Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Segmentation of White Blood Cells Using Deep Learning

Grimmeiss Grahm, Sophia LU and Nilsson, Desiré LU (2019) In Master's Theses in Mathematical Sciences FMAM05 20192
Mathematics (Faculty of Engineering)
Abstract (Swedish)
The white blood cell count and differential is an important part of diagnosing a number of medical conditions. Instead of doing this by manual microscopy, CellaVision’s technology has automated the process of finding and classifying white blood cells. To support a diagnosis it is desired that the system can produce cytoplasm-to-nucleus-ratio. This ratio is calculated from a segmented image where the pixels are labelled as background, cytoplasm, or nucleus. The system used today, using active contours, does not always produce perfect segmentations for all cells, and it would therefore be beneficial to improve the segmentation. Using machine learning, we have constructed a network for segmenting white blood cell images. This model, with some... (More)
The white blood cell count and differential is an important part of diagnosing a number of medical conditions. Instead of doing this by manual microscopy, CellaVision’s technology has automated the process of finding and classifying white blood cells. To support a diagnosis it is desired that the system can produce cytoplasm-to-nucleus-ratio. This ratio is calculated from a segmented image where the pixels are labelled as background, cytoplasm, or nucleus. The system used today, using active contours, does not always produce perfect segmentations for all cells, and it would therefore be beneficial to improve the segmentation. Using machine learning, we have constructed a network for segmenting white blood cell images. This model, with some small modifications, produces both binary (cell and background) and multi-class (cytoplasm, nucleus and background) segmentations. The model is a U-net inspired by work previously done on other similar segmentation tasks. The network reached an IoU of 93.9% in the binary case, and in the muli-class case 82.8% and 94.5% for the cytoplasm and nucleus respectively. The main challenges were to separate neighbouring cells and cells in a cluster.

Over all the network performed better than the active contour method in difficult images, and in cases where neither were good, the network was usually better. If the network was trained more on images that are difficult to segment, the resulting segmentations of these images could be improved. (Less)
Please use this url to cite or link to this publication:
author
Grimmeiss Grahm, Sophia LU and Nilsson, Desiré LU
supervisor
organization
alternative title
Segmentering av vita blodceller med hjälp av maskininlärning
course
FMAM05 20192
year
type
H2 - Master's Degree (Two Years)
subject
publication/series
Master's Theses in Mathematical Sciences
report number
LUFTMA-3395-2019
ISSN
1404-6342
other publication id
2019:E64
language
English
id
8998594
date added to LUP
2020-01-23 13:24:01
date last changed
2020-01-23 13:24:01
@misc{8998594,
  abstract     = {{The white blood cell count and differential is an important part of diagnosing a number of medical conditions. Instead of doing this by manual microscopy, CellaVision’s technology has automated the process of finding and classifying white blood cells. To support a diagnosis it is desired that the system can produce cytoplasm-to-nucleus-ratio. This ratio is calculated from a segmented image where the pixels are labelled as background, cytoplasm, or nucleus. The system used today, using active contours, does not always produce perfect segmentations for all cells, and it would therefore be beneficial to improve the segmentation. Using machine learning, we have constructed a network for segmenting white blood cell images. This model, with some small modifications, produces both binary (cell and background) and multi-class (cytoplasm, nucleus and background) segmentations. The model is a U-net inspired by work previously done on other similar segmentation tasks. The network reached an IoU of 93.9% in the binary case, and in the muli-class case 82.8% and 94.5% for the cytoplasm and nucleus respectively. The main challenges were to separate neighbouring cells and cells in a cluster.

Over all the network performed better than the active contour method in difficult images, and in cases where neither were good, the network was usually better. If the network was trained more on images that are difficult to segment, the resulting segmentations of these images could be improved.}},
  author       = {{Grimmeiss Grahm, Sophia and Nilsson, Desiré}},
  issn         = {{1404-6342}},
  language     = {{eng}},
  note         = {{Student Paper}},
  series       = {{Master's Theses in Mathematical Sciences}},
  title        = {{Segmentation of White Blood Cells Using Deep Learning}},
  year         = {{2019}},
}