lsqcurvefit.m function  (MathWorks Inc)

Journal: IEEE transactions on ultrasonics, ferroelectrics, and frequency control

Article Title: On the Effects of Spatial Sampling Quantization in Super-Resolution Ultrasound Microvessel Imaging

doi: 10.1109/TUFFC.2018.2832600

Figure Lengend Snippet: Summary of root-mean-square error (RMSE) of wire target localization

Article Snippet: For the parametric Gaussian fitting-based localization, a parametric fitting in a least-squares sense (i.e., Matlab function “lsqcurvefit.m”) was applied on the original data to derive an analytical solution of the wire signal modeled as a 1D Gaussian function.

Techniques:

Journal: IEEE transactions on ultrasonics, ferroelectrics, and frequency control

Article Title: On the Effects of Spatial Sampling Quantization in Super-Resolution Ultrasound Microvessel Imaging

doi: 10.1109/TUFFC.2018.2832600

Figure Lengend Snippet: SR vessel density images of the flow channel obtained from different microbubble localization methods (b–g, localization methods indicated in the subtitles). The axial and lateral beamforming resolution was both 0.25 λ. For each SR image, a magnified view of a local region inside the channel was displayed (as indicated by the white box on the top left image). To facilitate better visualization of the pixelated SR images, square root compression was applied to each image followed by a modest 2D Gaussian smoothing filter (3 × 3 window, σ = 0.5). The smoothing filter was only applied to the non-zoomed background image. No smoothing filtering was applied to the zoomed local SR images to facilitate better comparisons among various conditions. For the reference data in (a), direct microbubble localization was performed on oversampled data.

Article Snippet: For the parametric Gaussian fitting-based localization, a parametric fitting in a least-squares sense (i.e., Matlab function “lsqcurvefit.m”) was applied on the original data to derive an analytical solution of the wire signal modeled as a 1D Gaussian function.

Techniques:

Journal: IEEE transactions on ultrasonics, ferroelectrics, and frequency control

Article Title: On the Effects of Spatial Sampling Quantization in Super-Resolution Ultrasound Microvessel Imaging

doi: 10.1109/TUFFC.2018.2832600

Figure Lengend Snippet: SR vessel density images of the flow channel obtained from different microbubble localization methods (b–g, localization methods indicated in the subtitles). The axial and lateral beamforming resolution was both 1 λ. For each SR image, a magnified view of a local region inside the channel was displayed (as indicated by the white box on the top left image). To facilitate better visualization of the pixelated SR images, square root compression was applied to each image followed by a modest 2D Gaussian smoothing filter (3 × 3 window, σ = 0.5). The smoothing filter was only applied to the non-zoomed background image. No smoothing filtering was applied to the zoomed local SR images to facilitate better comparisons among various conditions. For the reference data in (a), direct microbubble localization was performed on oversampled data.

Article Snippet: For the parametric Gaussian fitting-based localization, a parametric fitting in a least-squares sense (i.e., Matlab function “lsqcurvefit.m”) was applied on the original data to derive an analytical solution of the wire signal modeled as a 1D Gaussian function.

Techniques:

Journal: IEEE transactions on ultrasonics, ferroelectrics, and frequency control

Article Title: On the Effects of Spatial Sampling Quantization in Super-Resolution Ultrasound Microvessel Imaging

doi: 10.1109/TUFFC.2018.2832600

Figure Lengend Snippet: SR vessel density images of the flow channel obtained from different microbubble localization methods (b–g, localization methods indicated in the subtitles). The axial and lateral beamforming resolution was 0.5 λ and 1 λ, respectively. For each SR image, a magnified view of a local region inside the channel was displayed (as indicated by the white box on the top left image). To facilitate better visualization of the pixelated SR images, square root compression was applied to each image followed by a modest 2D Gaussian smoothing filter (3 × 3 window, σ = 0.5). The smoothing filter was only applied to the non-zoomed background image. No smoothing filtering was applied to the zoomed local SR images to facilitate better comparisons among various conditions. For the reference data in (a), direct microbubble localization was performed on oversampled data.

Article Snippet: For the parametric Gaussian fitting-based localization, a parametric fitting in a least-squares sense (i.e., Matlab function “lsqcurvefit.m”) was applied on the original data to derive an analytical solution of the wire signal modeled as a 1D Gaussian function.

Techniques: