Analysis and Visualization of Tissue-motion of Cranial Ultrasonogram Image Sequences for Newborn Babies

Abstract

Cranial ultrasound scans are very essential part of the routine investigation of neonatal intensive care. The scan ultrasound image sequence are not only used for real-time diagnosis but also recorded as moving images in video recorder as ultrasonographic movie for fitture diagnosis, offline analysis of ultrasonogram images and the study of tissue velocity in neonatal cranium. The tissue motion in neonatal cranium is an important physical parameter which is considered in discussing the pulsation strength of newborn baby for pediatric diagnosis. The artery pulsation has a strong correlation with the blood flow in the newborn baby head and tissue-motion has a relationship with artery pulsation. Optical flow technique finds an excellent application in determining the tissue motion A velocity quantitatively in cranial ultrasonograrn of newborn baby. Cranial ultrasonograrn of newborn babies for different coronal and sagittal sections are studied and the tissue motion velocity of neonatal cranium is analyzed using the optical flow techniques. In order to calculate tissue motion velocity, gradient-based approaches of optical flow techniques with different optical flow optimizations are used. Ultrasonogram image sequences of 32 frames, 640 x 480 pixels/frame, 8bits/pixel, 33 ms/frame for different coronal and sagival sections are used. Whole ultrasonogram is not useful for analysis and small portion of ultrasonogram images are selected during analysis in the region inside cranial bone. Tissue- motions are estimated in different coronal sections and their errors are also estimated. Further, the time variant tissue-motions are analyzed using discrete Fourier transform of optical flow velocity. The pulsation is observed in the time variant tissue motion images and strong pulsation is occurred in the harmonic frequencies of tissue-motion that has a relation to the heartbeat frequency of a newborn baby which is helpful for pediatric diagnosis. Pulsation amplitudes, caused by artery pulsation due to blood now are also analyzed and visualized from a video stream of ultrasound image sequence by using I D fast Fourier transform in brightness mode for future diagnosis using direct pixel value. A new imaging technique, named pulsation amplitude image is proposed. In the proposed method, the tissue motion of typical coronal and sagittal sections of normal and asphyxiated neonates are analyzed and visualized. Significant differences have been observed in cranial tissue motion between normal and abnormal neonate. The results obtained in the present study lead to determine pulsation amplitude that strongly contribute pediatricians in diagnosis of newborn baby’s ischemic diseases.