Martin P. Mienkina

Multispectral photoacoustic coded excitation imaging using unipolar orthogonal Golay codes.

We present a method to speed up the acquisition of multispectral photoacoustic data sets by using unipolar orthogonal Golay codes as excitation sequences for the irradiation system. Multispectral photoacoustic coded excitation (MS-PACE) allows acquiring photoacoustic data sets for two irradiation wavelengths simultaneously and separating them afterwards, thus improving the SNR or speeding up the measurement. We derive an analytical estimation of the SNR improvement using MS-PACE compared to time equivalent averaging. We demonstrate the feasibility of the method by successfully imaging a phantom composed of two dyes using unipolar orthogonal Golay codes as excitation sequence for two high power laser diodes operating at two different wavelengths. The experimental results show very good agreement with the theoretical predictions.

Experimental evaluation of photoacoustic coded excitation using unipolar golay codes

Q-switched Nd:YAG lasers are commonly used as light sources for photoacoustic imaging. However, laser diodes are attractive as an alternative to Nd:YAG lasers because they are less expensive and more compact. Although laser diodes deliver about three orders of magnitude less light pulse energy than Nd:YAG lasers (tens of microjoules compared with tens of millijoules), their pulse repetition frequency (PRF) is four to five orders of magnitude higher (up to 1 MHz compared with tens of hertz); this enables the use of averaging to improve SNR without compromising the image acquisition rate. In photoacoustic imaging, the PRF is limited by the maximum acoustic time-of-flight. This limit can be overcome by using coded excitation schemes in which the coding eliminates ambiguities between echoes induced by subsequent pulses. To evaluate the benefits of photoacoustic coded excitation (PACE), the performance of unipolar Golay codes is investigated analytically and validated experimentally. PACE imaging of a copper slab using laser diodes at a PRF of 1 MHz and a modified clinical ultrasound scanner is successfully demonstrated. Considering laser safety regulations and taking into account a comparison between a laser diode system and Nd:YAG systems with respect to SNR, we conclude that PACE is feasible for small animal imaging.

Evaluation of Ferucarbotran (Resovist®) as a photoacoustic contrast agent / Evaluation von Ferucarbotran (Resovist®) als photoakustisches Kontrastmittel

Abstract Photoacoustic imaging combines the resolution of ultrasound imaging with the contrast of optical imaging, while maintaining a penetration depth up to a few centimeters. Inorganic gold nanorods can be employed as photoacoustic contrast agents. However, the toxicological properties of such nanoparticles are still under investigation. At the same time, there is an increasing need for clinically established photoacoustic contrast agents. In this paper, therefore, we investigate the photoacoustic properties of Ferucarbotran, which is a clinically established nanoscale contrast agent for magnetic resonance imaging. Gelatin phantoms containing cubes with different gelatin-Ferucarbotran mixture concentrations were prepared and irradiated by a Nd:YAG laser (1064 nm). First, the photoacoustic signals were acquired by a single element ultrasound transducer (7.5 MHz) and evaluated quantitatively. In a second setup, photoacoustic imaging of Ferucarbotran with a modified clinical scanner was demonstrated. The experiments showed that in order to achieve a 6 dB gain of received photoacoustic signal energy, compared to the sensitivity threshold of the used system, a Ferucarbotran concentration of 1.9 μmol Fe/ml is needed. The photoacoustic imaging was successful and showed a contrast-to-background ratio of 15.7 dB for a concentration of 11.63 μmol Fe/ml. However, for imaging in tissue the signal-to-noise ratio has to be increased. Zusammenfassung Die Photoakustik kombiniert die Auflösung der Ultraschallbildgebung mit dem Kontrast der optischen Bildgebung, wobei eine Eindringtiefe von bis zu einigen Zentimetern erhalten bleibt. Insbesondere können anorganische Gold-Nano-Rods auf Grund ihrer besonders starken optischen Absorption vorteilhaft als photoakustische Kontrastmittel eingesetzt werden. Die toxikologischen Eigenschaften dieser Partikel sind jedoch noch nicht eingehend geklärt. Gleichzeitig entsteht aber ein zunehmender Bedarf an klinisch zugelassenen photoakustischen Kontrastmitteln. Daher sollen ihm Rahmen dieser Arbeit die photoakustischen Eigenschaften von Ferucarbotran, einem klinisch eingesetzten Nanopartikel-Kontrastmittel für die Magnetresonanztomographie, evaluiert werden. Es wurden Gelatinephantome hergestellt, die einen Kubus mit verschiedenen Ferucarbotrankonzentrationen erhielten und mittels eines Nd:YAG-Lasers (1064 nm) bestrahlt wurden. Zunächst wurden photoakustische Signale durch einen Einzelelementultraschallwandler (7,5 MHz) aufgenommen und quantitativ ausgewertet. Anschließend wurde die photoakustische Bildgebung von Ferucarbotran mit einem modifizierten klinischen Ultraschallgerät demonstriert. Die quantitativen Experimente zeigen, dass, um einen Gewinn der Signalenergie aus dem Ferucarbotranblock von 6 dB gegenüber der Sensitivitätsgrenze des eingesetzten Systems zu erhalten, eine Konzentration von 1,9 μmol Fe/ml notwendig ist. Weiterhin wurden die Phantome erfolgreich photoakustisch abgebildet und ein Kontrast der Zielregion zum Hintergrund von 15,7 dB für eine Konzentration von 11,63 μmol Fe/ml erreicht. Für die Bildgebung im Gewebe muss das Signal-Rausch-Verhältnis des Systems jedoch noch weiter verbessert werden.

Simulation study of photoacoustic coded excitation using Golay Codes

Photoacoustics (PA) is a new imaging modality based on the generation of ultrasound due to laser irradiation. Instead of commonly used Q-switched Nd:YAG lasers the application of pulsed laser diodes as light sources for photoacoustic imaging is proposed. The high pulse repetition frequency of laser diodes enables the usage of coded excitation schemes for SNR improvement. Therefore, the performance of photoacoustic coded excitation (PACE) using Golay Codes is evaluated by a simulation study based on experimental data. Golay Codes consist of 2 bipolar, complementary code sequences. For PACE the codes must be split into 4 unipolar sequences and processed further. First, experiments are conducted to provide the PA impulse response of the system under investigation by irradiating a spherical absorber by a pulsed laser diode and acquiring the acoustic response. Then, based on these measurements, the acoustic response to a coded laser excitation is simulated by the superposition of PA impulse responses according to the coding scheme. For each code length the acquisition duration is computed, the equivalent number of possible averages is determined, and the coding gain compared to averaging is calculated. The SNR gain of the Golay Codes with respect to a single impulse response ranges from 5.87 dB for a 4 bit sequence to 26.97 dB for a 512 bit sequence. No range sidelobes are detectable and the acoustic response of the optical absorber is equivalent to the signal obtained from averaging. Assuming an acquisition depth of 9 cm, the comparison of the coding schemes with equivalent averaging procedures shows that for coding PRFs lower than 60 kHz Golay coding exhibits a lower SNR than averaging for sequences up to 512 bit. For a PRF of 250 kHz and a coding length of 512 bit the coding gain rises to 6.20 dB. Additionally, the coding gain is theoretically derived and is in good agreement with the simulations. Since state of the art high power laser diode drivers exhibit PRF up to 1 MHz, using coded excitation for photoacoustic imaging can improve the SNR in future set-ups.

Multispectral photoacoustic coded excitation using pseudorandom codes

Photoacoustic imaging (PAI) combines high ultrasound resolution with optical contrast. Laser-generated ultrasound is potentially beneficial for cancer detection, blood oxygenation imaging, and molecular imaging. PAI is generally performed using solid state Nd:YAG lasers in combination with optical parametric oscillators. An alternative approach uses laser diodes with higher pulse repetition rates but lower power. Thus, improvement in signal-to-noise ratio (SNR) is a key step towards applying laser diodes in PAI. To receive equivalent image quality using laser diodes as with Nd:YAG lasers, the lower power must be compensated by averaging, which can be enhanced through coded excitation. In principle, perfect binary sequences such as orthogonal Golay codes can be used for this purpose when acquiring data at multiple wavelengths. On the other hand it was shown for a single wavelength that sidelobes can remain invisible even if imperfect sequences are used. Moreover, SNR can be further improved by using an imperfect sequence compared to Golay codes. Here, we show that pseudorandom sequences are a good choice for multispectral photoacoustic coded excitation (MSPACE). Pseudorandom sequences based upon maximal length shift register sequences (m-sequences) are introduced and analyzed for the purpose of use in MSPACE. Their gain in SNR exceeds that of orthogonal Golay codes for finite code lengths. Artefacts are introduced, but may remain invisible depending on SNR and code length.

Evaluation of Simplex Codes for Photoacoustic Coded Excitation

In biomedical photoacoustics (PA) ultrasound is generated by laser irradiation of tissue. Usually, PA signals are generated by Nd:YAG lasers, but it would be attractive to use less expensive laser diodes instead. Although laser diodes exhibit low pulse energy, the pulse repetition frequency (PRF) of pulsed laser diodes is much higher than the PRF of Nd:YAG lasers thus averaging can be used to increase the SNR. The applied PRF for averaging is limited by the acoustical time-offlight. To further increase the SNR we propose to use coded excitation for pulsed PA imaging. In optical time-domain reflectometry, which is similar to PA imaging, Simplex codes are successfully used for coding. Therefore, we evaluate their performance for photoacoustic coded excitation by a simulation study based on experimental data.

Hybrid 3D Sono/PET in a mouse.

Dear Sir, Hybrid imaging technologies such as PET/CT enable the assessment of complementary aspects of diseases, namely morphology, function and molecular pathways, in a onestop-shop fashion. So far, a 3D hybrid approach combining ultrasound and PET has never been demonstrated. Such an approach would be of particular interest, since these modalities show excellent complementary properties. PET is well known for its ability to assess molecular pathways noninvasively in vivo in both preclinical and clinical studies with very high sensitivity. However, PET cannot assess morphology and is therefore ideally combined with morphological imaging devices in hybrid systems. Ultrasound offers excellent spatial and temporal resolution, and thus provides information on both organ morphology and function. Most importantly, ultrasound is quick, cost-effective, uses no additional radiation and is widely available in health care institutions. In addition to the complementary diagnostic information that it provides, ultrasound can improve the accuracy of PET, e.g. by providing a means for partial volume correction. We therefore tested the feasibility of ultrasound and PET hybrid imaging (Sono/PET) in a small animal scenario. PET and ultrasound scans were performed serially in a wild-type mouse anaesthetised with isoflurane. For constant and rigid positioning during both scans, the mouse was placed and fixated in a dedicated plastic mould that was compatible with both ultrasound and PET imaging. Ultrasound images were acquired using a 5-12 MHz linear array broadband transducer on a Philips HDI5000 device in combination with a PC-controlled positioning system, which allows defined positioning of the transducer in 3D space. A total of 51 transaxial ultrasound images were acquired while moving the transducer step-by-step along the abdomen and thorax of the mouse with a pitch of 0.5 mm. PET scans were initiated on a high-resolution small animal PET device (quadHIDAC, Oxford Positron Ltd., UK) 20 min after intravenous injection of 12 MBq FFDG for 20 min, continuously monitoring vital signs (temperature, ECG, respiratory rate). Images were reconstructed into a 150×150×300 matrix with an isotropic voxel size of 0.4 mm using an iterative 3D OSEM reconstruction algorithm. The stack of DICOM ultrasound images were converted into a 3D volume. Ultrasound and PET images were fused by rigid transformation using the open source software Amide [1]. As shown in Fig. 1, 3D Sono/PET hybrid imaging is feasible in a mouse. Further improvement of this new hybrid imaging technique can be achieved by respiratory and cardiac triggering, providing both improved spatial coregistration between the modalities and optimised image quality. Sono/PET hybrid imaging should therefore become Eur J Nucl Med Mol Imaging (2007) 34:1706–1707 DOI 10.1007/s00259-007-0501-7

Estimation of time of flight for ultrasonic reflex-transmission tomography with active contour models

The histological significance of acoustic speed, together with acoustic attenuation, in differentiating breast carcinomas from healthy tissue is well founded. A reflex-transmission system comprising a metallic reflector and a linear array of a standard ultrasound system was developed, the two fixtures being mounted in such a way that they can describe a whole revolution about the object to be imaged thus enabling a tomographic reconstruction, whose accuracy depends largely on that of the time of flight data estimated on the basis of the first arrival signals reflected from the metallic plate. The estimation is an extremely intricate task, its multiple reflections, attenuation, refraction and speckle restrain the utility of conventional techniques like threshold detection and cross correlation. The problem was eluded by using active contour models, as they allow prior information to be formulated in the form of a cost functional, whose minimum will correspond to the optimum estimate for the time of flight.

Photoacoustic coded excitation using periodically perfect sequences

Photoacoustic imaging is based on the generation of ultrasound using laser irradiation. Nd:YAG laser systems are commonly employed for this purpose, but cheap and handy pulsed laser diodes can be an attractive alternative. They emit significantly lower pulse energies, but fast averaging is feasible due to high repetition rates. Averaging is limited by the time-of-flight of the acoustic signal, but coded excitation can be used to overcome this limit. Here, we examine the performance of difference set based sequences with perfect correlation properties (periodically perfect sequences, PPS). PPS can be used for continuous, artifact free acquisition, the acquisition scheme is simpler than for all previously reported coding strategies. The coding gain reached for periodic imaging is higher than for Golay codes and Legendre sequences.

Compact semiconductor laser sources for photoacoustic imaging

We present a multi-wavelength semiconductor laser source for photoacoustic imaging. We discuss the abilities of the system and its limitations. In detail we analyze how this laser diode system might be used to increase the spectral contrast of ultrasonic systems. In a first test set-up we prove in principle the spectral sensitivity of this device.