Ryuichi Shinomura

Diagnostic results for breast disease by real-time elasticity imaging system

We have previously reported our real-time elasticity imaging system and a preliminary application to breast tissue diagnosis. We now propose several post processing algorithms to stabilize the elasticity imaging, introduce a method for scoring its clinical usefulness, and report the current status of the scoring method on breast tissue diagnosis. Our newly implemented post processing algorithms are: (1) frame-to-frame smoothing; (2) adaptive contrast optimization; (3) noisy-frame rejection; (4) noisy-region reduction. Using these algorithms, more than 20% of intensity fluctuations (SD) in strain images can be reduced. Our newly introduced scoring method is based on the imaging pattern of the low-strain region inside the hypoechoic region in the B-mode image. We classify 5 grades of elasticity score ranging from 1 (no strain-zero brightness region; benign) to 5 (broader strain-zero brightness region; malignant). As the result of applying 137 patients with breast diseases, this method provides a sensitivity of 87%, a specificity of 91%, and an accuracy of 89%.

A 120-MHz Ultrasound Probe for Tissue Imaging

Ultrasound transducers with center frequency above 100 MHz are expected to be used for future diagnostic tissue characterization because of their high lateral resolution. We have fabricated a 120-MHz transducer that consists of a ZnO piezoelectric film on a sapphire substrate that has a concave acoustic lens. The lateral resolution was calculated as 13 microns. The insertion loss of the transducer, defined as the difference between the received voltage and the transmitted one, was -45 dB. The 6-dB handwidth of the received signal was approximately 40 MHz. The transducer was mounted in a rod-shaped probe to ensure contact with in vivo tissue, because of the low penetration of ultrasound in the high frequency region. While the probe is rotated and moved along its axis mechanically, the transducer receives backscattered ultrasound from the surrounding tissue on a cylindrical plane that is kept a constant distance from the probe surface. The feasibility of this high-frequency tissue imaging probe has been demonstrated by obtaining preliminary images of an in vitro bovine kidney.

Multi‐element ultrasonic transducer

A multi-element ultrasonic transducer in which elements are arrayed and in which a plate-shaped piezoelectric material has its one face formed with a uniform electrode and its other face formed alternately with electrodes corresponding to the respective elements and electrodes for separating the elements. These electrodes for the element separation are connected the uniform electrode opposed thereto and is fed with a ground potential. On the other hand, the electrodes corresponding to the respective elements are fed individually with transmitting and receiving signals independently of the elements so that the electronic scanning or focusing operations can be achieved.

Two-directional phase aberration correction using a two-dimensional array

The aim of this paper is to show the effectiveness of two-directional phase aberration correction using real-time signals obtained from a human body. The authors constructed a two-dimensional array and the real-time data acquisition system for this purpose. The array has ten elements in both the scan and elevation directions. It is placed in contact with a human body and signals reflected from the subjects liver are digitized by parallel A/D converters and stored in memory. The data acquisition for each image is completed in 52.4 ms. A cross-sectional image of the liver is then constructed off-line with a computer. Images of the liver before and after the phase aberration correction show that the image was significantly improved after the compensation.

Intracorporeal imaging and differentiation of living tissue with an ultra-high-frequency ultrasound probe

Intraoperative diagnostic tissue differentiation is expected to be useful clinically. We have fabricated a 3-mm diameter rod-shaped ultrasound (US) probe mounted with a 120-MHz transducer whose lateral resolution is the same as the cellular size of 13 microm. The probe can image a microscopic structure (i.e., the cellular arrangement inside intracorporeal living tissue). We imaged normal kidney tissue of a living mouse and tumor tissue implanted in another mouse kidney. We anesthetized the mice, exteriorized the kidneys, and punctured the kidneys with the probe. Renal corpuscle-like structures were seen in the healthy kidney, but a wavy spindle-like structure was seen in the tumor. The similarity between the ultrasonic images and histological sections taken from the imaged organs demonstrates the possibility of real-time tissue differentiation by ultra-high-frequency US.

Frequency scanning, front viewing, alternating polarity transducer array for ultrasound imaging system

A frequency scanning beam steering technique using an alternating polarity transducer array is investigated for presentation of B-mode ultrasound images for front viewing intracavitary applications. Arrays of alternating polarity PZT elements were computer simulated and fabricated to characterize their beam steering and resolution capabilities. A spherical lens was mounted in front of the array to improve focusing. A personal computer based system was used to test and generate sector images of wire targets using these arrays.

Estimation of pressure-distribution effects upon elasticity imaging

We have already proposed the real-time strain imaging techniques (ca. 12 fps) for stable diagnosis in freehand and image classification method for the breast tissue diagnosis. For the objective diagnosis, most important problem is come from the fact that the strain image patterns were strongly depending on the magnitude and uniformity of the compression. In this paper, we investigate the pressure-distribution effects on the elasticity imaging using the tissue-mimicking elastography phantoms. We prepared 5 types of elastography phantoms with different stiffness. Phantoms #1-#3 have uniform stiffness overall in the rectangular solids and #4-#5 include small rectangular solids with different stiffness from surroundings. The phantom stiffness was estimated from the distribution of pressure and induced strain using our elasticity imaging system as follows. (1) The Youngs moduli of first three phantoms were estimated as 6±1 kPa, 27±4 kPa, and 108±23 kPa, respectively. (2) In the case of phantoms #4 and #5, outer parts were estimated as 6±1 kPa and 8±2 kPa, and inclusion parts were measured as 12±1 kPa and 58±14 kPa, respectively. (3) For the special case of above phantom #1, we applied non-uniform pressure (3 times difference between each lateral limits). Even in that case, we successfully reduced the effect of non-uniformity from 235 % to 5 % in elasticity imaging.

Ultrasound imaging system using combinational coded excitation

Increasing penetration to a deep body is one of the most important targets for clinical ultrasound imaging. To this end, coded excitation technique is known as a fundamental solution to improve signal to noise ratio (SNR) without degradation of axial resolution. In order to improve the SNR, the coded excitation requires a longer pulse containing many wavelengths. However, a nonlinear characteristic of signal processing such as dynamic beam forming degrades the resolution. Decoding error tends to appear at different distances where focusing parameters vary at the long pulse decoded. At a previous symposium, our group proposed a new approach, sub-aperture decoding, to enhance performance of coded excitation, in which the combinational codes with multiple codes with relatively small size taps are transmitted and then received signals are decoded with these combinational codes. In other words, this approachs transmission combines and compounds two or more codes of relatively small code length, and decoding processes divide into two or more decoding filters correspond to each code. In this report, we report more detailed simulation and preliminary experimental results for this approach of combinational coded excitation. Using this new coding approach, higher SNR without degrading lateral resolution is achieved. In the case of 3/spl times/5 codes, the combinational coding technique improves SNR up to 11.4 dB experimentally and this result is relatively coincidental with calculated value (11.5 dB).

Clinical evaluation of real-time phase-aberration correction system [medical ultrasound]

We have developed a phase-aberration correction system that correlates signals in real time. In this paper we evaluate the clinical performance of the system in vivo. This system 1) constructs a cross-sectional image and, 2) calculates the correlation of signals at neighboring sensor elements. Both 1) and 2) are carried out in real time. The system was used for imaging of living tissue. First the beam was formed using initial focus delay settings and a real-time cross-sectional image was constructed. The correlation between neighboring signals was calculated simultaneously with the beam formation and the time difference between the pairs of signals was acquired. The time difference was then used to compensate for the initial delay. The image of the living tissue was substantially improved after the compensation. A further experiment is in progress to collect a statistically significant number of clinical results.

Optimum correlator arrangement for two-directional phase-aberration correction [in medical US imaging]

We have determined by computer simulation the optimum correlator arrangement when correcting two-directional phase-aberration with a two-dimensional (2D) array. In the simulation, a 2D phase screen was estimated from the correlation of neighboring signals in the 2D transducer. The difference between a given phase screen and the corresponding estimated phase screen was evaluated to find the optimum correlator arrangement. The optimum arrangement is to locate an equal number of correlators in both the scan and elevation directions with the distance between the neighboring correlators being 0.2 to 0.4 times the phase-screen correlation length.