Blood noise reduction in intravascular ultrasound imaging

Abstract

Scattering from red blood cells (blood noise) increases significantly as the ultrasound frequency is increased above 10 MHz. This reduces the contrast between the vessel wall and the lumen in intravascular ultrasound imaging which makes it difficult to localize the vessel wall and plaque. A blood noise filter based on beam tilting and digital lateral low pass filtering is described. Beam tilting introduces a Doppler shift from blood which results in a frequency separation of the vessel wall signal and the blood noise. The performance of the filter is investigated by simulations and by in vitro experiments. The filter is found to be effective for blood velocities exceeding approximately 50 cm s 01 at a 20 MHz ultrasound frequency with a beam tilt angle of 10 degrees and a frame rate of 15 f.p.s. By increasing the system frequency to 40 MHz, increase the beam tilt angle to 15 degrees and reduce the frame rate to 10 f.p.s., the filter is effective for blood velocities below 10 cm s 01 .

Multi-frequency processing for lumen enhancement with wideband intravascular ultrasound

“…Most of the previously reported IVUS lumen enhancement techniques have taken advantage of the fast and random frame-to-frame movement of blood speckles in contrast to the slow and consistent motion from the tissue [1][2][3]. This motion contrast is also the most important visual cue to physicians unless saline is injected to flush out the lumen.…”

“…One technical challenge of using 40 MHz or higher frequencies is the strong scattering from blood that may obscure the arterial lumen and make image interpretation difficult. To overcome this issue, several groups have reported signal or image processing techniques that utilize the time-varying characteristics of the flowing blood in contrast to the stationary arterial tissues [1][2][3]. These motionbased processing techniques, however, are not as effective under lower blood flow conditions.…”

3D ultrasound in cardiology

“…Various sources of noise, including thermal noise of the amplification circuits, acoustic noise, phase effects (noise speckle), as well as the type, size and texture of the transducer, constitute elements of distortion that generate artifacts in ultrasound images [3] . However, image quality, thanks to technological evolution, is continuously improving [4][5][6][7][8][9] . The greatest advantage of 2D ultrasound is its flexibility, which allows the specialist to manipulate the transducer and visualize the desired ana-tomical section.…”

A State-of-the-Art Review on Segmentation Algorithms in Intravascular Ultrasound (IVUS) Images

“…The term blood noise is associated with scattering from red blood cells inside the lumen. To the best of our knowledge, the authors in [79] were the first to employ signal-processing techniques in lateral direction for BNR in IVUS images acquired with a 20-MHz transducer. The framework exploited the Doppler shift in the blood power spectrum which could be teased out from the vessel wall spectrum centered at the zero frequency using a low-pass filter.…”