing approach to automatically analyze duo-channel microscopic images of muscular fiber nuclei and cytoplasm. Nuclei and cytoplasm play a critical role in determining the health and functioning of muscular fibers as changes of nuclei and cytoplasm manifest in many diseases such as muscular dystrophy and hypertrophy. Quantitative evaluation of muscle fiber nuclei and cytoplasm thus is of great importance to researchers in musculoskeletal studies. The proposed computational approach consists of steps of image processing to segment and delineate cytoplasm and identify nuclei in two-channel images. Morphological operations like skeletonization is applied to extract the length of cytoplasm for quantification. We tested the approach on real images and found that it can achieve high accuracy, objectivity, and robustness.

Keywords: Cytoplasm; Muscle fiber; Nuclei; Quantification; Segmentation; Skeletonization.

Publication types

  • Research Support, Non-U.S. Gov't
  • Research Support, U.S. Gov't, Non-P.H.S.

MeSH terms

  • Algorithms*
  • Animals
  • Cells, Cultured
  • Image Enhancement / methods*
  • Image Interpretation, Computer-Assisted / methods*
  • Male
  • Mice
  • Mice, Inbred C57BL
  • Microscopy / methods*
  • Muscle Fibers, Skeletal / cytology*
  • Pattern Recognition, Automated / methods*
  • Reproducibility of Results
  • Sensitivity and Specificity