William D. Richard

Frameless stereotactic ultrasonography: Method and applications

In stereotactic neurosurgery, computed tomography (CT) and magnetic resonance (MR) images are registered in a coordinate system defined with respect to the skull. By intraoperatively tracking the coordinate position of a surgical instrument, various displays can be formed which show the position of the instrument in the MR and/or CT images. However, the accuracy of this display varies because intracranial structures may shift or warp from their position prior to surgery. Ultrasonic imaging systems provide real-time images of the brain, but structures in these images are difficult to interpret because the images are based on ultrasonic echoes. A method has been developed for the real-time registration of these images. With this registration, software continuously updates a corresponding image constructed from the set of MR and/or CT images used for guidance. By developing this second view of the structures in the ultrasound image, the surgeon can easily interpret the ultrasound image, and it becomes possible to determine the extent of the intra-operative structure shift between the two images.

Automated texture-based segmentation of ultrasound images of the prostate

Segmenting two-dimensional images of the prostate into prostate and nonprostate regions is required when forming a three-dimensional image of the prostate from a set of parallel two-dimensional images. The texture-based segmentation method presented here is a pixel classifier based on four texture energy measures associated with each pixel in the image. An automated clustering procedure is used to label each pixel in the image with the label of its most probable class. The segmented images produced as the result of applying the algorithm to an example image are presented and discussed. The automated segmentation algorithm has been found to hold promise as an automated segmentation method.

Segmenting ultrasound images of the prostate using neural networks

This paper describes a method for segmenting transrectal ultrasound images of the prostate using feedforward neural networks. Segmenting two-dimensional images of the prostate into prostate and nonprostate regions is required when forming a three-dimensional image of the prostate from a set of parallel two-dimensional images. Three neural network architectures are presented as examples and discussed. Each of these networks was trained using a small portion of a training image segmented by an expert sonographer. The results of applying the trained networks to the entire training image and to adjacent images in the two-dimensional image set are presented and discussed. The final network architecture was also trained with additional data from two other images in the set. The results of applying this retrained network to each of the images in the set are presented and discussed.

A High-Frame Rate High-Frequency Ultrasonic System for Cardiac Imaging in Mice

We report the development of a high-frequency (30-50 MHz), real-time ultrasonic imaging system for cardiac imaging in mice. This system is capable of producing images at 130 frames per second (fps) with a spatial resolution of less than 50 mum. A novel mechanical sector probe was developed that utilizes a magnetic drive mechanism and custom-built servo controller for high speed and accuracy. Additionally, a very light-weight (< 0.28 g), single-element transducer was constructed and used to reduce the mass load on the motor. The imaging electronics were triggered according to the angular position of the transducer in order to compensate for the varying speed of the sector motor. This strategy ensured the production of equally spaced scan lines with minimal jitter. Wire phantom testing showed that the system axial and lateral resolutions wore 48 mum and 72 mum, respectively. In vivo experiments showed that high-frequency ultrasonic imaging at 130 fps is capable of showing a detailed depiction of a beating mouse heart.

Real-Time Ultrasonic Scan Conversion via Linear Interpolation of Oversampled Vectors

Scan conversion is required in order to display conventional B-mode ultrasonic signals, which are acquired along radii at varying angles, on standard Cartesian-coordinate video monitors. For real-time implementations, either nearest-neighbor or bilinear interpolation is usually used in scan conversion. If the sampling rate along each radius is high enough, however, the gray-scale value of a given pixel can be interpolated accurately using the nearest samples on two adjacent vectors. The required interpolation then reduces to linear interpolation. Oversampling by a factor of 2 along with linear interpolation was superior to bilinear interpolation of vectors sampled to match pixel-to-pixel spacing in 6 representative B-mode images. A novel 8-bit linear interpolation algorithm was implemented as a CMOS VLSI circuit using a readily available, high-level synthesis tool. The circuit performed 30 million interpolations per second. Arithmetic results produced by the 8-bit interpolator on 7-bit samples were virtually identical to IEEE-format, single-precision, floating-point results.

A High-Frame Rate Duplex Ultrasound Biomicroscopy for Small Animal Imaging In vivo

Much of the current knowledge of human cardiovascular pathologies and treatment strategies has been gained from understanding the cardiac physiologies and functions in small animal models, such as mice, rats, and zebraflsh. In this paper, we present the development of a high-frame-rate duplex ultrasound biomicroscopy (UBM) capable of B-mode imaging and pulsed-wave (PW) Doppler measurement for in vivo cardiovascular investigation in small animals. A frame rate of 200 frames per second (fps) was accomplished at a view of 5 mm times 8 mm, using a novel highspeed sector probe and specially designed lightweight transducers. In a reduced lateral view of 1.2 mm, a frame rate of 400 fps was achieved to examine more detailed cardiac motion. The UBM utilized transducers with different center frequencies (40-75 MHz) and geometries, which made it useful for various applications in small animal cardiac imaging. The highest spatial resolution the UBM achieved was 25 mum times 56 mum. In addition, the image-guided PW Doppler implemented in the UBM demonstrated the detection of the velocity of a moving wire as low as 0.1 mm/s, and flow in a polyimide tube as small as 200 mum in diameter. Furthermore, the UBM achieved a 15-muV minimal detectable signal and a 60-dB dynamic range using a low-cost PCB-based design. Finally, sample in vivo cardiac images of mouse and zebraflsh hearts were given. These results showed that the UBM integrated with B-mode and PW Doppler is useful to investigate the pathophysiological mechanism in the cardiovascular studies.

The Smart Port Card: An Embedded Unix Processor Architecture for Network Management and Active Networking

This paper describes the architecture of the Smart Port Card (SPC) designed for use with the Washington University Gigabit Switch. The SPC uses an embedded Intel Pentium processor running open-source NetBSD to support network management and active networking applications. The SPC physically connects between a switch port and a normal link adapter, allowing cell streams to be processed as they enter or leave the switch. In addition to the hardware architecture, this paper describes current and future applications for the SPC.

The Washington University multimedia system

The Washington University multimedia system (MMS) is a complete multimedia system capable of transmitting and receiving JPEG-compressed video, CD-quality audio, and high-resolution radiological images, in addition to normal network traffic, over the Washington University broadband ATM network. The MMS consists of an ATMizer and three multimedia subsystems. The ATMizer implements the host interface, the interface to the ATM network, and the interface to the three multimedia subsystems. This paper describes the architecture of the MMS, the software used with the system, and the applications which have been developed to demonstrate the capability and applicability of broadband ATM networks for multimedia applications.

A low-cost B-mode USB ultrasound probe.

The Universal Serial Bus (USB) is now the ubiquitous interface bus of choice for connecting peripherals to personal computers and laptops. USB 2.0 is a half-duplex bus running at 480 Mb/s and each peripheral can draw as much as 500 mA of current at a nominal 5 V from the USB connector. We have developed a family of USB-based, B-mode probes that connect directly to a personal computer or laptop and that draw as little as 250 mA (1.25 W) when forming ten 5 MHz images/second. The pulser/receiver, high voltage supply, analog-to-digital converter, servo and USB interface are implemented on a small circuit board inside the probe body. After raw data are transferred to the computer, gain compensation, interpolation, filtering and other data processing are performed by the host processor. This gives flexibility to developers and allows enhancements to the system to be incorporated via software updates. In addition, the raw data are available for storage and later postprocessing. There are several advantages to this architectural approach to B-mode imaging, including low cost, portability and optimal signal-to-noise performance. This paper describes the advantages of the architecture of the probe family, discusses the hardware/software division of the required processing steps and presents example images from a 12.5 MHz ophthalmic probe.

A method for three-dimensional prostate imaging using transrectal ultrasound

This paper describes a method for forming three-dimensional images of the prostate using transrectal ultrasound. This method extracts three-dimensional images of the prostate from sets of two-dimensional ultrasound images obtained via a special-purpose transrectal ultrasound probe. Each two-dimensional image is segmented and the results used to form a three-dimensional image of the prostate. A method for segmenting two-dimensional images of the prostate based on the Laplacian-of-Gaussian edge operator is described. The three-dimensional imaging method described provides a new, noninvasive method for monitoring gland pathology during radiation therapy.