Automatic localization of the left ventricular blood pool centroid in short axis cardiac cine MR images

Li Kuo Tan, Yih Miin Liew, Einly Lim, Yang Faridah Abdul Aziz, Kok Han Chee, Robert A. McLaughlin

    Research output: Contribution to journalArticle

    1 Citation (Scopus)

    Abstract

    In this paper, we develop and validate an open source, fully automatic algorithm to localize the left ventricular (LV) blood pool centroid in short axis cardiac cine MR images, enabling follow-on automated LV segmentation algorithms. The algorithm comprises four steps: (i) quantify motion to determine an initial region of interest surrounding the heart, (ii) identify potential 2D objects of interest using an intensity-based segmentation, (iii) assess contraction/expansion, circularity, and proximity to lung tissue to score all objects of interest in terms of their likelihood of constituting part of the LV, and (iv) aggregate the objects into connected groups and construct the final LV blood pool volume and centroid. This algorithm was tested against 1140 datasets from the Kaggle Second Annual Data Science Bowl, as well as 45 datasets from the STACOM 2009 Cardiac MR Left Ventricle Segmentation Challenge. Correct LV localization was confirmed in 97.3% of the datasets. The mean absolute error between the gold standard and localization centroids was 2.8 to 4.7 mm, or 12 to 22% of the average endocardial radius. [Figure not available: see fulltext.]

    Original languageEnglish
    Pages (from-to)1053-1062
    Number of pages10
    JournalMedical and Biological Engineering and Computing
    Volume56
    Issue number6
    Early online date17 Nov 2017
    DOIs
    Publication statusPublished - Jun 2018

    Fingerprint

    Blood
    Tissue

    Cite this

    Tan, Li Kuo ; Liew, Yih Miin ; Lim, Einly ; Abdul Aziz, Yang Faridah ; Chee, Kok Han ; McLaughlin, Robert A. / Automatic localization of the left ventricular blood pool centroid in short axis cardiac cine MR images. In: Medical and Biological Engineering and Computing. 2018 ; Vol. 56, No. 6. pp. 1053-1062.
    @article{05c374a18b784aceae136bb9a13cbf80,
    title = "Automatic localization of the left ventricular blood pool centroid in short axis cardiac cine MR images",
    abstract = "In this paper, we develop and validate an open source, fully automatic algorithm to localize the left ventricular (LV) blood pool centroid in short axis cardiac cine MR images, enabling follow-on automated LV segmentation algorithms. The algorithm comprises four steps: (i) quantify motion to determine an initial region of interest surrounding the heart, (ii) identify potential 2D objects of interest using an intensity-based segmentation, (iii) assess contraction/expansion, circularity, and proximity to lung tissue to score all objects of interest in terms of their likelihood of constituting part of the LV, and (iv) aggregate the objects into connected groups and construct the final LV blood pool volume and centroid. This algorithm was tested against 1140 datasets from the Kaggle Second Annual Data Science Bowl, as well as 45 datasets from the STACOM 2009 Cardiac MR Left Ventricle Segmentation Challenge. Correct LV localization was confirmed in 97.3{\%} of the datasets. The mean absolute error between the gold standard and localization centroids was 2.8 to 4.7 mm, or 12 to 22{\%} of the average endocardial radius. [Figure not available: see fulltext.]",
    keywords = "Automatic localization, Cardiovascular MRI, Cine MRI, Left ventricle, Segmentation",
    author = "Tan, {Li Kuo} and Liew, {Yih Miin} and Einly Lim and {Abdul Aziz}, {Yang Faridah} and Chee, {Kok Han} and McLaughlin, {Robert A.}",
    year = "2018",
    month = "6",
    doi = "10.1007/s11517-017-1750-7",
    language = "English",
    volume = "56",
    pages = "1053--1062",
    journal = "MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING",
    issn = "0140-0118",
    publisher = "Springer-Verlag London Ltd.",
    number = "6",

    }

    Automatic localization of the left ventricular blood pool centroid in short axis cardiac cine MR images. / Tan, Li Kuo; Liew, Yih Miin; Lim, Einly; Abdul Aziz, Yang Faridah; Chee, Kok Han; McLaughlin, Robert A.

    In: Medical and Biological Engineering and Computing, Vol. 56, No. 6, 06.2018, p. 1053-1062.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Automatic localization of the left ventricular blood pool centroid in short axis cardiac cine MR images

    AU - Tan, Li Kuo

    AU - Liew, Yih Miin

    AU - Lim, Einly

    AU - Abdul Aziz, Yang Faridah

    AU - Chee, Kok Han

    AU - McLaughlin, Robert A.

    PY - 2018/6

    Y1 - 2018/6

    N2 - In this paper, we develop and validate an open source, fully automatic algorithm to localize the left ventricular (LV) blood pool centroid in short axis cardiac cine MR images, enabling follow-on automated LV segmentation algorithms. The algorithm comprises four steps: (i) quantify motion to determine an initial region of interest surrounding the heart, (ii) identify potential 2D objects of interest using an intensity-based segmentation, (iii) assess contraction/expansion, circularity, and proximity to lung tissue to score all objects of interest in terms of their likelihood of constituting part of the LV, and (iv) aggregate the objects into connected groups and construct the final LV blood pool volume and centroid. This algorithm was tested against 1140 datasets from the Kaggle Second Annual Data Science Bowl, as well as 45 datasets from the STACOM 2009 Cardiac MR Left Ventricle Segmentation Challenge. Correct LV localization was confirmed in 97.3% of the datasets. The mean absolute error between the gold standard and localization centroids was 2.8 to 4.7 mm, or 12 to 22% of the average endocardial radius. [Figure not available: see fulltext.]

    AB - In this paper, we develop and validate an open source, fully automatic algorithm to localize the left ventricular (LV) blood pool centroid in short axis cardiac cine MR images, enabling follow-on automated LV segmentation algorithms. The algorithm comprises four steps: (i) quantify motion to determine an initial region of interest surrounding the heart, (ii) identify potential 2D objects of interest using an intensity-based segmentation, (iii) assess contraction/expansion, circularity, and proximity to lung tissue to score all objects of interest in terms of their likelihood of constituting part of the LV, and (iv) aggregate the objects into connected groups and construct the final LV blood pool volume and centroid. This algorithm was tested against 1140 datasets from the Kaggle Second Annual Data Science Bowl, as well as 45 datasets from the STACOM 2009 Cardiac MR Left Ventricle Segmentation Challenge. Correct LV localization was confirmed in 97.3% of the datasets. The mean absolute error between the gold standard and localization centroids was 2.8 to 4.7 mm, or 12 to 22% of the average endocardial radius. [Figure not available: see fulltext.]

    KW - Automatic localization

    KW - Cardiovascular MRI

    KW - Cine MRI

    KW - Left ventricle

    KW - Segmentation

    UR - http://www.scopus.com/inward/record.url?scp=85034231051&partnerID=8YFLogxK

    U2 - 10.1007/s11517-017-1750-7

    DO - 10.1007/s11517-017-1750-7

    M3 - Article

    VL - 56

    SP - 1053

    EP - 1062

    JO - MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING

    JF - MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING

    SN - 0140-0118

    IS - 6

    ER -