Publication detail

White Blood Cell Segmentation Using Fully Convolutional Neural Networks

HESKO, B. HARABIŠ, V. KRÁLÍK, M.

Original Title

White Blood Cell Segmentation Using Fully Convolutional Neural Networks

English Title

White Blood Cell Segmentation Using Fully Convolutional Neural Networks

Type

journal article - other

Language

en

Original Abstract

In medicine, the identification and counting of white blood cells are used for diagnosing diseases like inflammation, malignancy or leukaemia. In this paper, we propose a novel approach to white blood cell segmentation. On two different white blood cell datasets, two networks, PSPNet and U-Net are trained to perform simultaneous nucleus and cytoplasm segmentation. Compared to ground truth, our segmentations are almost identical, with smoother borders. When comparing overall cell segmentation with current methods, our networks are achieving similar (or better) results in evaluated metrics, with intersection over union reaching around 0.95 for both networks. DICE coefficient is higher than 0.96 for both networks and both datasets, which is a promising result of the segmentation.

English abstract

In medicine, the identification and counting of white blood cells are used for diagnosing diseases like inflammation, malignancy or leukaemia. In this paper, we propose a novel approach to white blood cell segmentation. On two different white blood cell datasets, two networks, PSPNet and U-Net are trained to perform simultaneous nucleus and cytoplasm segmentation. Compared to ground truth, our segmentations are almost identical, with smoother borders. When comparing overall cell segmentation with current methods, our networks are achieving similar (or better) results in evaluated metrics, with intersection over union reaching around 0.95 for both networks. DICE coefficient is higher than 0.96 for both networks and both datasets, which is a promising result of the segmentation.

Keywords

White blood cell, segmentation, deep learning, convolutional neural networks

Released

31.10.2018

Pages from

1

Pages to

9

Pages count

8

BibTex


@article{BUT150871,
  author="Branislav {Hesko} and Vratislav {Harabiš} and Martin {Králík}",
  title="White Blood Cell Segmentation Using Fully Convolutional Neural Networks",
  annote="In medicine, the identification and counting of white blood cells are used for diagnosing diseases like inflammation, malignancy or leukaemia. In this paper, we propose a novel approach to white blood cell segmentation. On two different white blood cell datasets, two networks, PSPNet and U-Net are trained to perform simultaneous nucleus and cytoplasm segmentation. Compared to ground truth, our segmentations are almost identical, with smoother borders. When comparing overall cell segmentation with current methods, our networks are achieving similar (or better) results in evaluated metrics, with intersection over union reaching around 0.95 for both networks. DICE coefficient is higher than 0.96 for both networks and both datasets, which is a promising result of the segmentation.",
  chapter="150871",
  howpublished="online",
  number="5",
  volume="20",
  year="2018",
  month="october",
  pages="1--9",
  type="journal article - other"
}