The purpose of this work was to demonstrate that a clinical ultrasound transducer array can practically detect thermoacoustic pulses induced by irradiation by very high frequency (VHF) electromagnetic energy. This is an important step because thermoacoustic signal strength is directly proportional to the specific absorption rate (SAR), which is lower in the VHF regime than in microwave or optical regimes. A 96-channel transducer array (P4-1) providing 3 cm coverage was incorporated into a benchtop thermoacoustic imaging system for imaging fresh surgical specimens. Thermoacoustic signal was generated by 700 ns irradiation pulses with 11 kV/m electric field strength and 108 MHz carrier frequency. To improve SNR 1024 pulses were averaged at a 250 Hz repetition rate. Two sets of sinograms were acquired, separated by a 2 cm translation along the tomographic axis and reconstructed over a 6 x 6 x 5 cm3 volume. Contrast and in-plane resolution were measured by imaging a homogeneous cylindrical phantom and an 80- micron wire designed to highlight E-field polarization effects. FWHM of the in-plane point spread function varied from 250 microns to 1.1 mm, depending upon transducer used and phantom orientation relative to the electric field. Several fresh human prostates were imaged immediately after surgery. Rudimentary comparison to histology was performed and volumetric reconstruction of the multi-channel P4-1 data visualizes anatomic features that are rarely seen in ultrasound, CT, or MRI. The single element transducer provided superior image contrast, but with inferior resolution.
Ex vivo imaging of fresh prostate specimens was performed to test the hypothesis that the thermoacoustic (TA) contrast mechanism generated with very high frequency electromagnetic (EM) irradiation is sensitive to prostate cancer. Ex vivo imaging was performed immediately after radical prostatectomy, performed as part of normal care. Irradiation pulsewidth was 700 ns and duty cycle was extremely low. Typical specific absorption rate (SAR) throughout the prostate was 70-90 kW/kg during pulsing, but time-averaged SAR was below 2 W/kg. TA pressure pulses generated by rapid heating due to EM energy deposition were detected using single element transducers. 15g/L glycine powder mixed into DI water served as acoustic couplant, which was chilled to prevent autolysis. Spatial encoding was performed by scanning in tomographic “step-and-shoot” mode, with 3 mm translation between slices and 1.8-degree rotation between tomographic views. Histology slides for 3 cases scanned with 2.25 MHz transducers were marked for comparison to TA reconstructions. These three cases showed little, moderate, and severe involvement in the histology levels surrounding the verumontanum. TA signal strength decreased with percent cancerous involvement. When VHF is used for tissue heating, the TA contrast mechanism is driven by ionic content and we observed suppressed TA signal from diseased prostate tissue in the peripheral zone. For the 45 regions of interest analyzed, a reconstruction value of 0.4 mV provides 100% sensitivity but only 29% specificity.
Thermoacoustic (TA) imaging provides a novel contrast mechanism that may enable visualization of cancerous lesions
which are not robustly detected by current imaging modalities. Prostate cancer (PCa) is the most notorious example.
Imaging entire prostate glands requires 6 cm depth penetration. We therefore excite TA signal using submicrosecond
VHF pulses (100 MHz). We will present reconstructions of fresh prostates imaged in a well-controlled benchtop TA
imaging system. Chilled glycine solution is used as acoustic couplant. The urethra is routinely visualized as signal
dropout; surgical staples formed from 100-micron wide wire bent to 3 mm length generate strong positive signal.
The unique electrical properties of Single-Walled Carbon Nanotubes make them good candidates for thermoacoustic
contrast agents. Theoretical considerations suggest that nanotubes are capable of greatly increasing a material's
absorption of electromagnetic radiation. We describe these properties and discuss our measurements of aqueous
nanotube solutions and nanotube-infused tissue mimicking phantoms. We discuss results and the difficulties currently
associated with making these measurements on nanotubes.
As in MRI, orientation of the patient relative to field lines significantly impacts signal generated. Oblong inclusions generate far less thermoacoustic signal when oriented with long axis perpendicular to E field lines of radiofrequency excitation. We present simulated and low-power measured results below.
Thermoacoustic signal excitation is a function of intrinsic tissue properties and illuminating electric field. We
have designed a water-filled testbed propagating a controlled electric field with respect to pulse shape, power, and
polarization. Illuminating with a known and carefully controlled E field will enable quantitative measurement of
the thermoacoustic contrast mechanism.
Thermoacoustic contrast under RF illumination is a function of electrical conductivity and applied electric field, as well
as mechanical and thermal properties. An array of phantom objects complement our RF testbed which illuminates with a
controlled E field. All rely on electrical conductivity to generate TCT contrast. Inexpensive and easily fabricated
resolution phantoms are highly conductive small inclusions with similar acoustic and thermal properties as background
water bath. Tissue mimicking phantoms developed for microwave imaging have acoustic and dielectric properties
similar to fat and muscle. Thermal properties are measured to completely quantify expected TCT contrast for
Radiofrequency (RF) pulses used to generate thermoacoustic computerized tomography (TCT) signal couple directly
into the pulser-receiver and oscilloscope, swamping true TCT signal. We use a standard RF enclosure housing both RF
amplifier and object being imaged. This is similar to RF shielding of magnetic resonance imaging (MRI) suites and
protects electronics outside from stray RF. Unlike MRI, TCT receivers are ultrasound transducers, which must also be
shielded from RF. A transducer housing that simultaneously shields RF and permits acoustic transmission was
developed specifically for TCT. We compare TCT signals measured with and without RF shielding.
Ideal transducers have perfectly uniform response to incoming pressure waves, regardless of frequency. In practice, piezoelectric transducers are designed with a particular center frequency (CF) and are most sensitive to signals with strong frequency content near CF, bandpass filtering signal - and reconstructed images. We characterized the frequency dependent receive sensitivity of three single-element transducers with CFs ranging from 1 to 3.5 MHz. The resulting sensitivity response curves are applied to ideal
thermo/photo/opto-acoustic (TPOAT) signals generated by ideal spherical absorbers to show the impact of transducer frequency response on measured TPOAT data and reconstructed images.
Filtered backprojection (FBP) reconstruction is the method of choice for diagnostic xray CT, despite
the fact that backprojection is computationally costly. FBP image quality is superior over fast Fourier
reconstruction techniques because interpolation errors are localized and the backprojector applies the
Radon transform, annihilating all measurement errors orthogonal to its range. We discuss computational
complexity, sampling rates, and quadrature techniques for FBP reconstruction of thermo/photo/optoacoustic
Governing equations for ultrasonic propagation in three spatial dimensions with attenuation obeying a
frequency power law are derived. Quadratic attenuation corresponds to a partial differential equation
of degree four whose operator factors into a product of two parabolic operators, and impulse response
is related to the heat kernel. The solutions satisfy primitive causality, but not relativistic causality. For
powers that are not even integers, the waves satisfy integral-differential equations.
A recovery algorithm is given for the 2 X 2 X 2 problem in diffuse tomography. This three dimensional algorithm is computationally more complex and yields relatively more information than its two dimensional counterpart.
The idea of diffuse tomography has been recently introduced as a way of modeling an imaging problem using photons with very low energy. In this paper, we discuss how to use `Pluecker relations' to simplify the general equations. Using Pluecker relations, we are now able to reduce the original system of equations to a pair of equations before we run out of computing power.
The idea of diffuse tomography has been recently introduced as a way of modeling an imaging problem using photons with very low energy. It can be seen as a far reaching extension of the standard tomographic problem where photons are assumed to travel in a straight line. Although any real-life application will require the solution to the three-dimensional problem, we start with the two-dimensional problem.