Multispectral optoacoustic tomography

Computational techniques such as spectral unmixing deconvolute the ultrasound waves emitted by these different absorbers, allowing each emitter to be visualized separately in the target tissue.

[1][6] Like optical microscopy, they use focused light to form images and offers fundamentally the same capabilities (submicrometer resolution, <1mm penetration depth).

This term denotes images formed by combining raw measurements from multiple points around the specimen in a mathematical inversion scheme.

This process is analogous to x-ray computed tomography, except that tomographic mathematical models describe light and sound propagation in tissues.

MSOT detects photoechoes, i.e. ultrasound waves generated by thermo-elastic expansion of a sample (e.g. tissue) after absorption of transient electromagnetic energy.

Background in images can be reduced by exploiting differences in time (baseline subtraction) and in absorption spectra of the various photoabsorbers (spectral unmixing).

MSOT has the potential to provide multi-parametric information involving the three spatial dimensions (x, y, z), time, optical wavelength spectrum and ultrasound frequency range.

[2] This dimensionality has been made possible by key advances in laser source and detector technology, computed tomography and unmixing techniques.

Optimal tomographic imaging is therefore achieved by recording time-resolved pressure waves along a closed surface volumetrically surrounding the target tissue.

[33] Early optoacoustic imaging involved scanning a single ultrasound detector along one or two dimensions, resulting in acquisition times of several seconds, minutes or longer.

(b) Maximal-intensity projections along the axial direction following single-wavelength illumination before (upper) and after injection of two concentrations of contrast agent (10nmol in the middle and 50 nmol lower), indocyanine green.

A key strength of MSOT is its ability to resolve the photoechoes obtained in response to excitation with different wavelengths of illuminating light.

The endogenous photoabsorbers most often imaged are oxy- and deoxy-hemoglobin, key players in oxygen metabolism, myoglobin, lipids, melanin and water.

[34][35] Through spectral unmixing and other techniques, MSOT data can be used to generate separate images based on the contrast provided by different photoabsorbers.

In other words, a single MSOT data collection run provides separate images showing the distribution of oxy- or deoxy-hemoglobin.

Macroscopic MSOT typically uses detectors operating in the frequency range from 0.1 to 10 MHz, allowing imaging depths of approximately 1–5 cm and resolution of 0.1–1 mm.

Such mesoscopy can analyze morphology and biological processes such as inflammation in greater detail than macroscopy, revealing, for example, microvasculature networks in skin and epithelial tissues or the microenvironment within a tumor.

Sensitivity also depends on the ultrasound detector employed, the amount of light energy applied, the voxel size and spectral unmixing method.

[2][48] Organic dyes, such as the fluorochromes indocyanine green and methylene blue, are non-specific, approved for clinical use, and suitable for perfusion imaging.

[51] Light-absorbing nanoparticles offer potential advantages over organic dyes because of their ability to produce stronger photoechoes and their lower photosensitivity.

Gold nanoparticles generate strong optoacoustic signals due to plasmon resonance, and their absorption spectrum can be tuned by modifying their shape.

[9] This transgenic approach is not limited to fluorescent proteins: infecting tissue with a vaccinia virus carrying the tyrosinase gene allows in situ production of melanin, which generates strong optoacoustic signal for MSOT.

[61] Several optoacoustic studies[19][62][63] have aimed to improve on the poor sensitivity of X-ray mammography in dense breast tissue and the low specificity of ultrasound imaging.

MSOT studies of breast cancer typically focus on detecting the increased vascular density and correspondingly high hemoglobin concentration thought to occur in and around tumors.

[48] The hemoglobin distribution in carotid arteries of healthy humans has recently been imaged in real time using a hand-held device similar to diagnostic ultrasound systems currently in the clinic.

The handheld MSOT probe shown here to measure photoechoes from hemoglobin, allows more sensitive detection of small blood vessels than Doppler ultrasound already in the clinic.

[2][48] Miniaturized optoacoustic devices are also expected to offer interesting possibilities for intravascular imaging [72-74], improving our ability to detect atherosclerosis and stent-related biomarkers.

Optoacoustic imaging is likely to be well suited to this application, since it can detect lipids, neovasculature, hemoglobin oxygenation and contrast agents that mark inflammation.

(A) Indocyanine green (ICG) is injected and accumulates inside the sentinel lymph node, which is detected using a hand-held two-dimensional MSOT device.

In both cases, strong melanin signal from the skin can be seen Optoacoustic imaging in general and MSOT in particular may address a number of challenges for surgical procedures by providing real-time visualization below the tissue surface.