UltraArt™ Universal Image Processing

UltraArt™ Universal Image Processing is a state of-the-art speckle reduction algorithm, developed by Siemens Healthineers to reduce speckle, enhance ridges of directional structures and sharpen edges prior to image interpretation without losing important features (e.g., small, low-contrast structures and texture information), thus improving image detail and contrast resolution.

White paper UltraArt Universal Image Processing ACUSON Redwood ultrasound system Mikyoung Park, MSEE, Siemens Medical Solutions USA, Inc., Ultrasound SIEMENS Healthineers White paper · UltraArt Introduction Due to the nature of coherent image formation of UltraArt universal image processing is a state of-the-art ultrasound, speckle is an inherent property of ultrasound speckle reduction algorithm, developed by Siemens images. Speckle is seen as a random granular pattern Healthineers to reduce speckle, enhance ridges of caused by the constructive and destructive interference directional structures and sharpen edges prior to image of back scattered echoes from scatterers much smaller interpretation without losing important features (e.g., than the ultrasound imaging wavelength. The presence small, low-contrast structures and texture information), of speckle degrades the detail and contrast resolution thus improving image detail and contrast resolution. of ultrasound images and ‘can make’ diagnosis more difficult. Speckle Reduction Techniques Various techniques have been developed and 2. Post-processing-based techniques implemented to reduce speckle in research institutions and the ultrasound industry. These techniques can be Speckle reduction achieved through single image categorized into two broad areas: (1) Compounding/ frame post-processing techniques include linear filtering Averaging-based techniques and (2) Post-processing [1–3], median filtering [4], Lee filtering [5], Wiener based techniques. filtering [6], anisotropic diffusion filtering [7–9], and wavelet denoising filtering [10–11]. Siemens Healthineers 1. Compounding/Averaging-based techniques DTCE [12], ContextVision [13], GE SRI [14] and Philips XRES [15] are commercially available proprietary post- Speckle reduction is achieved by combining uncorrelated processing techniques for this purpose. images. In spatial compounding, several overlapping uncorrelated component image frames are acquired Current speckle reduction techniques reduce speckle from different view angles, and then combined into a to various extents. However, it is common to see single compounded image. In frequency compounding, excessive blurring and an artificial image appearance. uncorrelated sub-images are obtained by either varying As a result, important features may be lost or obscured the center frequency on transmission or by dividing the after performing speckle reduction. spectrum of RF signals on reception, and recombining to form a frequency-averaged image. In temporal averaging (known as persistence), successive image frames that are uncorrelated through tissue or probe motion are combined into a single image. 2 UltraArt · White paper UltraArt universal image processing UltraArt universal image processing is a state-of-the- In terms of clinical workflow, UltraArt universal image art post-processing algorithm developed by Siemens processing is designed to improve: Healthineers utilizing our in-depth understanding • of ultrasound image characteristics and clinical Plunkability: reduces speckle and increases continuity applications / anatomies to improve diagnostic of specular reflectors, and thus improves contrast confidence. It is implemented not only for B-mode resolution but also for M-mode, Doppler and contrast modes • Usability: reduces complexity, redundancy and on the new ACUSON Redwood ultrasound system. interplay of image processing features, and unifies them under a simple and intuitive user interface • Enables robust image presentation and aesthetics User Interface (UI) Traditional post-processing user controls are often non- intuitive, complex and redundant which can prevent users from choosing an ideal combination. UltraArt universal image processing introduces a simple and intuitive image-based graphical user interface. Traditional user controls are replaced by a set of images, each corresponding to a different set of image processing parameters applied to a common input image. The user picks one of the images from the set, based on the desired outcome on the touch screen interface, which determines the UltraArt universal image processing parameters for the exam until a different selection is made. Figure 1 illustrates the UltraArt universal image processing UI. Figure 1: The ACUSON Redwood ultrasound system with the UltraArt universal image processing UI. 3 White paper · UltraArt A ? 10L4 Patient 2D Review Report End Exam Off Change ROI 2 Figure 2: Simple and intuitive image-based UI Clarify showing different levels of image Off enhancement. W Algorithm Description There are three major components of image enhance- directional structures. Image analysis identifies the ment for UtraArt universal image processing: speckle location, strength and direction of edges in the images reduction, ridge enhancement and edge sharpening. at multiple spatial scales. Edges are enhanced and/or Speckle is a granular pattern most noticeable within smoothed via adaptive sharpening/smoothing filters. echogenic area (e.g., tissue). Ridges are formed where Finally, local contrast is enhanced by amplifying the the gray value reaches local maxima in a given direction difference between the luminance values of each pixel (e.g., bright structures). Edges are borders between and its local region. areas of high and low gray values. Two edges can be detected at either side of a ridge. Sequoia Figure 2 illustrates the major image components. UltraArt universal image processing is designed to control the enhancement of these three major components independently so that ultrasound images can be customized to meet clinical needs. The philosophy of UltraArt universal image processing optimization is preserving information and maintaining natural tissue texture, rather than drastic filtering. Figure 3 illustrates the UltraArt universal image processing signal path. Speckle is reduced by utilizing speckle statistics and echogenicity similarity information while preserving edges of structures and, most importantly, preserving authentic tissue texture (e.g., liver tissue). Anisotropic filters are used to detect and enhance ridges of Figure 3: Example areas of speckle (yellow circle), ridges (red solid line) and edges (green dash line) are highlighted in the thyroid image above. 4 UltraArt · White paper Touch Screen Interface T n (m, n) m Input Speckle Ridge Edge Local Output Image Reduction Enhance Enhance Adaptive DR Image Figure 4: Block diagram describing independent processing components and intermittent results using UltraArt universal image processing. UltraArt Universal Image Processing Clinical Applications Abdomen Figure 5 shows a transverse cross section of (a) a liver with and improving contrast resolution in liver structures and hepatic vein and (c) a liver with longitudinal section of the the kidney (e.g., the echo differentiation between the kidney. Original images show a granular pattern in tissue continuity of cortex, medulla, and columns). Noise in the with noise observed inside the vein and kidney pyramid. vein and kidney pyramid is also suppressed, and structure In UltraArt universal image processing applied images, definition (vessel wall, diaphragm, interface between speckle is reduced while maintaining natural tissue texture kidney and liver, etc.) is improved. 5C1 5C1 Abdomen Abdomen TIB 0.21 TIB 0.21 TIC 1.20 TIC 1.20 TIS 0.21 TIS 0.21 MI 1.29 MI 1.29 28 fps 28 fps 98% 98% 2D 2D H High H High 0 dB -3 dB DR 70 DR 70 LD 3 LD 3 UA Off UA 3 -10cm z -10cm Z Fr 242 Fr 107 Figure 5: a) Original image using a curved transducer, (b) Image using UltraArt universal image processing. 5 White paper · UltraArt Thyroid Figure 6 shows thyroid images. In UltraArt universal while enhancing contrast resolution, details and edge image processing applied images, speckle in the sterno- structures. Structure boundaries (e.g., thyroid lobes, cleidomastoid muscle and thyroid gland is suppressed trachea, intima media) are better defined. Thyroid TIB 0.22 TIC 0.54 TIS 0.22 MI 1.15 H High TEQ 3 dB EO 3 9P DR 65 LD 1 UA Off 3cm 3cm Fr 454 Figure 6: a) Original image using a linear transducer, (b) Image using UltraArt universal image processing. Figure 7 shows (a) four-chamber echocardiography Contrast resolution in the differentiation of the image. In UltraArt universal image processing applied myocardium, endocardium, and pericardium is improved, image (b), clutter noise in the chamber is well and the continuity of the ticuspid and mitral valves is suppressed as compared to the original image (a). better defined. Cardiac TIB 0.90 Cardi TIC 3.26 TIC 3.2 TIS 0.90 MI 1.14 MI 1.1 51 fps 2D H High TEQ 1 dB DR 60 02 Fr 1062 42bpm Lead Il 45bpm Lead Il 77 01 2020 15 2020.01.17 15:53:09 Figure 7: a) Original image using a vector transducer, (b) Image using UltraArt universal image processing. 6 UltraArt · White paper Conclusion UltraArt universal image processing is an excellent diagnostic confidence and consistency across different ultrasound image processing algorithm that improves users by producing images of higher quality and by image quality across clinical applications. Clinical avoiding improper combinations of individual post- evaluations suggest that it reduces speckle while processing parameters. By allowing real time modification maintaining natural tissue texture, improves the of the level of speckle in the image, UltraArt universal continuity of structures, and creates better defined image processing empowers the user to customize the borders. All these enhancements improve image detail image to their unique preferences. In this way, UltraArt and contrast resolution. universal image processing on the ACUSON Redwood system helps to expand precision medicine. Plunkability and usability improvements by UltraArt universal image processing can help in improving Please note that the learning material is for training purposes only! For the proper use of the software or hardware, please always use the Operator Manual or Instructions for Use (hereinafter collectively “Operator Manual”) issued by Siemens Healthineers. This material is to be used as training material only and shall by no means substitute the Operator Manual. Any material used in this training will not be updated on a regular basis and does not necessarily reflect the latest version of the software and hardware available at the time of the training. The Operator's Manual shall be used as your main reference, in particular for relevant safety information like warnings and cautions. Note: Some functions shown in this material are optional and might not be part of your system. Certain products, product related claims or functionalities (hereinafter collectively “Functionality”) may not (yet) be commercially available in your country. Due to regulatory requirements, the future availability of said Functionalities in any specific country is not guaranteed. Please contact your local Siemens Healthineers sales representative for the most current information. The reproduction, transmission or distribution of this training or its contents is not permitted without express written authority. Offenders will be liable for damages. All names and data of patients, parameters and configuration dependent designations are fictional and examples only. All rights, including rights created by patent grant or registration of a utility model or design, are reserved. Copyright © Siemens Healthcare GmbH 2020 Siemens Healthineers Headquarters\Siemens Healthcare GmbH\Henkestr. 127\ 91052 Erlangen, Germany\Telephone: +49 9131 84-0\ 7 References 1. Dave Hale, “Recursive gaussian filters”, CWP-546, 2006. 8. K. Abd-Elmoniem, A.-B. Youssef, Y. Kadah, ‘‘Real-time speckle 2. L. Chen, G. Lu and D. Zhang, ‘’Effects of different gabor filter reduction and coherence enhancement in ultrasound imaging parameters on image retrieval by texture’’, In Proc. of IEEE via nonlinear anisotropic diffusion,’’ IEEE Trans. Biomed. Eng., 10th International Conference on Multi-Media Modelling, pp. vol. 49, no. 9, pp. 997–1014, 2002. 273-278, 2004. 9. Y. Yongjian, S.T. Acton, ‘‘Speckle reducing anisotropic diffusion,’’ 3. V. Shiv Naga Prasad and Justin Domke, ‘’Gabor filter IEEE Trans. Image Process., vol. 11, no. 11, pp. 1260–1270, visualization’’, Technical Report, University of Maryland, 2005. 2002. 4. T. Huang, G. Yang, G. Tang, ‘‘A fast two-dimensional median 10. S. Zhong, V. Cherkassky, ‘‘Image denoising using wavelet filtering algorithm,’’ IEEE Trans. Acoustics Speech Signal Proces., thresholding and model selection,’’ Proc. IEEE Int. Conf. Image vol. 27, no. 1, pp. 13–18, 1979. Process., Vancouver,Canada, pp.1–4, 2000. 5. Lee JS. “Speckle analysis and smoothing of synthetic aperture 11. A. Achim, A. Bezerianos, P. Tsakalides, ‘‘Novel Bayesian RADAR images.” Comput Graph Image Process. 17(1): 24–32, multiscalemethod for speckle removal in medical ultrasound 1981. images,’’ IEEE Trans.Med. Imag., vol. 20, no. 8, pp. 772–783, 2001. 6. Robert S. Caprari, Alvin S. Goh, and Emily K. Moffatt, “Noise and speckle reduction in synthetic aperture radar imagery by 12. nonparametric Wiener filtering,” Applied Optics. Vol.39, Issue 13. 35, pp.6633-6640, 2000. 14. 7. S. Jin, Y. Wang, J. Hiller, ‘‘An adaptive non-linear diffusion algorithm for filtering medical images,’’ IEEE Trans. Inform. 15. Technol. Biomed., vol. 4, no. 4, pp. 298–305, 2000. Standalone clinical images may have been cropped to better visualize pathology. The products/features mentioned in this document may not be commercially available in all countries. Due to regulatory reasons, their future availability cannot be guaranteed. Please contact your local Siemens Healthineers organization for further details. ACUSON Redwood and UltraArt universal image processing are trademarks of Siemens Medical Solutions USA, Inc. Siemens Healthineers Headquarters Legal Manufacturer Siemens Healthcare GmbH Siemens Medical Solutions USA, Inc. Henkestr. 127 Ultrasound 91052 Erlangen, Germany 22010 S.E. 51st Street Phone: +49 9131 84-0 Issaquah, WA 98029, USA Phone: 1-888-826-9702 Published by Siemens Medical Solutions USA, Inc. · Order No. A91US-633-1C-4A00 · Printed in Germany · 8584 0920 © Siemens Medical Solutions USA, Inc., 2020

  • ultraart
  • Dynamic range
  • speckle
  • Ultrasound 2D