Limits...
A 3D freehand ultrasound system for multi-view reconstructions from sparse 2D scanning planes.

Yu H, Pattichis MS, Agurto C, Beth Goens M - Biomed Eng Online (2011)

Bottom Line: Volume measurements from multi-view 3D reconstructions are found to be consistently and significantly more accurate than measurements from single view reconstructions.In clinical in-vivo cardiac experiments, we show that volume estimates of the left ventricle from multi-view reconstructions are found to be in better agreement with clinical measures than measures from single view reconstructions.Multi-view 3D reconstruction from sparse 2D freehand B-mode images leads to more accurate volume quantification compared to single view systems.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Electrical and Computer Engineering, University of New Mexico, Albuquerque, NM 87131, USA. pattichis@ece.unm.edu.

ABSTRACT

Background: A significant limitation of existing 3D ultrasound systems comes from the fact that the majority of them work with fixed acquisition geometries. As a result, the users have very limited control over the geometry of the 2D scanning planes.

Methods: We present a low-cost and flexible ultrasound imaging system that integrates several image processing components to allow for 3D reconstructions from limited numbers of 2D image planes and multiple acoustic views. Our approach is based on a 3D freehand ultrasound system that allows users to control the 2D acquisition imaging using conventional 2D probes.For reliable performance, we develop new methods for image segmentation and robust multi-view registration. We first present a new hybrid geometric level-set approach that provides reliable segmentation performance with relatively simple initializations and minimum edge leakage. Optimization of the segmentation model parameters and its effect on performance is carefully discussed. Second, using the segmented images, a new coarse to fine automatic multi-view registration method is introduced. The approach uses a 3D Hotelling transform to initialize an optimization search. Then, the fine scale feature-based registration is performed using a robust, non-linear least squares algorithm. The robustness of the multi-view registration system allows for accurate 3D reconstructions from sparse 2D image planes.

Results: Volume measurements from multi-view 3D reconstructions are found to be consistently and significantly more accurate than measurements from single view reconstructions. The volume error of multi-view reconstruction is measured to be less than 5% of the true volume. We show that volume reconstruction accuracy is a function of the total number of 2D image planes and the number of views for calibrated phantom. In clinical in-vivo cardiac experiments, we show that volume estimates of the left ventricle from multi-view reconstructions are found to be in better agreement with clinical measures than measures from single view reconstructions.

Conclusions: Multi-view 3D reconstruction from sparse 2D freehand B-mode images leads to more accurate volume quantification compared to single view systems. The flexibility and low-cost of the proposed system allow for fine control of the image acquisition planes for optimal 3D reconstructions from multiple views.

Show MeSH

Related in: MedlinePlus

Multi-view reconstruction with registration for the calibrated phantom. (a) short-axis view image, (b) long-axis view image, (c) short-axis view reconstruction, (d) long-axis view reconstruction, (e) two-view reconstruction without registration, and (f) two-view reconstruction with automatic registration.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3037343&req=5

Figure 9: Multi-view reconstruction with registration for the calibrated phantom. (a) short-axis view image, (b) long-axis view image, (c) short-axis view reconstruction, (d) long-axis view reconstruction, (e) two-view reconstruction without registration, and (f) two-view reconstruction with automatic registration.

Mentions: An example of the phantom image from the short-axis and long-axis views is shown in Figure 9a-b. Forty image slices were used with a region of interest (ROI) of 70 × 70 pixels in the short-axis view and 40 images with 110 × 60 pixels ROI in the long-axis view. Single view reconstructions are shown in Figure 9c-d respectively. We note that the long-axis view reconstruction appears to miss some of the data at the extremes of the z-axis. Such extreme edge artifacts are clearly unavoidable in single view reconstruction. The two-view reconstruction without registration is shown in Figure 9e. Without multi-view registration, the accurate volume, even the 3D shape of the object could not be obtained correctly. The reconstruction with volumetric registration is shown in Figure 9f. The improvement is significant.


A 3D freehand ultrasound system for multi-view reconstructions from sparse 2D scanning planes.

Yu H, Pattichis MS, Agurto C, Beth Goens M - Biomed Eng Online (2011)

Multi-view reconstruction with registration for the calibrated phantom. (a) short-axis view image, (b) long-axis view image, (c) short-axis view reconstruction, (d) long-axis view reconstruction, (e) two-view reconstruction without registration, and (f) two-view reconstruction with automatic registration.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3037343&req=5

Figure 9: Multi-view reconstruction with registration for the calibrated phantom. (a) short-axis view image, (b) long-axis view image, (c) short-axis view reconstruction, (d) long-axis view reconstruction, (e) two-view reconstruction without registration, and (f) two-view reconstruction with automatic registration.
Mentions: An example of the phantom image from the short-axis and long-axis views is shown in Figure 9a-b. Forty image slices were used with a region of interest (ROI) of 70 × 70 pixels in the short-axis view and 40 images with 110 × 60 pixels ROI in the long-axis view. Single view reconstructions are shown in Figure 9c-d respectively. We note that the long-axis view reconstruction appears to miss some of the data at the extremes of the z-axis. Such extreme edge artifacts are clearly unavoidable in single view reconstruction. The two-view reconstruction without registration is shown in Figure 9e. Without multi-view registration, the accurate volume, even the 3D shape of the object could not be obtained correctly. The reconstruction with volumetric registration is shown in Figure 9f. The improvement is significant.

Bottom Line: Volume measurements from multi-view 3D reconstructions are found to be consistently and significantly more accurate than measurements from single view reconstructions.In clinical in-vivo cardiac experiments, we show that volume estimates of the left ventricle from multi-view reconstructions are found to be in better agreement with clinical measures than measures from single view reconstructions.Multi-view 3D reconstruction from sparse 2D freehand B-mode images leads to more accurate volume quantification compared to single view systems.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Electrical and Computer Engineering, University of New Mexico, Albuquerque, NM 87131, USA. pattichis@ece.unm.edu.

ABSTRACT

Background: A significant limitation of existing 3D ultrasound systems comes from the fact that the majority of them work with fixed acquisition geometries. As a result, the users have very limited control over the geometry of the 2D scanning planes.

Methods: We present a low-cost and flexible ultrasound imaging system that integrates several image processing components to allow for 3D reconstructions from limited numbers of 2D image planes and multiple acoustic views. Our approach is based on a 3D freehand ultrasound system that allows users to control the 2D acquisition imaging using conventional 2D probes.For reliable performance, we develop new methods for image segmentation and robust multi-view registration. We first present a new hybrid geometric level-set approach that provides reliable segmentation performance with relatively simple initializations and minimum edge leakage. Optimization of the segmentation model parameters and its effect on performance is carefully discussed. Second, using the segmented images, a new coarse to fine automatic multi-view registration method is introduced. The approach uses a 3D Hotelling transform to initialize an optimization search. Then, the fine scale feature-based registration is performed using a robust, non-linear least squares algorithm. The robustness of the multi-view registration system allows for accurate 3D reconstructions from sparse 2D image planes.

Results: Volume measurements from multi-view 3D reconstructions are found to be consistently and significantly more accurate than measurements from single view reconstructions. The volume error of multi-view reconstruction is measured to be less than 5% of the true volume. We show that volume reconstruction accuracy is a function of the total number of 2D image planes and the number of views for calibrated phantom. In clinical in-vivo cardiac experiments, we show that volume estimates of the left ventricle from multi-view reconstructions are found to be in better agreement with clinical measures than measures from single view reconstructions.

Conclusions: Multi-view 3D reconstruction from sparse 2D freehand B-mode images leads to more accurate volume quantification compared to single view systems. The flexibility and low-cost of the proposed system allow for fine control of the image acquisition planes for optimal 3D reconstructions from multiple views.

Show MeSH
Related in: MedlinePlus