Fast Super Resolution Ultrasound Imaging by Tracking of Erythrocytes using Di-erent Velocity Constraints
Publikation: Bidrag til tidsskrift › Konferenceartikel › Forskning › fagfællebedømt
One of the integral parts of super-resolution ultrasound imaging (SRI) is particle tracking, and researchers are trying to improve SRI with different tracking methods. This paper presents tracking for a new approach for SUper Resolution ultrasound imaging using Erythrocytes (SURE), which uses the erythrocytes as the target instead of fragile microbubbles (MBs). The acquisition of the SURE data can be accomplished in seconds due to the abundance of erythrocytes as targets. The nearest-neighbor (NN) algorithm was used to track erythrocytes, since this is one of the successful approaches for MB tracking in SRI. The erythrocyte targets were tracked to create SURE intensity map by three NN trackers with a constraint on the maximum velocities of 20, 40, 80 mm/s. Each tracker generated a different track map that depict different details. By combining the outputs of three trackers, and inserting them into one map, it was demonstrated that the combination of trajectories from different velocities carried more information from all the maps. Also, an image fusion method using discrete wavelet transform is applied on the intensity maps of these three dfferent velocities. Qualitative and quantitative experiments were conducted to exhibit the enhancements of the combined and the fused intensity map. Comparing three intensity maps with their combination and fusion and also a magnified area showed the fused one had better visual appearance than the others. Standard deviation (SD), average gradient (AG), entropy (H) and spatial frequency (SF) were used for quantitative comparison. According to SD, AG, H, and SF, the fused intensity map had 10%, 46%, 2%, and 46% higher scores compared to the combined intensity map. Also, the fused intensity map had 19% and 22% higher scores compared to the intensity map with maximum velocities of 80 mm/s according to AG and SF, respectively.
|Tidsskrift||Progress in Biomedical Optics and Imaging - Proceedings of SPIE|
|Status||Udgivet - 2022|
|Begivenhed||Medical Imaging 2022: Ultrasonic Imaging and Tomography - Virtual, Online|
Varighed: 21 mar. 2022 → 27 mar. 2022
|Konference||Medical Imaging 2022: Ultrasonic Imaging and Tomography|
|Periode||21/03/2022 → 27/03/2022|
|Sponsor||The Society of Photo-Optical Instrumentation Engineers (SPIE)|
The work is supported by European Research Council’s (ERC) Synergy Grant 854796.
© 2022 SPIE. All rights reserved.