Two-dimensional ARMA modeling for breast cancer detection and classification

Jerzy Zielinski, Nidhal Bouaynaya, Dan Schonfeld

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    9 Citations (Scopus)

    Abstract

    Computer aided diagnosis (CAD) paradigms have gained currency for discriminating malignant from benign lesions in ultrasound breast images. But even the most sophisticated investigators often rely on one-dimensional representations of the image in terms of its scanlines. Such vector representations are convenient because of the mathematical tractability of one-dimensional time-series. However, they fail to take into account the spatial correlations between the pixels, which is crucial in tumor detection and classification in breast images. In this paper, we propose a CAD system for tumor detection and classification (cancerous v.s. benign) in ultrasound breast images based on a two-dimensional Auto-Regressive-Moving-Average (ARMA) model of the breast image. First, we show, using the Wold decomposition theorem, that ultrasound breast images can be accurately modeled by two-dimensional ARMA random fields. As in the 1D case, the 2D ARMA parameter estimation problem is much more difficult than its 2D AR counterpart, due to the non-linearity in estimating the 2D moving average (MA) parameters. We propose to estimate the 2D ARMA parameters using a two-stage Yule-Walker Least-Squares algorithm. The estimated parameters are then used as the basis for statistical inference and biophysical interpretation of the breast image. We evaluate the performance of the 2D ARMA vector features in real ultrasound images using a k-means classifier. Our results suggest that the proposed CAD system based on a two-dimensional ARMA model leads to parameters that can accurately segment the ultrasound breast image into three regions: healthy tissue, benign tumor, and cancerous tumor. Moreover, the specificity and sensitivity of the proposed two-dimensional CAD system is superior to its one-dimensional homologue.

    Original languageEnglish (US)
    Title of host publication2010 International Conference on Signal Processing and Communications, SPCOM 2010
    DOIs
    StatePublished - Oct 29 2010
    Event2010 International Conference on Signal Processing and Communications, SPCOM 2010 - Bangalore, India
    Duration: Jul 18 2010Jul 21 2010

    Publication series

    Name2010 International Conference on Signal Processing and Communications, SPCOM 2010

    Other

    Other2010 International Conference on Signal Processing and Communications, SPCOM 2010
    CountryIndia
    CityBangalore
    Period7/18/107/21/10

    Fingerprint

    Computer aided diagnosis
    Ultrasonics
    Tumors
    Parameter estimation
    Time series
    Classifiers
    Pixels
    Tissue
    Decomposition

    All Science Journal Classification (ASJC) codes

    • Computer Networks and Communications
    • Signal Processing

    Cite this

    Zielinski, J., Bouaynaya, N., & Schonfeld, D. (2010). Two-dimensional ARMA modeling for breast cancer detection and classification. In 2010 International Conference on Signal Processing and Communications, SPCOM 2010 [5560514] (2010 International Conference on Signal Processing and Communications, SPCOM 2010). https://doi.org/10.1109/SPCOM.2010.5560514
    Zielinski, Jerzy ; Bouaynaya, Nidhal ; Schonfeld, Dan. / Two-dimensional ARMA modeling for breast cancer detection and classification. 2010 International Conference on Signal Processing and Communications, SPCOM 2010. 2010. (2010 International Conference on Signal Processing and Communications, SPCOM 2010).
    @inproceedings{b1e24839d72d4cfa9241c09763438024,
    title = "Two-dimensional ARMA modeling for breast cancer detection and classification",
    abstract = "Computer aided diagnosis (CAD) paradigms have gained currency for discriminating malignant from benign lesions in ultrasound breast images. But even the most sophisticated investigators often rely on one-dimensional representations of the image in terms of its scanlines. Such vector representations are convenient because of the mathematical tractability of one-dimensional time-series. However, they fail to take into account the spatial correlations between the pixels, which is crucial in tumor detection and classification in breast images. In this paper, we propose a CAD system for tumor detection and classification (cancerous v.s. benign) in ultrasound breast images based on a two-dimensional Auto-Regressive-Moving-Average (ARMA) model of the breast image. First, we show, using the Wold decomposition theorem, that ultrasound breast images can be accurately modeled by two-dimensional ARMA random fields. As in the 1D case, the 2D ARMA parameter estimation problem is much more difficult than its 2D AR counterpart, due to the non-linearity in estimating the 2D moving average (MA) parameters. We propose to estimate the 2D ARMA parameters using a two-stage Yule-Walker Least-Squares algorithm. The estimated parameters are then used as the basis for statistical inference and biophysical interpretation of the breast image. We evaluate the performance of the 2D ARMA vector features in real ultrasound images using a k-means classifier. Our results suggest that the proposed CAD system based on a two-dimensional ARMA model leads to parameters that can accurately segment the ultrasound breast image into three regions: healthy tissue, benign tumor, and cancerous tumor. Moreover, the specificity and sensitivity of the proposed two-dimensional CAD system is superior to its one-dimensional homologue.",
    author = "Jerzy Zielinski and Nidhal Bouaynaya and Dan Schonfeld",
    year = "2010",
    month = "10",
    day = "29",
    doi = "10.1109/SPCOM.2010.5560514",
    language = "English (US)",
    isbn = "9781424471362",
    series = "2010 International Conference on Signal Processing and Communications, SPCOM 2010",
    booktitle = "2010 International Conference on Signal Processing and Communications, SPCOM 2010",

    }

    Zielinski, J, Bouaynaya, N & Schonfeld, D 2010, Two-dimensional ARMA modeling for breast cancer detection and classification. in 2010 International Conference on Signal Processing and Communications, SPCOM 2010., 5560514, 2010 International Conference on Signal Processing and Communications, SPCOM 2010, 2010 International Conference on Signal Processing and Communications, SPCOM 2010, Bangalore, India, 7/18/10. https://doi.org/10.1109/SPCOM.2010.5560514

    Two-dimensional ARMA modeling for breast cancer detection and classification. / Zielinski, Jerzy; Bouaynaya, Nidhal; Schonfeld, Dan.

    2010 International Conference on Signal Processing and Communications, SPCOM 2010. 2010. 5560514 (2010 International Conference on Signal Processing and Communications, SPCOM 2010).

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    TY - GEN

    T1 - Two-dimensional ARMA modeling for breast cancer detection and classification

    AU - Zielinski, Jerzy

    AU - Bouaynaya, Nidhal

    AU - Schonfeld, Dan

    PY - 2010/10/29

    Y1 - 2010/10/29

    N2 - Computer aided diagnosis (CAD) paradigms have gained currency for discriminating malignant from benign lesions in ultrasound breast images. But even the most sophisticated investigators often rely on one-dimensional representations of the image in terms of its scanlines. Such vector representations are convenient because of the mathematical tractability of one-dimensional time-series. However, they fail to take into account the spatial correlations between the pixels, which is crucial in tumor detection and classification in breast images. In this paper, we propose a CAD system for tumor detection and classification (cancerous v.s. benign) in ultrasound breast images based on a two-dimensional Auto-Regressive-Moving-Average (ARMA) model of the breast image. First, we show, using the Wold decomposition theorem, that ultrasound breast images can be accurately modeled by two-dimensional ARMA random fields. As in the 1D case, the 2D ARMA parameter estimation problem is much more difficult than its 2D AR counterpart, due to the non-linearity in estimating the 2D moving average (MA) parameters. We propose to estimate the 2D ARMA parameters using a two-stage Yule-Walker Least-Squares algorithm. The estimated parameters are then used as the basis for statistical inference and biophysical interpretation of the breast image. We evaluate the performance of the 2D ARMA vector features in real ultrasound images using a k-means classifier. Our results suggest that the proposed CAD system based on a two-dimensional ARMA model leads to parameters that can accurately segment the ultrasound breast image into three regions: healthy tissue, benign tumor, and cancerous tumor. Moreover, the specificity and sensitivity of the proposed two-dimensional CAD system is superior to its one-dimensional homologue.

    AB - Computer aided diagnosis (CAD) paradigms have gained currency for discriminating malignant from benign lesions in ultrasound breast images. But even the most sophisticated investigators often rely on one-dimensional representations of the image in terms of its scanlines. Such vector representations are convenient because of the mathematical tractability of one-dimensional time-series. However, they fail to take into account the spatial correlations between the pixels, which is crucial in tumor detection and classification in breast images. In this paper, we propose a CAD system for tumor detection and classification (cancerous v.s. benign) in ultrasound breast images based on a two-dimensional Auto-Regressive-Moving-Average (ARMA) model of the breast image. First, we show, using the Wold decomposition theorem, that ultrasound breast images can be accurately modeled by two-dimensional ARMA random fields. As in the 1D case, the 2D ARMA parameter estimation problem is much more difficult than its 2D AR counterpart, due to the non-linearity in estimating the 2D moving average (MA) parameters. We propose to estimate the 2D ARMA parameters using a two-stage Yule-Walker Least-Squares algorithm. The estimated parameters are then used as the basis for statistical inference and biophysical interpretation of the breast image. We evaluate the performance of the 2D ARMA vector features in real ultrasound images using a k-means classifier. Our results suggest that the proposed CAD system based on a two-dimensional ARMA model leads to parameters that can accurately segment the ultrasound breast image into three regions: healthy tissue, benign tumor, and cancerous tumor. Moreover, the specificity and sensitivity of the proposed two-dimensional CAD system is superior to its one-dimensional homologue.

    UR - http://www.scopus.com/inward/record.url?scp=77958493825&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=77958493825&partnerID=8YFLogxK

    U2 - 10.1109/SPCOM.2010.5560514

    DO - 10.1109/SPCOM.2010.5560514

    M3 - Conference contribution

    AN - SCOPUS:77958493825

    SN - 9781424471362

    T3 - 2010 International Conference on Signal Processing and Communications, SPCOM 2010

    BT - 2010 International Conference on Signal Processing and Communications, SPCOM 2010

    ER -

    Zielinski J, Bouaynaya N, Schonfeld D. Two-dimensional ARMA modeling for breast cancer detection and classification. In 2010 International Conference on Signal Processing and Communications, SPCOM 2010. 2010. 5560514. (2010 International Conference on Signal Processing and Communications, SPCOM 2010). https://doi.org/10.1109/SPCOM.2010.5560514