Comparison of traditional image processing and deep learning approaches for classification of white blood cells in peripheral blood smear images

Roopa B. Hegde, Keerthana Prasad, Harishchandra Hebbar, Brij Mohan Kumar Singh

    Research output: Contribution to journalArticle

    Abstract

    Automated classification and morphological analysis of white blood cells has been addressed since last four decades, but there is no optimal method which can be used as decision support system in laboratories due to biologically complex nature of the cells. Automated blood cell analysis facilitates quick and objective results and can also handle massive amount of data without compromising with efficiency. In the present study, we demonstrate classification of white blood cells into six types namely lymphocytes, monocytes, neutrophils, eosinophils, basophils and abnormal cells. We provide the comparison of traditional image processing approach and deep learning methods for classification of white blood cells. We evaluated neural network classifier results for hand-crafted features and obtained the average accuracy of 99.8%. We also used full training and transfer learning approaches of convolutional neural network for the classification. An accuracy around 99% was obtained for full training CNN.

    Original languageEnglish
    Pages (from-to)382-392
    Number of pages11
    JournalBiocybernetics and Biomedical Engineering
    Volume39
    Issue number2
    DOIs
    Publication statusPublished - 01-04-2019

    Fingerprint

    Image processing
    Blood
    Cells
    Neural networks
    Lymphocytes
    Decision support systems
    Classifiers
    Deep learning

    All Science Journal Classification (ASJC) codes

    • Biomedical Engineering

    Cite this

    @article{e314fa32f5e941119a6b4a66ec727a8c,
    title = "Comparison of traditional image processing and deep learning approaches for classification of white blood cells in peripheral blood smear images",
    abstract = "Automated classification and morphological analysis of white blood cells has been addressed since last four decades, but there is no optimal method which can be used as decision support system in laboratories due to biologically complex nature of the cells. Automated blood cell analysis facilitates quick and objective results and can also handle massive amount of data without compromising with efficiency. In the present study, we demonstrate classification of white blood cells into six types namely lymphocytes, monocytes, neutrophils, eosinophils, basophils and abnormal cells. We provide the comparison of traditional image processing approach and deep learning methods for classification of white blood cells. We evaluated neural network classifier results for hand-crafted features and obtained the average accuracy of 99.8{\%}. We also used full training and transfer learning approaches of convolutional neural network for the classification. An accuracy around 99{\%} was obtained for full training CNN.",
    author = "Hegde, {Roopa B.} and Keerthana Prasad and Harishchandra Hebbar and Singh, {Brij Mohan Kumar}",
    year = "2019",
    month = "4",
    day = "1",
    doi = "10.1016/j.bbe.2019.01.005",
    language = "English",
    volume = "39",
    pages = "382--392",
    journal = "Biocybernetics and Biomedical Engineering",
    issn = "0208-5216",
    publisher = "PWN-Polish Scientific Publishers",
    number = "2",

    }

    Comparison of traditional image processing and deep learning approaches for classification of white blood cells in peripheral blood smear images. / Hegde, Roopa B.; Prasad, Keerthana; Hebbar, Harishchandra; Singh, Brij Mohan Kumar.

    In: Biocybernetics and Biomedical Engineering, Vol. 39, No. 2, 01.04.2019, p. 382-392.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Comparison of traditional image processing and deep learning approaches for classification of white blood cells in peripheral blood smear images

    AU - Hegde, Roopa B.

    AU - Prasad, Keerthana

    AU - Hebbar, Harishchandra

    AU - Singh, Brij Mohan Kumar

    PY - 2019/4/1

    Y1 - 2019/4/1

    N2 - Automated classification and morphological analysis of white blood cells has been addressed since last four decades, but there is no optimal method which can be used as decision support system in laboratories due to biologically complex nature of the cells. Automated blood cell analysis facilitates quick and objective results and can also handle massive amount of data without compromising with efficiency. In the present study, we demonstrate classification of white blood cells into six types namely lymphocytes, monocytes, neutrophils, eosinophils, basophils and abnormal cells. We provide the comparison of traditional image processing approach and deep learning methods for classification of white blood cells. We evaluated neural network classifier results for hand-crafted features and obtained the average accuracy of 99.8%. We also used full training and transfer learning approaches of convolutional neural network for the classification. An accuracy around 99% was obtained for full training CNN.

    AB - Automated classification and morphological analysis of white blood cells has been addressed since last four decades, but there is no optimal method which can be used as decision support system in laboratories due to biologically complex nature of the cells. Automated blood cell analysis facilitates quick and objective results and can also handle massive amount of data without compromising with efficiency. In the present study, we demonstrate classification of white blood cells into six types namely lymphocytes, monocytes, neutrophils, eosinophils, basophils and abnormal cells. We provide the comparison of traditional image processing approach and deep learning methods for classification of white blood cells. We evaluated neural network classifier results for hand-crafted features and obtained the average accuracy of 99.8%. We also used full training and transfer learning approaches of convolutional neural network for the classification. An accuracy around 99% was obtained for full training CNN.

    UR - http://www.scopus.com/inward/record.url?scp=85063330344&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85063330344&partnerID=8YFLogxK

    U2 - 10.1016/j.bbe.2019.01.005

    DO - 10.1016/j.bbe.2019.01.005

    M3 - Article

    VL - 39

    SP - 382

    EP - 392

    JO - Biocybernetics and Biomedical Engineering

    JF - Biocybernetics and Biomedical Engineering

    SN - 0208-5216

    IS - 2

    ER -