Detection and display of acoustic window for guiding and training cardiac ultrasound users

Huang, Sheng‐Wen; Radulescu, E.G.; Wang, Shougang; Thiele, Karl; Prater, David; Maxwell, Douglas; Rafter, Patrick; Dupuy, Clément; Drysdale, Jeremy; Erkamp, Ramon

doi:10.1117/12.2044051

Cited by 5 publications

(4 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several groups [7]–[9] have developed methods to detect acoustic blockage using beamformed data. In [7], the edges of blocked parts of the field-of-view (FOV) are detected based on low coherence between beamsummed data from different subapertures.…”

Section: Introductionmentioning

confidence: 99%

“…In [7], the edges of blocked parts of the field-of-view (FOV) are detected based on low coherence between beamsummed data from different subapertures. In [8], the lateral power spectrum at the focus is used to infer shape of the active aperture and the locations of acoustic obstacles.…”

Section: Introductionmentioning

confidence: 99%

“…In [8], the lateral power spectrum at the focus is used to infer shape of the active aperture and the locations of acoustic obstacles. The methods in [7] and [8] are designed to provide guidance on probe positioning in order to increase the acoustic window. Ballard et al [9] used echogenecity (integrated backscatter) to determine locations of ribs prior to applying adaptive compensation on a dual-mode (imaging plus therapy) array.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Blocked Elements in 1-D and 2-D Arrays—Part I: Detection and Basic Compensation on Simulated and In Vivo Targets

Jakovljevic

Pinton

Dahl

et al. 2017

IEEE Trans. Ultrason., Ferroelect., Freq. Contr.

View full text Add to dashboard Cite

During a transcostal ultrasound scan, ribs and other highly attenuating and/or reflective tissue structures can block parts of the array. Blocked elements tend to limit the acoustic window and impede visualization of structures of interest. Here, we demonstrate a method to detect blocked elements and we measure the loss of image quality they introduce in simulation and in vivo. We utilize a fullwave simulation tool and a clinical ultrasound scanner to obtain element signals from fully sampled matrix arrays during simulated and in vivo transcostal liver scans, respectively. The elements that were blocked by a rib showed lower average signal amplitude and lower average nearest-neighbor cross correlation than the elements in the remainder of the 2-D aperture. The growing receive-aperture B-mode images created from the element data indicate that the signals on blocked elements are dominated by noise and that turning them OFF has a potential to improve visibility of liver vasculature. Adding blocked elements to the growing receive apertures for five in vivo transcostal acquisitions resulted in average decrease in vessel contrast and contrast to noise ratio of 19% and 10%, respectively.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Blocked Elements in 1-D and 2-D Arrays—Part I: Detection and Basic Compensation on Simulated and In Vivo Targets

Jakovljevic

Pinton

Dahl

et al. 2017

IEEE Trans. Ultrason., Ferroelect., Freq. Contr.

View full text Add to dashboard Cite

show abstract

“…Beyond conventional methods in estimating ultrasound image quality [3], [4], [5], [6], [7], most recent methods are based on deep learning to assess image quality. Wu et al [8] proposed a convolution neural networks (CNN) for fetal US quality assessment scheme in order to control US image quality.…”

mentioning

confidence: 99%

On Modelling Label Uncertainty in Deep Neural Networks: Automatic Estimation of Intra- Observer Variability in 2D Echocardiography Quality Assessment

Liao

Girgis

Abdi

et al. 2020

IEEE Trans. Med. Imaging

View full text Add to dashboard Cite

Uncertainty of labels in clinical data resulting from intra-observer variability can have direct impact on the reliability of assessments made by deep neural networks. In this paper, we propose a method for modelling such uncertainty in the context of 2D echocardiography (echo), which is a routine procedure for detecting cardiovascular disease at point-of-care. Echo imaging quality and acquisition time is highly dependent on the operator's experience level. Recent developments have shown the possibility of automating echo image quality quantification by mapping an expert's assessment of quality to the echo image via deep learning techniques. Nevertheless, the observer variability in the expert's assessment can impact the quality quantification accuracy. Here, we aim to model the intra-observer variability in echo quality assessment as an aleatoric uncertainty modelling regression problem with the introduction of a novel method that handles the regression problem with categorical labels. A key feature of our design is that only a single forward pass is sufficient to estimate the level of uncertainty for the network output. Compared to the 0.11±0.09 absolute error (in a scale from 0 to 1) archived by the conventional regression method, the proposed method brings the error down to 0.09 ± 0.08, where the improvement is statistically significant and equivalents to 5.7% test accuracy improvement. The simplicity of the proposed approach means that it could be generalized to other applications of deep learning in medical imaging, where there is often uncertainty in clinical labels.

show abstract

Quality Assessment of Echocardiographic Cine Using Recurrent Neural Networks: Feasibility on Five Standard View Planes

Abdi

Luong

Tsang

et al. 2017

Medical Image Computing and Computer Assisted Intervention − MICCAI 2017

View full text Add to dashboard Cite

Detection and display of acoustic window for guiding and training cardiac ultrasound users

Cited by 5 publications

References 3 publications

Blocked Elements in 1-D and 2-D Arrays—Part I: Detection and Basic Compensation on Simulated and In Vivo Targets

Blocked Elements in 1-D and 2-D Arrays—Part I: Detection and Basic Compensation on Simulated and In Vivo Targets

On Modelling Label Uncertainty in Deep Neural Networks: Automatic Estimation of Intra- Observer Variability in 2D Echocardiography Quality Assessment

Quality Assessment of Echocardiographic Cine Using Recurrent Neural Networks: Feasibility on Five Standard View Planes

Contact Info

Product

Resources

About