Todd Fjield

Design and evaluation of a feedback based phased array system for ultrasound surgery

A driving system has been designed for phased array ultrasound applicators. The system is designed to-operate in the bandwidth 1.2 to 1.8 MHz, with independent channel power control up to 60 W (8 bit resolution) for each array element. To reduce power variation between elements, the system utilizes switching regulators in a feedback loop to automatically adjust the DC supply of a class D/E power converter. This feedback reduces the RF electrical power variation from 20% to 1% into a 16 element array. DC-to-RF efficiencies close to 70% for all power levels eliminates the need for large heat sinks. In addition to power control, each channel may be phase shifted 360/spl deg/ with a minimum of 8 bit resolution. To ensure proper operation while driving ultrasound arrays with varying element sizes, each RF driving channel implements phase feedback such that proper phase of the driving signal is produced either at the amplifier output before the matching circuitry or after the matching circuitry at the transducer face. This feedback has been experimentally shown to increase the focal intensities by 20 to 25% of two tested phased arrays without array calibration using a hydrophone.

The combined concentric-ring and sector-vortex phased array for MRI guided ultrasound surgery

MRI guided ultrasound surgery requires small surgical equipment volumes to facilitate the treatment of larger patients in the limited space of a conventional MRI magnet. In addition, large focal volumes are required to reduce the treatment time of large tumors. The concentric-ring array is capable of moving the focus in one dimension, and previous studies have shown that a circular array composed of radial sectors is capable of producing enlarged focal volumes. These two array designs may be combined to create an array that is capable of both enlarging the focus and moving the focus along the axis of the array. Simulations were performed to predict the performance and capabilities of various combined array designs by using numerical routines to, calculate the acoustic power field, temperature distribution, and accumulated thermal dose. The results shown predict that the combined array can create necrosed tissue volumes over 30 times larger than the concentric-ring array while maintaining focal range. The simulation results were verified with an experimental array consisting of 13 rings and 4 sectors. In addition, simulations were performed where multiple focal patterns were cycled in the time domain to create an optimized heating pattern characterized by uniform thermal dose over the volume of the lesion. Such heating patterns resulted in a 40/spl deg/C lower maximum temperature compared to single mode sonications while producing the same necrosed tissue volume, and yielded a rate of necrosis of 26.4 cm/sup 3//h.

Feasibility of using ultrasound phased arrays for MRI monitored noninvasive surgery

The purpose of this paper was to evaluate the in vivo feasibility of using phased arrays for MRI guided ultrasound surgery. Two different array concepts were investigated: a spherically curved concentric ring array to move the focus along the central axis and a spherically curved 16 square element array to make the focus larger. Rabbit thigh muscles were exposed in vivo in a 1.5 T MRI scanner to evaluate the array performance. The results showed that both of the arrays performed as expected, and the focus could be moved and enlarged. In addition, adequate power could be delivered from the arrays to necrose in vivo muscle tissue in 10 s. This study was the first implementation of phased arrays for MRI guided ultrasound surgery. The results demonstrate that phased arrays have significant potential for noninvasive tissue coagulation.

A parametric study of the concentric‐ring transducer design for MRI guided ultrasound surgery

Noninvasive surgery using high-powered, focused ultrasound transducers in conjunction with magnetic resonance imaging has been shown to be feasible in previous studies. For clinical treatments, the geometry of standard MRI equipment limits the space available for ultrasound surgical equipment. This space requirement can be reduced in one dimension by using phased arrays to control the focal depth, thus eliminating the space required for the motion of a fixed focus transducer. Because of its symmetry, an annular array is ideal for changing the focal depth. Previous works have simulated, built, and characterized various concentric-ring transducers; however, no study has thoroughly examined the potential and limitations of the concentric-ring design for MRI guided ultrasound surgery. The present work is a systematic examination of the capabilities of the concentric-ring array, using numerical simulations to predict the power field, temperature distribution, and accumulated thermal dose. The results presented here illustrate the effects of ring size, center-to-center spacing configurations, number of rings, and radius of curvature on transducer performance. A 10-cm radius of curvature transducer with 14 evenly spaced rings has been built and characterized in order to verify the accuracy of the numerical simulations. The pressure-squared fields produced by this transducer are in excellent agreement with the simulated fields.

Comparison of modelled and observed in vivo temperature elevations induced by focused ultrasound: implications for treatment planning.

Two numerical models for predicting the temperature elevations resulting from focused ultrasound heating of muscle tissue were tested against experimental data. Both models use the Rayleigh-Sommerfeld integral to calculate the pressure field from a source distribution. The first method assumes a source distribution derived from a uniformly radiating transducer whereas the second uses a source distribution obtained by numerically projecting pressure field measurements from an area near the focus backward toward the transducer surface. Both of these calculated ultrasound fields were used as heat sources in the bioheat equation to calculate the temperature elevation in vivo. Experimental results were obtained from in vivo rabbit experiments using eight-element sector-vortex transducers at 1.61 and 1.7 MHz and noninvasive temperature mapping with MRI. Results showed that the uniformly radiating transducer model over-predicted the peak temperature by a factor ranging from 1.4 to 2.8, depending on the operating mode. Simulations run using the back-projected sources were much closer to experimental values, ranging from 1.0 to 1.7 times the experimental results, again varying with mode. Thus, a significant improvement in the treatment planning can be obtained by using actual measured ultrasound field distributions in combination with backward projection.

Design and experimental verification of thin acoustic lenses for the coagulation of large tissue volumes

Large focal volumes are desired in ultrasound surgery to reduce the total treatment time when large tumours are thermally coagulated. Phased arrays are capable of producing enlarged focal volumes in addition to providing the ability for on-line modification of focal shape and location. Although phased arrays have several advantages over their non-phased counterparts, the complexity of these arrays also presents some disadvantages regarding cost and complexity. One less costly alternative is the use of thin acoustic lenses to alter the field shape of a single-focus transducer. Four polystyrene lenses have been designed using the sector-vortex principle developed for phased arrays by Cain and Umemura. Measurements of the acoustic fields produced with the lenses are in good agreement with the simulated fields. The transmission measurements through each of the four lenses ranged from 76% to 84%, and over 52 W of total acoustic power has been delivered through each of the lenses during in vivo experiments without any damage to the lenses or the transducer. The in vivo results showed an increase in rate of necrosis to 10.1 +/- 1.4 cm3 h-1 using the mode 4 lens, or 5.2 +/- 0.7 times higher than the focused transducer alone.

Low-profile lenses for ultrasound surgery

Several flat lens designs have been simulated that focus a planar transducer at various depths and angular deflections. The use of flat lenses with planar phased arrays to create multiple focus patterns has also been explored. The simulations have shown that the discrete element size is practical to machine. Simulations predict that a flat polystyrene lens with an f number of 1.0 can produce a focal peak intensity that is 80% that of a geometrically focused transducer with the same focal depth, including the attenuation through the lens, by utilizing four discrete phase steps. These predictions were verified with a simplified experimental lens utilizing only two phase steps. Simulations predicted an attenuation within the lens of 18% and a focal peak intensity of 38%. Measurements resulted in an attenuation of 30% and a focal peak intensity of 30%. Experimental lens studies have indicated that sufficient power can be transmitted through the lenses for these designs to be feasible for use in ultrasound surgery.

In vivo verification of the acoustic model used to predict temperature elevations for MRI guided ultrasound surgery

Previous studies have shown that larger focal volumes created by phased-arrays result in lower overall treatment times when multiple sonications are required to necrose large tissue volumes. To fully utilize and optimize complex phased arrays, accurate theoretical models are required for prediction of the temperature elevation in vivo. So far, numerical simulation models have been shown to be relatively accurate for predicting coagulated tissue volumes (C. Damiano et al., 1995). However, there have not been studies investigating the accuracy of the actual temperature elevations in vivo. In the current experiment, an 8 element sector-vortex array has been constructed and tested in rabbit thigh in vivo. Temperature sensitive MRI imaging sequences were used to monitor the temperature elevations. A comparison of the theoretically calculated temperature elevations with the temperature elevations as monitored with the MRI resulted in a discrepancy. The theoretical temperature elevation for a single focus pattern was 2.7/spl plusmn/19% times higher than the measured, while for a mode 4 sonication, the theoretical value was only 1.1/spl plusmn/21% higher than the measured. These results indicate that the theoretical model used to predict temperature elevations in vivo neglects some phenomenon that is dependent on focal volume.

Experimental verification of the sectored annular phased array for MRI guided ultrasound surgery

To meet two of the requirements for MRI Guided Ultrasound Surgery, namely small surgical equipment and large focal volumes, a combined array encompassing the design parameters of the concentric-ring array and the sector-vortex array has been proposed. Simulations will show that the sectored annular array is capable of producing larger necrosed tissue volumes than the concentric-ring alone, while maintaining the ability of the concentric-ring array to move the focal volume in the axial direction of the array. These simulations are verified by measurements of the acoustic fields produced by an experimental array in water. In addition, the constructed array produced the necessary power required to coagulate tissue in rabbit thigh in vivo, while the temperature elevation was monitored using MRI.

Method of reduction of the number of driving system channels for phased-array transducers using isolation transformers

Phased-array technology offers an incredible advantage to therapeutic ultrasound due to the ability to electronically steer foci, create multiple foci, or to create an enlarged focal region by using phase cancellation. However, to take advantage of this flexibility, the phased-arrays generally consist of many elements. Each of these elements requires its own radio-frequency generator with independent amplitude and phase control, resulting in a large, complex, and expensive driving system. A method is presented here where in certain cases the number of amplifier channels can be reduced to a fraction of the number of transducer elements, thereby simplifying the driving system and reducing the overall system complexity and cost, by using isolation transformers to produce 180/spl deg/ phase shifts.