Proc. SPIE. 7564, Photons Plus Ultrasound: Imaging and Sensing 2010
KEYWORDS: Real time imaging, Imaging systems, Image resolution, Photoacoustic tomography, Hemodynamics, Acquisition tracking and pointing, Photoacoustic spectroscopy, In vivo imaging, Absorption, Brain
For the first time, the hemodynamics within the entire cerebral cortex of a mouse were studied by using photoacoustic
tomography (PAT) non-invasively and in real time. The PAT system, based on a 512-element full-ring
array with cylindrical focusing, received the PA signal primarily from a slice of about 2 mm thickness. This
system can provide not only high resolution brain vasculature images but also hemodynamic functional images.
We recorded the wash-in process of a photoacoustic contrast agent in a mouse brain in real time. Our results
demonstrated that PAT is a powerful imaging modality to study real-time small animal neurofunctional activities
that cause changes in hemodynamics.
For the first time, the hemodynamics within the entire cerebral cortex of a mouse were studied by using photoacoustic tomography (PAT) in real time. The PAT system, based on a 512-element full-ring ultrasound array, received photoacoustic signals primarily from a slice of 2-mm thickness. This system can provide high-resolution brain vasculature images. We also monitored the fast wash-in process of a photoacoustic contrast agent in the mouse brain. Our results demonstrated that PAT is a powerful imaging modality that can be potentially used to study small animal neurofunctional activities.
We present the application of a curved array photoacoustic tomographic imaging system that can provide rapid, high-resolution photoacoustic imaging of small animal brains. The system is optimized to produce a B-mode, 90-deg field-of-view image at sub-200-µm resolution at a frame rate of ~1 frame/second when a 10-Hz pulse repetition rate laser is employed. By rotating samples, a complete 360-deg scan can be achieved within 15 s. In previous work, two-dimensional (2-D) ex vivo mouse brain cortex imaging has been reported. We report three-dimensional (3-D) small animal brain imaging obtained with the curved array system. The results are presented as a series of 2-D cross-sectional images. Besides structural imaging, the blood oxygen saturation of the animal brain cortex is also measured in vivo. In addition, the system can measure the time-resolved relative changes in blood oxygen saturation level in the small animal brain cortex. Last, ultrasonic gel coupling, instead of the previously adopted water coupling, is conveniently used in near-real-time 2-D imaging.
We present results of investigations of the application of a priori information and sparse
or limited-view algorithms to simultaneously improve imaging quality and timeresolution
in photoacoustic tomography. Modified versions of existing MRI/CT
algorithms such as constrained backprojection and keyhole imaging are evaluated as well
as a new Wiener estimation methods for extrapolation of missing data from reference
data sets. Simulations indicate the effectiveness of the approaches for accurate tracking
of dynamic photoacoustic events for data sets with limited views (< 90 degrees) or
tomographic views with up to 1/64 of the full data set. We present experimental data of
contrast uptake and washout using a 512-element curved transducer with 1:8 electronic
multiplexing that demonstrates high-resolution tomographic imaging with a temporal
resolution of better than 150 milliseconds using these methods.
A 512-element photoacoustic tomography system for small animal imaging using a ring
ultrasound array has been developed. The system features a 5 MHz piezocomposite
transducer array formed into a complete circular aperture. Custom receiver electronics
consisting of dedicated preamplifiers, 8:1 multiplexed post-amplifiers, and a 64-channel
data acquisition module provide full tomographic imaging at up to 8 frames/second. We
present details of the system design along with characterization results of the resolution,
imaging volume, and sensitivity. Small animal imaging performance is demonstrated
through images of mice brain vasculature at different depths and real-time spectroscopic
scans. This system enables real-time tomographic imaging for functional photoacoustic
studies for the first time.
We present the application of an optimized curved array photoacoustic tomographic imaging system, which can provide
rapid, high-resolution photoacoustic imaging of small animal brains. The system can produce a B-mode, 90-degree
field-of-view image at sub-200 μm resolution at a frame rate of ~1 frame/second when a 10-Hz pulse repetition rate
laser is employed. By rotating samples, a complete 360-degree scan can be achieved within 15 seconds. In previous
work, two-dimensional ex vivo mouse brain cortex imaging has been reported. In the current work, we report three-dimensional
small animal brain imaging obtained with the curved array system. The results are presented as a series of
two-dimensional cross-sectional images. Besides structural imaging, the blood oxygen saturation of the animal brain
cortex is also measured in vivo. In addition, the system can measure the time-resolved relative changes in blood oxygen
saturation level in the small animal brain cortex. Finally, ultrasonic gel coupling, instead of the previously adopted
water coupling, is conveniently used in near-real-time 2D imaging.
We present systematic characterization of a photoacoustic imaging system optimized for rapid, high-resolution tomographic imaging of small animals. The system is based on a 128-element ultrasonic transducer array with a 5-MHz center frequency and 80% bandwidth shaped to a quarter circle of 25 mm radius. A 16-channel data-acquisition module and dedicated channel detection electronics enable capture of a 90-deg field-of-view image in less than 1 s and a complete 360-deg scan using sample rotation within 15 s. Measurements on cylindrical phantom targets demonstrate a resolution of better than 200 µm and high-sensitivity detection of 580-µm blood tubing to depths greater than 3 cm in a turbid medium with reduced scattering coefficient µ=7.8 cm−1. The system is used to systematically investigate the effects of target size, orientation, and geometry on tomographic imaging. As a demonstration of these effects and the system imaging capabilities, we present tomographic photoacoustic images of the brain vasculature of an ex vivo mouse with varying measurement aperture. For the first time, according to our knowledge, resolution of sub-200-µm vessels with an overlying turbid medium of greater than 2 cm depth is demonstrated using only intrinsic biological contrast.
We report experimental imaging results with mice using an array-based photoacoustic tomography system designed for small animal imaging. The system features a 128-element curved transducer array with stage rotation to enable complete two-dimensional tomographic
imaging in less than 15 seconds. High fidelity imaging of ex vivo mouse brain vasculature was achieved with resolution of vessels less than 200 microns in diameter in the cortex as well as the
cerebellum. Images obtained using varying measurement surface angular spans clearly illustrate the impact on feature definition with orientation. The high sensitivity of the system was
demonstrated by images of the brain vasculature with an overlying turbid medium (μa=0.03 cm<sup>-1</sup> and μs'~7 cm<sup>-1</sup> at 780 nm) of over 2 cm depth. In phantom experiments, high-quality images of blood tubing in a turbid medium were achieved at depths greater than 3 cm for incident fluences of less than 15 mJ/cm<sup>2</sup>. These results illustrate the suitability for near real-time small animal imaging of deep tissue with high definition.
We report the first experimental investigations of photoacoustic guidance of diffusive optical
tomography for detection and characterization of optical contrast targets. The hybrid system
combined an 8-source, 10-detector reflection mode frequency domain DOT imager with either
orthogonal and reflection-geometry photoacoustic systems. The PAT subsystems imaged two-dimensional
cross-sections to define centers and radii of regions of interest for a dual-zone mesh
DOT imaging algorithm. Phantom absorbers, 1 cm in diameter, of high and low contrast, were
spaced 1.5 to 2.5 cm apart at depths ranging from 1 to 2 cm in a turbid medium. Without PAT
guidance, the absorber DOT images in many cases were merged and indistinguishable. With
PAT guidance, the two targets were well resolved and the reconstructed absorption coefficients
improved to 86-130% of the true values. In addition, using both pulse-echo and photoacoustic
image detection, the photoacoustic guidance correctly distinguished mechanical from optical
contrast providing more specific target information and reconstruction accuracy.
In principle, absorbed energy profiles can be exactly reconstructed from photoacoustic measurements on a
closed surface. Clinical applications, however, involve compromises due to transducer focus, frequency
characteristics, and incomplete measurement apertures. These tradeoffs introduce artifacts and errors in
reconstructed absorption distributions that affect quantitative interpretations as well as qualitative contrast
between features. The quantitative effects of target geometry, limited measurement surfaces, and bandpass
transducer frequency response have been investigated using a ring transducer system designed for small
animal imaging. The directionality of photoacoustic radiation is shown to increase with target aspect ratio,
producing proportionate overestimates of absorption values for two-dimension apertures less than
approximately 150 degrees. For all target geometries and orientations, mean absorption values approach
the full view values for hemicircular measurement surfaces although the true spatial uniformity is
recovered only with the complete surface. The bandpass transducer frequency spectrum produces a peaked
amplitude response biased toward spatial features ranging from 1 to 8 times the system resolution. We
discuss the implications of these results for design of clinical systems.
We have developed and tested a photoacoustic imaging system based on a 128 element curved-phased
ultrasonic array, which spans a quarter of a complete circle with a radius of curvature equal to 25mm. The center
frequency of the array is 5 MHz with 60% bandwidth. The physical dimensions of the elements are 10x0.3mm (elevation
x azimuth) with an elevation focus of 19mm. Earlier we reported acoustic measurements of the axial and lateral
resolutions of the system that were limited by the impulse response of the narrowband source used in the test. In this
paper we discuss photoacoustic characterization of the system including resolution and sensitivity. The array forms the
building block for a 512-element ring designed for complete tomographic imaging of small animals. Imaging results of
phantoms will be compared with simulations.
Real-time photoacoustic imaging requires ultrasonic array receivers and parallel data acquisition systems for the simultaneous detection of weak photoacoustic signals. In this paper, we introduce a newly completed ultrasonic receiving array system and report preliminary results of our measured point spread function. The system employs a curved ultrasonic phased array consisting of 128-elements, which span a quarter of a complete circle. The center frequency of the array is 5 MHz and the bandwidth is greater than 60%. In order to maximize the signal-to-noise ratio for photoacoustic signal detection, we utilized special designs for the analog front-end electronics. First, the 128 transducer-element signals were routed out using a 50-Ohm impedance matching PCB board to sustain signal integrity. We also utilize 128 low-noise pre-amplifiers, connected directly to the ultrasonic transducer, to amplify the weak photoacoustic signals before they were multiplexed to a variable-gain multi-stage amplifier chain. All front-end circuits were placed close to the transducer array to minimize signal lose due to cables and therefore improve the signal-to-noise ratio. Sixteen analog-to-digital converters were used to sample signals at a rate of 40 mega-samples per second with a resolution of 10-bits per sample. This allows us to perform a complete electronic scan of all 128 elements using just eight laser pulses.