Adam C. Luchies
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Adam C. Luchies has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Brett Byram
Department of Biomedical Engineering, Vanderbilt University, Nashville, TN, USA
5 shared publications Michael L. Oelze
Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, Illinois 61801, USA
4 shared publications- Jaime Tierney
Department of Biomedical Engineering, Vanderbilt University, Nashville, TN, USA
2 shared publications Matthew Berger
Department of Electrical Engineering and Computer Science, Vanderbilt University, Nashville, TN, USA
2 shared publicationsBrett C. Byram
Department of Biomedical Engineering, Vanderbilt University, Nashville, TN, USA
1 shared publication- Christopher Khan
Vanderbilt University
1 shared publication
Latest Publications
- 30 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 41 (1) , 158-171
Training Deep Network Ultrasound Beamformers With Unlabeled In Vivo Data
Conventional delay-and-sum (DAS) beamforming is highly efficient but also suffers from various sources of image degradation. Several adaptive beamformers have been proposed to address this problem, including more recently proposed deep learning methods. With deep learning, adaptive beamforming is typically framed as a regression problem, where clean ground-truth physical information is used for training. Because it is difficult to know ground truth information in vivo , training data are usually simulated. However, deep networks trained on simulations can produce suboptimal in vivo image quality because of a domain shift between simulated and in vivo data. In this work, we propose a novel domain adaptation (DA) scheme to correct for domain shift by incorporating unlabeled in vivo data during training. Unlike classification tasks for which both input domains map to the same target domain, a challenge in our regression-based beamforming scenario is that domain shift exists in both the input and target data. To solve this problem, we leverage cycle-consistent generative adversarial networks to map between simulated and in vivo data in both the input and ground truth target domains. Additionally, to account for separate as well as shared features between simulations and in vivo data, we use augmented feature mapping to train domain-specific beamformers. Using various types of training data, we explore the limitations and underlying functionality of the proposed DA approach. Additionally, we compare our proposed approach to several other adaptive beamformers. Using the DA DNN beamformer, consistent in vivo image quality improvements are achieved compared to established techniques.
- 30 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 41 (1) , 158-171
Training Deep Network Ultrasound Beamformers With Unlabeled In Vivo Data
Conventional delay-and-sum (DAS) beamforming is highly efficient but also suffers from various sources of image degradation. Several adaptive beamformers have been proposed to address this problem, including more recently proposed deep learning methods. With deep learning, adaptive beamforming is typically framed as a regression problem, where clean ground-truth physical information is used for training. Because it is difficult to know ground truth information in vivo , training data are usually simulated. However, deep networks trained on simulations can produce suboptimal in vivo image quality because of a domain shift between simulated and in vivo data. In this work, we propose a novel domain adaptation (DA) scheme to correct for domain shift by incorporating unlabeled in vivo data during training. Unlike classification tasks for which both input domains map to the same target domain, a challenge in our regression-based beamforming scenario is that domain shift exists in both the input and target data. To solve this problem, we leverage cycle-consistent generative adversarial networks to map between simulated and in vivo data in both the input and ground truth target domains. Additionally, to account for separate as well as shared features between simulations and in vivo data, we use augmented feature mapping to train domain-specific beamformers. Using various types of training data, we explore the limitations and underlying functionality of the proposed DA approach. Additionally, we compare our proposed approach to several other adaptive beamformers. Using the DA DNN beamformer, consistent in vivo image quality improvements are achieved compared to established techniques.
- 29 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
- Vol. 68 (7) , 2370-2385
Evaluating Input Domain and Model Selection for Deep Network Ultrasound Beamforming
Improving ultrasound B-mode image quality remains an important area of research. Recently, there has been increased interest in using deep neural networks (DNNs) to perform beamforming to improve image quality more efficiently. Several approaches have been proposed that use different representations of channel data for network processing, including a frequency-domain approach that we previously developed. We previously assumed that the frequency domain would be more robust to varying pulse shapes. However, frequency- and time-domain implementations have not been directly compared. In addition, because our approach operates on aperture domain data as an intermediate beamforming step, a discrepancy often exists between network performance and image quality on fully reconstructed images, making model selection challenging. Here, we perform a systematic comparison of frequency- and time-domain implementations. In addition, we propose a contrast-to-noise ratio (CNR)-based regularization to address previous challenges with model selection. Training channel data were generated from simulated anechoic cysts. Test channel data were generated from simulated anechoic cysts with and without varied pulse shapes, in addition to physical phantom and in vivo data. We demonstrate that simplified time-domain implementations are more robust than we previously assumed, especially when using phase preserving data representations. Specifically, 0.39- and 0.36-dB median improvements in in vivo CNR compared to DAS were achieved with frequency- and time-domain implementations, respectively. We also demonstrate that CNR regularization improves the correlation between training validation loss and simulated CNR by 0.83 and between simulated and in vivo CNR by 0.35 compared to DNNs trained without CNR regularization.
- 29 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
- Vol. 68 (7) , 2370-2385
Evaluating Input Domain and Model Selection for Deep Network Ultrasound Beamforming
Improving ultrasound B-mode image quality remains an important area of research. Recently, there has been increased interest in using deep neural networks (DNNs) to perform beamforming to improve image quality more efficiently. Several approaches have been proposed that use different representations of channel data for network processing, including a frequency-domain approach that we previously developed. We previously assumed that the frequency domain would be more robust to varying pulse shapes. However, frequency- and time-domain implementations have not been directly compared. In addition, because our approach operates on aperture domain data as an intermediate beamforming step, a discrepancy often exists between network performance and image quality on fully reconstructed images, making model selection challenging. Here, we perform a systematic comparison of frequency- and time-domain implementations. In addition, we propose a contrast-to-noise ratio (CNR)-based regularization to address previous challenges with model selection. Training channel data were generated from simulated anechoic cysts. Test channel data were generated from simulated anechoic cysts with and without varied pulse shapes, in addition to physical phantom and in vivo data. We demonstrate that simplified time-domain implementations are more robust than we previously assumed, especially when using phase preserving data representations. Specifically, 0.39- and 0.36-dB median improvements in in vivo CNR compared to DAS were achieved with frequency- and time-domain implementations, respectively. We also demonstrate that CNR regularization improves the correlation between training validation loss and simulated CNR by 0.83 and between simulated and in vivo CNR by 0.35 compared to DNNs trained without CNR regularization.
- 15 June 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
- Vol. 67 (11) , 2321-2335
Assessing the Robustness of Frequency-Domain Ultrasound Beamforming Using Deep Neural Networks
We study training deep neural network (DNN) frequency-domain beamformers using simulated and phantom anechoic cysts and compare to training with simulated point target responses. Using simulation, physical phantom, and in vivo scans, we find that training DNN beamformers using anechoic cysts provided comparable or improved image quality compared with training DNN beamformers using simulated point targets. The proposed method could also be adapted to generate training data from in vivo scans. Finally, we evaluated the robustness of DNN beamforming to common sources of image degradation, including gross sound speed errors, phase aberration, and reverberation. We found that DNN beamformers maintained their ability to improve image quality even in the presence of the studied sources of image degradation. Overall, the results show the potential of using DNN beamforming to improve ultrasound image quality.
- 15 June 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
- Vol. 67 (11) , 2321-2335
Assessing the Robustness of Frequency-Domain Ultrasound Beamforming Using Deep Neural Networks
We study training deep neural network (DNN) frequency-domain beamformers using simulated and phantom anechoic cysts and compare to training with simulated point target responses. Using simulation, physical phantom, and in vivo scans, we find that training DNN beamformers using anechoic cysts provided comparable or improved image quality compared with training DNN beamformers using simulated point targets. The proposed method could also be adapted to generate training data from in vivo scans. Finally, we evaluated the robustness of DNN beamforming to common sources of image degradation, including gross sound speed errors, phase aberration, and reverberation. We found that DNN beamformers maintained their ability to improve image quality even in the presence of the studied sources of image degradation. Overall, the results show the potential of using DNN beamforming to improve ultrasound image quality.
- 1 October 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 568-570
An improved training scheme for deep neural network ultrasound beamforming
Deep neural networks have been shown to be effective adaptive beamformers for ultrasound imaging. However, when training with traditional ${\mathcal{L}_p}$ norm loss functions, model selection is difficult because lower loss values are not always associated with higher image quality. This ultimately limits the maximum achievable image quality with this approach and raises concerns about the optimization objective. In an effort to align the optimization objective with the image quality metrics of interest, we implemented a novel ultrasound-specific loss function based on the spatial lag-one coherence and signal-to-noise ratio of the delayed channel data in the short-time Fourier domain. We employed the R-Adam optimizer with look ahead and cyclical learning rate to make the training more robust to initialization and local minima, leading to better model performance and more reliable convergence. With our custom loss function and optimization scheme, we achieved higher contrast-to-noise-ratio, higher speckle signal-to-noise-ratio, and more accurate contrast ratio reconstruction than with previous deep learning and delay-and-sum beamforming approaches.- 1 October 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 568-570
An improved training scheme for deep neural network ultrasound beamforming
Deep neural networks have been shown to be effective adaptive beamformers for ultrasound imaging. However, when training with traditional ${\mathcal{L}_p}$ norm loss functions, model selection is difficult because lower loss values are not always associated with higher image quality. This ultimately limits the maximum achievable image quality with this approach and raises concerns about the optimization objective. In an effort to align the optimization objective with the image quality metrics of interest, we implemented a novel ultrasound-specific loss function based on the spatial lag-one coherence and signal-to-noise ratio of the delayed channel data in the short-time Fourier domain. We employed the R-Adam optimizer with look ahead and cyclical learning rate to make the training more robust to initialization and local minima, leading to better model performance and more reliable convergence. With our custom loss function and optimization scheme, we achieved higher contrast-to-noise-ratio, higher speckle signal-to-noise-ratio, and more accurate contrast ratio reconstruction than with previous deep learning and delay-and-sum beamforming approaches.- 18 February 2019
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 64 (4) , 045018
Training improvements for ultrasound beamforming with deep neural networks
This paper investigates practical considerations of training ultrasound deep neural network (DNN) beamformers. First, we studied training DNNs using the combination of multiple point target responses instead of single point target responses. Next, we demonstrated the effect of different hyperparameter settings on ultrasound image quality for simulated scans. This study also showed that DNN beamforming was robust to electronic noise. Next, we showed that mean squared error validation loss was not a good predictor for image quality for simulation, phantom, and in vivo scans. As an alternative to validation loss for selecting DNN beamformers, we studied image quality in physical phantom and in vivo scans and demonstrated that DNN beamformer image quality in these settings was correlated to DNN beamformer image quality in simulated images. These findings suggest that simulated image quality can be used to select DNN beamformers. Finally, we studied the effect of dataset size on DNN beamformer image quality in simulation, physical phantom, and in vivo scans. We interpret the results in terms of recent work on the scaling of deep learning. Overall, the results in this paper show that DNN beamforming has significant potential for improving B-mode image quality.
- 18 February 2019
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 64 (4) , 045018
Training improvements for ultrasound beamforming with deep neural networks
This paper investigates practical considerations of training ultrasound deep neural network (DNN) beamformers. First, we studied training DNNs using the combination of multiple point target responses instead of single point target responses. Next, we demonstrated the effect of different hyperparameter settings on ultrasound image quality for simulated scans. This study also showed that DNN beamforming was robust to electronic noise. Next, we showed that mean squared error validation loss was not a good predictor for image quality for simulation, phantom, and in vivo scans. As an alternative to validation loss for selecting DNN beamformers, we studied image quality in physical phantom and in vivo scans and demonstrated that DNN beamformer image quality in these settings was correlated to DNN beamformer image quality in simulated images. These findings suggest that simulated image quality can be used to select DNN beamformers. Finally, we studied the effect of dataset size on DNN beamformer image quality in simulation, physical phantom, and in vivo scans. We interpret the results in terms of recent work on the scaling of deep learning. Overall, the results in this paper show that DNN beamforming has significant potential for improving B-mode image quality.
- 1 April 2018
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 143 (4) , 2172-2181
Effects of the container on structure function with impedance map analysis of dense scattering media
Quantitative ultrasound (QUS) can be used to estimate acoustic properties of tissue microstructure. In one approach to QUS, the backscatter coefficient (BSC) is utilized to quantify and classify tissue state. From the BSC, parametric models can be constructed to relate the frequency-dependent BSC to geometrical properties of the underlying tissue. However, most of these parametric models are based on analytic expressions (e.g., Gaussian function) and not on actual tissue morphology. Impedance map analysis has been proposed to help identify sources of ultrasonic scattering in tissues and to develop improved models of scattering. Previously, two-dimensional impedance maps (2DZMs) were demonstrated to provide tissue models of three-dimensional (3D) structures for sparse scattering media. In the current study, 2DZMs analysis of dense scatterer media combining the structure function with impedance map analysis was studied through a series of simulations. The simulation analysis demonstrated that the correlation coefficient and power spectrum could be estimated for a dense collection of spheres using 2DZMs. The current finding implies that 2DZMs can capture information about the 3D spatial positions of scatterers in addition to information about the size and shape of the scatterers for a dense scattering media, which is expected to be encountered in many tissues.
- 1 April 2018
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 143 (4) , 2172-2181
Effects of the container on structure function with impedance map analysis of dense scattering media
Quantitative ultrasound (QUS) can be used to estimate acoustic properties of tissue microstructure. In one approach to QUS, the backscatter coefficient (BSC) is utilized to quantify and classify tissue state. From the BSC, parametric models can be constructed to relate the frequency-dependent BSC to geometrical properties of the underlying tissue. However, most of these parametric models are based on analytic expressions (e.g., Gaussian function) and not on actual tissue morphology. Impedance map analysis has been proposed to help identify sources of ultrasonic scattering in tissues and to develop improved models of scattering. Previously, two-dimensional impedance maps (2DZMs) were demonstrated to provide tissue models of three-dimensional (3D) structures for sparse scattering media. In the current study, 2DZMs analysis of dense scatterer media combining the structure function with impedance map analysis was studied through a series of simulations. The simulation analysis demonstrated that the correlation coefficient and power spectrum could be estimated for a dense collection of spheres using 2DZMs. The current finding implies that 2DZMs can capture information about the 3D spatial positions of scatterers in addition to information about the size and shape of the scatterers for a dense scattering media, which is expected to be encountered in many tissues.
- 27 February 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 37 (9) , 2010-2021
Deep Neural Networks for Ultrasound Beamforming
We investigate the use of deep neural networks (DNNs) for suppressing off-axis scattering in ultrasound channel data. Our implementation operates in the frequency domain via the short-time Fourier transform. The inputs to the DNN consisted of the separated real and imaginary components (i.e. in-phase and quadrature components) observed across the aperture of the array, at a single frequency and for a single depth. Different networks were trained for different frequencies. The output had the same structure as the input and the real and imaginary components were combined as complex data before an inverse short-time Fourier transform was used to reconstruct channel data. Using simulation, physical phantom experiment, and in vivo scans from a human liver, we compared this DNN approach to standard delay-and-sum (DAS) beamforming and an adaptive imaging technique that uses the coherence factor. For a simulated point target, the side lobes when using the DNN approach were about 60 dB below those of standard DAS. For a simulated anechoic cyst, the DNN approach improved contrast ratio (CR) and contrast-to-noise (CNR) ratio by 8.8 dB and 0.3 dB, respectively, compared with DAS. For an anechoic cyst in a physical phantom, the DNN approach improved CR and CNR by 17.1 dB and 0.7 dB, respectively. For two in vivo scans, the DNN approach improved CR and CNR by 13.8 dB and 9.7 dB, respectively. We also explored methods for examining how the networks in this paper function.
- 27 February 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 37 (9) , 2010-2021
Deep Neural Networks for Ultrasound Beamforming
We investigate the use of deep neural networks (DNNs) for suppressing off-axis scattering in ultrasound channel data. Our implementation operates in the frequency domain via the short-time Fourier transform. The inputs to the DNN consisted of the separated real and imaginary components (i.e. in-phase and quadrature components) observed across the aperture of the array, at a single frequency and for a single depth. Different networks were trained for different frequencies. The output had the same structure as the input and the real and imaginary components were combined as complex data before an inverse short-time Fourier transform was used to reconstruct channel data. Using simulation, physical phantom experiment, and in vivo scans from a human liver, we compared this DNN approach to standard delay-and-sum (DAS) beamforming and an adaptive imaging technique that uses the coherence factor. For a simulated point target, the side lobes when using the DNN approach were about 60 dB below those of standard DAS. For a simulated anechoic cyst, the DNN approach improved contrast ratio (CR) and contrast-to-noise (CNR) ratio by 8.8 dB and 0.3 dB, respectively, compared with DAS. For an anechoic cyst in a physical phantom, the DNN approach improved CR and CNR by 17.1 dB and 0.7 dB, respectively. For two in vivo scans, the DNN approach improved CR and CNR by 13.8 dB and 9.7 dB, respectively. We also explored methods for examining how the networks in this paper function.
- 1 April 2016
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 139 (4) , 1557-1564
Using two-dimensional impedance maps to study weak scattering in sparse random media
Impedance maps (ZMs) have been proposed as a tool for modeling acoustic properties of tissue microstructure. Three-dimensional (3D) ZMs are constructed from a series of adjacent histological tissue slides that have been stained to emphasize acoustic impedance structures. The power spectrum of a 3DZM can be related to the ultrasound backscatter coefficient, which can be further reduced to a form factor. The goal of this study is to demonstrate the ability to estimate form factors using two-dimensional (2D) ZMs instead of 3DZMs, which have reduced computational and financial cost. The proposed method exploits the properties of isotropic media to estimate the correlation coefficient from slices before estimating the 3D volume power spectrum. Simulated sparse collections of objects were used to study the method by comparing the results obtained using 2DZMs to those predicted by theory. 2DZM analysis was conducted on normal rabbit liver histology and compared to 3DZM analysis of the same histology. The mean percent error between effective scatterer diameter estimates from 2DZMs and 3DZMs of rabbit liver histology was <10% when using three 2DZMs. The results suggest that 2DZMs are a feasible alternative to 3DZMs when estimating form factors for sparse collections of objects.
- 1 April 2016
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 139 (4) , 1557-1564
Using two-dimensional impedance maps to study weak scattering in sparse random media
Impedance maps (ZMs) have been proposed as a tool for modeling acoustic properties of tissue microstructure. Three-dimensional (3D) ZMs are constructed from a series of adjacent histological tissue slides that have been stained to emphasize acoustic impedance structures. The power spectrum of a 3DZM can be related to the ultrasound backscatter coefficient, which can be further reduced to a form factor. The goal of this study is to demonstrate the ability to estimate form factors using two-dimensional (2D) ZMs instead of 3DZMs, which have reduced computational and financial cost. The proposed method exploits the properties of isotropic media to estimate the correlation coefficient from slices before estimating the 3D volume power spectrum. Simulated sparse collections of objects were used to study the method by comparing the results obtained using 2DZMs to those predicted by theory. 2DZM analysis was conducted on normal rabbit liver histology and compared to 3DZM analysis of the same histology. The mean percent error between effective scatterer diameter estimates from 2DZMs and 3DZMs of rabbit liver histology was <10% when using three 2DZMs. The results suggest that 2DZMs are a feasible alternative to 3DZMs when estimating form factors for sparse collections of objects.
- 3 September 2014
- Published by SAGE Publications in Ultrasonic Imaging
- Vol. 37 (2) , 117-134
Backscatter Coefficient Estimation Using Tapers with Gaps
When using the backscatter coefficient (BSC) to estimate quantitative ultrasound parameters such as the effective scatterer diameter (ESD) and the effective acoustic concentration (EAC), it is necessary to assume that the interrogated medium contains diffuse scatterers. Structures that invalidate this assumption can affect the estimated BSC parameters in terms of increased bias and variance and decrease performance when classifying disease. In this work, a method was developed to mitigate the effects of echoes from structures that invalidate the assumption of diffuse scattering, while preserving as much signal as possible for obtaining diffuse scatterer property estimates. Backscattered signal sections that contained nondiffuse signals were identified and a windowing technique was used to provide BSC estimates for diffuse echoes only. Experiments from physical phantoms were used to evaluate the effectiveness of the proposed BSC estimation methods. Tradeoffs associated with effective mitigation of specular scatterers and bias and variance introduced into the estimates were quantified. Analysis of the results suggested that discrete prolate spheroidal (PR) tapers with gaps provided the best performance for minimizing BSC error. Specifically, the mean square error for BSC between measured and theoretical had an average value of approximately 1.0 and 0.2 when using a Hanning taper and PR taper respectively, with six gaps. The BSC error due to amplitude bias was smallest for PR (Nω = 1) tapers. The BSC error due to shape bias was smallest for PR (Nω = 4) tapers. These results suggest using different taper types for estimating ESD versus EAC.
- 3 September 2014
- Published by SAGE Publications in Ultrasonic Imaging
- Vol. 37 (2) , 117-134
Backscatter Coefficient Estimation Using Tapers with Gaps
When using the backscatter coefficient (BSC) to estimate quantitative ultrasound parameters such as the effective scatterer diameter (ESD) and the effective acoustic concentration (EAC), it is necessary to assume that the interrogated medium contains diffuse scatterers. Structures that invalidate this assumption can affect the estimated BSC parameters in terms of increased bias and variance and decrease performance when classifying disease. In this work, a method was developed to mitigate the effects of echoes from structures that invalidate the assumption of diffuse scattering, while preserving as much signal as possible for obtaining diffuse scatterer property estimates. Backscattered signal sections that contained nondiffuse signals were identified and a windowing technique was used to provide BSC estimates for diffuse echoes only. Experiments from physical phantoms were used to evaluate the effectiveness of the proposed BSC estimation methods. Tradeoffs associated with effective mitigation of specular scatterers and bias and variance introduced into the estimates were quantified. Analysis of the results suggested that discrete prolate spheroidal (PR) tapers with gaps provided the best performance for minimizing BSC error. Specifically, the mean square error for BSC between measured and theoretical had an average value of approximately 1.0 and 0.2 when using a Hanning taper and PR taper respectively, with six gaps. The BSC error due to amplitude bias was smallest for PR (Nω = 1) tapers. The BSC error due to shape bias was smallest for PR (Nω = 4) tapers. These results suggest using different taper types for estimating ESD versus EAC.
- 18 May 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
- Vol. 59 (5) , 893-904
Quantitative ultrasonic characterization of diffuse scatterers in the presence of structures that produce coherent echoes
Quantitative ultrasound (QUS) techniques that parameterize the backscattered power spectrum have demonstrated significant promise for ultrasonic tissue characterization. Some QUS parameters, such as the effective scatterer diameter (ESD), require the assumption that the examined medium contains uniform diffuse scatterers. Structures that invalidate this assumption can significantly affect the estimated QUS parameters and decrease performance when classifying disease. In this work, a method was developed to reduce the effects of echoes that invalidate the assumption of diffuse scattering. To accomplish this task, backscattered signal sections containing non-diffuse echoes were identified and removed from the QUS analysis. Parameters estimated from the generalized spectrum (GS) and the Rayleigh SNR parameter were compared for detecting data blocks with non-diffuse echoes. Simulations and experiments were used to evaluate the effectiveness of the method. Experiments consisted of estimating QUS parameters from spontaneous fibroadenomas in rats and from beef liver samples. Results indicated that the method was able to significantly reduce or eliminate the effects of nondiffuse echoes that might exist in the backscattered signal. For example, the average reduction in the relative standard deviation of ESD estimates from simulation, rat fibroadenomas, and beef liver samples were 13%, 30%, and 51%, respectively. The Rayleigh SNR parameter performed best at detecting nondiffuse echoes for the purpose of removing and reducing ESD bias and variance. The method provides a means to improve the diagnostic capabilities of QUS techniques by allowing separate analysis of diffuse and non-diffuse scatterers.
- 18 May 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
- Vol. 59 (5) , 893-904
Quantitative ultrasonic characterization of diffuse scatterers in the presence of structures that produce coherent echoes
Quantitative ultrasound (QUS) techniques that parameterize the backscattered power spectrum have demonstrated significant promise for ultrasonic tissue characterization. Some QUS parameters, such as the effective scatterer diameter (ESD), require the assumption that the examined medium contains uniform diffuse scatterers. Structures that invalidate this assumption can significantly affect the estimated QUS parameters and decrease performance when classifying disease. In this work, a method was developed to reduce the effects of echoes that invalidate the assumption of diffuse scattering. To accomplish this task, backscattered signal sections containing non-diffuse echoes were identified and removed from the QUS analysis. Parameters estimated from the generalized spectrum (GS) and the Rayleigh SNR parameter were compared for detecting data blocks with non-diffuse echoes. Simulations and experiments were used to evaluate the effectiveness of the method. Experiments consisted of estimating QUS parameters from spontaneous fibroadenomas in rats and from beef liver samples. Results indicated that the method was able to significantly reduce or eliminate the effects of nondiffuse echoes that might exist in the backscattered signal. For example, the average reduction in the relative standard deviation of ESD estimates from simulation, rat fibroadenomas, and beef liver samples were 13%, 30%, and 51%, respectively. The Rayleigh SNR parameter performed best at detecting nondiffuse echoes for the purpose of removing and reducing ESD bias and variance. The method provides a means to improve the diagnostic capabilities of QUS techniques by allowing separate analysis of diffuse and non-diffuse scatterers.