Toward patient-specific estimation of hepatic tumors respiratory motion: A finite element-based machine learning approach

    Research output: Contribution to conferencePaper

    Abstract

    Purpose
    To develop and evaluate an approach to estimate the respiratory-induced motion of hepatic lesions.

    Materials and methods
    In this study, a correlation between the collected surrogate signals and the liver tumor respiratory motion is obtained using learning-based algorithms. A robotic phantom is developed which simulates the respiratory motion of the liver, the diaphragm, and the abdomen skin in two directions as superior-inferior (SI) and anterior-posterior (AP). The surrogate signals are collected by means of optical markers attached to the abdomen skin and tracked by a digital camera, in addition to an inertial measurement unit (IMU) fixed to the hub of a needle that is inserted into the tissue. A finite element model (FEM) is developed to study the effect of tissue and tumor location parameters on target motion.

    Results
    The estimation error of linear regression for the SI and AP directions has been respectively 1.37% and 2.87%, and for quadratic polynomial regression have been 0.76% and 2.41% on the data from the experiments.

    Conclusion
    Using more than one surrogate signals resulted in a about 0.5-6.5% decrease of the motion estimation error compared to using only one of the surrogates.

    Keywords: hepatic tumor, motion estimation, respiratory motion, machine learning, robotic phantom, finite element analysis
    Original languageEnglish
    Publication statusPublished - 2019
    Event33rd International Conference on Computer Assisted Radiology and Surgery 2019 - Rennes Conference Center, Rennes, France
    Duration: 17 Jun 201921 Jun 2019
    Conference number: 33
    https://www.cars2019.org/

    Conference

    Conference33rd International Conference on Computer Assisted Radiology and Surgery 2019
    Abbreviated titleCARS 2019
    CountryFrance
    CityRennes
    Period17/06/1921/06/19
    Internet address

    Fingerprint

    Learning systems
    Tumors
    Motion estimation
    Liver
    Skin
    Robotics
    Tissue
    Units of measurement
    Digital cameras
    Diaphragms
    Linear regression
    Needles
    Error analysis
    Polynomials
    Finite element method
    Experiments

    Cite this

    Berijanian, M., Naghibi Beidokhti, H., Sirmaçek, B., & Abayazid, M. (2019). Toward patient-specific estimation of hepatic tumors respiratory motion: A finite element-based machine learning approach. Paper presented at 33rd International Conference on Computer Assisted Radiology and Surgery 2019, Rennes, France.
    Berijanian, Maryam ; Naghibi Beidokhti, Hamid ; Sirmaçek, Beril ; Abayazid, Momen . / Toward patient-specific estimation of hepatic tumors respiratory motion: A finite element-based machine learning approach. Paper presented at 33rd International Conference on Computer Assisted Radiology and Surgery 2019, Rennes, France.
    @conference{95e80c81d3954147b89b58143612b051,
    title = "Toward patient-specific estimation of hepatic tumors respiratory motion: A finite element-based machine learning approach",
    abstract = "PurposeTo develop and evaluate an approach to estimate the respiratory-induced motion of hepatic lesions.Materials and methodsIn this study, a correlation between the collected surrogate signals and the liver tumor respiratory motion is obtained using learning-based algorithms. A robotic phantom is developed which simulates the respiratory motion of the liver, the diaphragm, and the abdomen skin in two directions as superior-inferior (SI) and anterior-posterior (AP). The surrogate signals are collected by means of optical markers attached to the abdomen skin and tracked by a digital camera, in addition to an inertial measurement unit (IMU) fixed to the hub of a needle that is inserted into the tissue. A finite element model (FEM) is developed to study the effect of tissue and tumor location parameters on target motion.ResultsThe estimation error of linear regression for the SI and AP directions has been respectively 1.37{\%} and 2.87{\%}, and for quadratic polynomial regression have been 0.76{\%} and 2.41{\%} on the data from the experiments.ConclusionUsing more than one surrogate signals resulted in a about 0.5-6.5{\%} decrease of the motion estimation error compared to using only one of the surrogates.Keywords: hepatic tumor, motion estimation, respiratory motion, machine learning, robotic phantom, finite element analysis",
    author = "Maryam Berijanian and {Naghibi Beidokhti}, Hamid and Beril Sirma{\cc}ek and Momen Abayazid",
    year = "2019",
    language = "English",
    note = "33rd International Conference on Computer Assisted Radiology and Surgery 2019, CARS 2019 ; Conference date: 17-06-2019 Through 21-06-2019",
    url = "https://www.cars2019.org/",

    }

    Berijanian, M, Naghibi Beidokhti, H, Sirmaçek, B & Abayazid, M 2019, 'Toward patient-specific estimation of hepatic tumors respiratory motion: A finite element-based machine learning approach' Paper presented at 33rd International Conference on Computer Assisted Radiology and Surgery 2019, Rennes, France, 17/06/19 - 21/06/19, .

    Toward patient-specific estimation of hepatic tumors respiratory motion: A finite element-based machine learning approach. / Berijanian, Maryam; Naghibi Beidokhti, Hamid ; Sirmaçek, Beril ; Abayazid, Momen .

    2019. Paper presented at 33rd International Conference on Computer Assisted Radiology and Surgery 2019, Rennes, France.

    Research output: Contribution to conferencePaper

    TY - CONF

    T1 - Toward patient-specific estimation of hepatic tumors respiratory motion: A finite element-based machine learning approach

    AU - Berijanian, Maryam

    AU - Naghibi Beidokhti, Hamid

    AU - Sirmaçek, Beril

    AU - Abayazid, Momen

    PY - 2019

    Y1 - 2019

    N2 - PurposeTo develop and evaluate an approach to estimate the respiratory-induced motion of hepatic lesions.Materials and methodsIn this study, a correlation between the collected surrogate signals and the liver tumor respiratory motion is obtained using learning-based algorithms. A robotic phantom is developed which simulates the respiratory motion of the liver, the diaphragm, and the abdomen skin in two directions as superior-inferior (SI) and anterior-posterior (AP). The surrogate signals are collected by means of optical markers attached to the abdomen skin and tracked by a digital camera, in addition to an inertial measurement unit (IMU) fixed to the hub of a needle that is inserted into the tissue. A finite element model (FEM) is developed to study the effect of tissue and tumor location parameters on target motion.ResultsThe estimation error of linear regression for the SI and AP directions has been respectively 1.37% and 2.87%, and for quadratic polynomial regression have been 0.76% and 2.41% on the data from the experiments.ConclusionUsing more than one surrogate signals resulted in a about 0.5-6.5% decrease of the motion estimation error compared to using only one of the surrogates.Keywords: hepatic tumor, motion estimation, respiratory motion, machine learning, robotic phantom, finite element analysis

    AB - PurposeTo develop and evaluate an approach to estimate the respiratory-induced motion of hepatic lesions.Materials and methodsIn this study, a correlation between the collected surrogate signals and the liver tumor respiratory motion is obtained using learning-based algorithms. A robotic phantom is developed which simulates the respiratory motion of the liver, the diaphragm, and the abdomen skin in two directions as superior-inferior (SI) and anterior-posterior (AP). The surrogate signals are collected by means of optical markers attached to the abdomen skin and tracked by a digital camera, in addition to an inertial measurement unit (IMU) fixed to the hub of a needle that is inserted into the tissue. A finite element model (FEM) is developed to study the effect of tissue and tumor location parameters on target motion.ResultsThe estimation error of linear regression for the SI and AP directions has been respectively 1.37% and 2.87%, and for quadratic polynomial regression have been 0.76% and 2.41% on the data from the experiments.ConclusionUsing more than one surrogate signals resulted in a about 0.5-6.5% decrease of the motion estimation error compared to using only one of the surrogates.Keywords: hepatic tumor, motion estimation, respiratory motion, machine learning, robotic phantom, finite element analysis

    M3 - Paper

    ER -

    Berijanian M, Naghibi Beidokhti H, Sirmaçek B, Abayazid M. Toward patient-specific estimation of hepatic tumors respiratory motion: A finite element-based machine learning approach. 2019. Paper presented at 33rd International Conference on Computer Assisted Radiology and Surgery 2019, Rennes, France.