Publication detail

Imaging of Ultrasound Attenuation Coefficient

JIŘÍK, R., JAN, J.

Original Title

Imaging of Ultrasound Attenuation Coefficient

Czech Title

Zobrazování ultrazvukového koeficientu útlumu

English Title

Imaging of Ultrasound Attenuation Coefficient

Type

journal article

Language

en

Original Abstract

The paper studies imaging of tissue-specific ultrasound attenuation. Ultrasound attenuation parameters are closely related to the type and the pathological state of the tissue. Therefore, the estimated attenuation parameters can be used as a feature in quantitative tissue characterization. Ultrasound attenuation is also an important factor affecting spatial resolution of ultrasound images since the higher ultrasonic frequencies are attenuated more than the lower ones. Knowing the attenuation as a function of frequency and depth allows estimation of the depth-dependent point spread function component caused by attenuation. This component can be used to improve the spatial resolution of the ultrasound image through deconvolution. This paper introduces a method, for tracking the attenuation coefficient in B-mode ultrasonic images on a pixel-by-pixel basis. The resulting parametric images show the mean attenuation coefficient between the probe and a given pixel position, later referred to as the cumulative attenuation coefficient. The presented ultrasound attenuation imaging is based on a physical model of ultrasound radiofrequency signal. A-scan radiofrequency signals are first divided into short overlapping segments. Each short segment is then transformed from the time domain to the log-spectrum domain. The cumulative attenuation coefficient is estimated for each segment by means of the least-mean-squares approximation of the segment log-spectrum using its physical model. The proposed method was tested on images of a tissue-mimicking phantom and on clinical images of liver and thigh. The parametric images show some of the tissue structures. In selected regions of the parametric images, the local tissue-specific attenuation coefficient was estimated and compared with reference values. For phantom images, the reference values of the local attenuation coefficient were given by the specifications of the used phantom. For the clinical images of liver and thigh, the reference values were taken from literature. The estimated local attenuation coefficients were mostly in accordance with the reference values.

Czech abstract

The paper studies imaging of tissue-specific ultrasound attenuation. Ultrasound attenuation parameters are closely related to the type and the pathological state of the tissue. Therefore, the estimated attenuation parameters can be used as a feature in quantitative tissue characterization. Ultrasound attenuation is also an important factor affecting spatial resolution of ultrasound images since the higher ultrasonic frequencies are attenuated more than the lower ones. Knowing the attenuation as a function of frequency and depth allows estimation of the depth-dependent point spread function component caused by attenuation. This component can be used to improve the spatial resolution of the ultrasound image through deconvolution. This paper introduces a method, for tracking the attenuation coefficient in B-mode ultrasonic images on a pixel-by-pixel basis. The resulting parametric images show the mean attenuation coefficient between the probe and a given pixel position, later referred to as the cumulative attenuation coefficient. The presented ultrasound attenuation imaging is based on a physical model of ultrasound radiofrequency signal. A-scan radiofrequency signals are first divided into short overlapping segments. Each short segment is then transformed from the time domain to the log-spectrum domain. The cumulative attenuation coefficient is estimated for each segment by means of the least-mean-squares approximation of the segment log-spectrum using its physical model. The proposed method was tested on images of a tissue-mimicking phantom and on clinical images of liver and thigh. The parametric images show some of the tissue structures. In selected regions of the parametric images, the local tissue-specific attenuation coefficient was estimated and compared with reference values. For phantom images, the reference values of the local attenuation coefficient were given by the specifications of the used phantom. For the clinical images of liver and thigh, the reference values were taken from literature. The estimated local attenuation coefficients were mostly in accordance with the reference values.

English abstract

The paper studies imaging of tissue-specific ultrasound attenuation. Ultrasound attenuation parameters are closely related to the type and the pathological state of the tissue. Therefore, the estimated attenuation parameters can be used as a feature in quantitative tissue characterization. Ultrasound attenuation is also an important factor affecting spatial resolution of ultrasound images since the higher ultrasonic frequencies are attenuated more than the lower ones. Knowing the attenuation as a function of frequency and depth allows estimation of the depth-dependent point spread function component caused by attenuation. This component can be used to improve the spatial resolution of the ultrasound image through deconvolution. This paper introduces a method, for tracking the attenuation coefficient in B-mode ultrasonic images on a pixel-by-pixel basis. The resulting parametric images show the mean attenuation coefficient between the probe and a given pixel position, later referred to as the cumulative attenuation coefficient. The presented ultrasound attenuation imaging is based on a physical model of ultrasound radiofrequency signal. A-scan radiofrequency signals are first divided into short overlapping segments. Each short segment is then transformed from the time domain to the log-spectrum domain. The cumulative attenuation coefficient is estimated for each segment by means of the least-mean-squares approximation of the segment log-spectrum using its physical model. The proposed method was tested on images of a tissue-mimicking phantom and on clinical images of liver and thigh. The parametric images show some of the tissue structures. In selected regions of the parametric images, the local tissue-specific attenuation coefficient was estimated and compared with reference values. For phantom images, the reference values of the local attenuation coefficient were given by the specifications of the used phantom. For the clinical images of liver and thigh, the reference values were taken from literature. The estimated local attenuation coefficients were mostly in accordance with the reference values.

Keywords

Medical Ultrasound, Ultrasound Attenuation Coefficient, Attenuation Imaging, Parametric Imaging

RIV year

2004

Released

01.02.2004

Publisher

IOS Press

Pages from

172

Pages to

173

Pages count

2

BibTex


@article{BUT45665,
  author="Radovan {Jiřík} and Jiří {Jan}",
  title="Imaging of Ultrasound Attenuation Coefficient",
  annote="The paper studies imaging of tissue-specific ultrasound attenuation. Ultrasound attenuation parameters are closely related to the type and the pathological state of the tissue. Therefore, the estimated attenuation parameters can be used as a feature in quantitative tissue characterization. Ultrasound attenuation is also an important factor affecting spatial resolution of ultrasound images since the higher ultrasonic frequencies are attenuated more than the lower ones. Knowing the attenuation as a function of frequency and depth allows estimation of the depth-dependent point spread function component caused by attenuation. This component can be used to improve the spatial resolution of the ultrasound image through deconvolution.

This paper introduces a method, for tracking the attenuation coefficient in B-mode ultrasonic images on a pixel-by-pixel basis. The resulting parametric images show the mean attenuation coefficient between the probe and a given pixel position, later referred to as the cumulative attenuation coefficient. The presented ultrasound attenuation imaging is based on a physical model of ultrasound radiofrequency signal. A-scan radiofrequency signals are first divided into short overlapping segments. Each short segment is then transformed from the time domain to the log-spectrum domain. The cumulative attenuation coefficient is estimated for each segment by means of the least-mean-squares approximation of the segment log-spectrum using its physical model. 

The proposed method was tested on images of a tissue-mimicking phantom and on clinical images of liver and thigh. The parametric images show some of the tissue structures. In selected regions of the parametric images, the local tissue-specific attenuation coefficient was estimated and compared with reference values. For phantom images, the reference values of the local attenuation coefficient were given by the specifications of the used phantom. For the clinical images of liver and thigh, the reference values were taken from literature. The estimated local attenuation coefficients were mostly in accordance with the reference values.

",
  address="IOS Press",
  booktitle="Technology and Health Care, Special Issue: Abstracts of ESEM 2003",
  chapter="45665",
  institution="IOS Press",
  journal="Technology and Health Care, Int. Journal of Health Care Engineering",
  number="2",
  volume="12",
  year="2004",
  month="february",
  pages="172",
  publisher="IOS Press",
  type="journal article"
}