Identifying immature precursor cells in bone marrow pathological images based on distance transform and internal structures of cells

  • Heng Wu*
  • , Guitao Cao
  • , Ling Li
  • , Jun Shi
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Nowadays manual observation by microscopes is the major method to detect leukemia cells. However this method takes much effort of doctors and depends much on the doctors' experience. This paper proposes a new method to locate the immature precursor cells based on distance transform and internal structures of cells. At first, Otsu method is used to binarize the image; secondly, the image is processed through morphological operations; then an initial location of immature precursor cells in the bone marrow pathological images is done by distance transform operation; fourthly, the internal structure and the size of cells are taken into consideration to remove the wrong cells; finally, the target cells are further marked according to the smoothing feature of the image. Experiments show the method proposed in this paper has a good performance in identifying the immature precursor cells in bone marrow pathological images.

Original languageEnglish
Title of host publication2010 International Conference on Multimedia Technology, ICMT 2010
DOIs
StatePublished - 2010
Event2010 International Conference on Multimedia Technology, ICMT 2010 - Ningbo, China
Duration: 29 Oct 201031 Oct 2010

Publication series

Name2010 International Conference on Multimedia Technology, ICMT 2010

Conference

Conference2010 International Conference on Multimedia Technology, ICMT 2010
Country/TerritoryChina
CityNingbo
Period29/10/1031/10/10

Keywords

  • Distance transform
  • Immature precursor cells
  • Internal structure
  • Morphology
  • Pathological images of bone marrow

Fingerprint

Dive into the research topics of 'Identifying immature precursor cells in bone marrow pathological images based on distance transform and internal structures of cells'. Together they form a unique fingerprint.

Cite this