부산대학교 도서관

Ultrasonography is the primary imaging modality used for diagnostic assessment of fetal and maternal health during pregnancy. However, a large proportion of fetal ultrasound exams suffer from inadequate or lack of visualization of clinically relevant structures due to acoustical noise or clutter. These undesirable effects are even more pronounced in overweight and obese mothers. In previous work our group has introduced short-lag spatial coherence (SLSC) and harmonic spatial coherence imaging (HSCI) as methods for clutter reduction. These have been shown to be successful in improving image contrast and border delineation in both simulations and in vivo experiments on liver and cardiac imaging. We extend the application of these techniques to fetal ultrasound imaging. We collected individual channel RF data on 11 volunteers in their first trimester of pregnancy. The amount of intrinsic clutter varied across the volunteers. These data sets were used to generate matched fundamental B-mode, SLSC, harmonic B-mode, and HSCI images. These images were compared qualitatively by assessing image texture and target detectability and quantitatively using SNR, CNR, and contrast. SLSC and HSCI images showed significant improvements across all imaging metrics compared to B-mode and harmonic B-mode, respectively. The improvements in clutter suppression and improved conspicuity of target structures were greatest for poor quality B-mode images. Clutter levels in these images, measured in a region of amniotic fluid compared to a region of uniform speckle, were measured to be 15 dB for B-mode versus 27 dB for SLSC, and 17 dB for harmonic B-mode versus 30 dB for HSCI. CNR for the same set of images improved from 1.4 to 2.5 for B-mode to SLSC and from 1.4 to 3.1 for harmonic B-mode to HSCI.