The discovery of the PA effect also stimulated the development of various approaches for the visualization of the object, which was later called photoacoustic imaging (PAI)
[6]. The first low-resolution and contrast images were focused on obtaining information from various inorganic materials, such as metallic and semiconducting structures
[7]. The limitations of this technique were due to the inherent drawbacks of the systems related to focusing monochromatic optical radiation and positioning the beam on the sample. Subsequent technical progress has played an important role in the development of this technique. The modern PAI technique reconstructs the simultaneous real-time imaging of surface and subsurface structures with high contrast and micron-scale spatial resolution
[8][9][10]. In this section, the researchers introduce technical advances and the applicability of the photoacoustic technique in microscale imaging. In addition, the researchers briefly emphasize the role of nanostructures in improving the visualization of objects using the PAI technique.
PAI methods have several advantages over optical microscopy, namely (i) they provide a deeper imaging depth (up to hundreds of micrometres for PAM and up to several centimetres for PAT), (ii) they do not need optical sectioning to provide a 3D PA image, and (iii) they can analyze the samples that are too absorbing, too scattering, or too transparent for optical microscopy methods.
2.1. Photoacoustic Microscopy: Areas of Application, Advantages, and Drawbacks
At present, PAM is known to be one of the most powerful techniques for the non-destructive testing of surface flaws and subsurface inhomogeneities of the studied objects. This concept is provided by using a non-stationary laser with the ability to move its beam through the sample’s surface and detect the PA response This is achieved, for example, by moving the sample itself using stepper motors or manipulating the beam using acousto-optic deflectors. According to the process of PA signal registration, the most commonly used PAM detection schemes are the gas–microphone technique and a method that uses a piezoelectric transducer.
In the case of the low-frequency modulation of the laser (from Hz to kHz), the PA signal depends on (a) the optical absorption of the sample, (b) the generation and propagation of thermal waves, and (c) elastic waves. Thus, many factors contribute to the characterization of a sample based on the analysis of its PA response.
A combination of PAM with other methods can provide aggregate information about the target sample, complementing different types of imaging
[11][12][13]. As scanning modes, it is possible to use various methods for recording the PA response, methods such as optical images for the transmission or reflection of light, the photoelectric (PE) method, etc. The choice of the scanning mode depends on the possibility of adapting the system to the structural features of the sample, as well as the research objectives.
In particular, PAM, complemented by images in the PE mode of semiconductor materials, can visualize any irregularities that are elastically stressed regions, disturb the propagation of the thermal diffusion wave, and/or affect the optical absorption coefficient that is difficult to see with an optical microscope (for example, subsurface voids, microcracks, the crystallographic defects of substrates, and the delamination of layered materials)
[13]. Furthermore, the use of confocal microscopy coupled with the PA approach offers the possibility of estimating the size of the nanoparticle’s agglomeration
[14].
The use of a dual-mode PAM/optical imaging system has been implemented, for example, for visualizing organic objects, such as the Drosophila fly (see
Figure 2)
[12]. The obtained PA images complement the images based on the reflected light and reveal the features of the microscopic anatomy of the insect that can be accurately seen only with high-precision and expensive devices such as SEM.
Figure 2. PAT images of a Drosophila fly obtained by a bi-modal system in the reflected light and PA mode.
It should be noted that low-frequency PA microscopy, along with its advantages, also has several technical and engineering problems that need to be solved. Among them, motion artefacts (if a living object is used as a target), imaging field limitations, and low scanning speed play an important role. In addition, most modern PAM systems are applicable only for the visualization of small areas (up to several hundred square millimetres), limited both by instrumental factors (slow scanning speed, the capabilities of sensors for detecting the PA signal, etc.) and by the nature of photothermoacoustic processes (fast extinction of the heat-wave in the sample, etc.).
To overcome motion artefacts from living objects under study, they are typically well-fixed and/or in anaesthesia. However, the anaesthetic or uncomfortable fixation effect may affect physiological functions, such as neuron activities and metabolism, contrary to normal conditions. These factors can give a false picture of the vital signs of the object during various kinds of research. The issue of analyzing conscious living organisms using PAM methods at this stage has not been resolved and requires significant engineering solutions (the miniaturization of a cell, its incorporation into an organism, a significant reduction in scanning time, etc.).
2.2. Adaptation of Photoacoustic Tomography for Preclinical and Clinical Applications
Photoacoustic tomography (PAT) characterizes the optical properties of a studied sample from acoustic signals excited by the absorption of modulated electromagnetic radiation. The imaging of optical absorption reveals important information about the internal environment in phantom or biological tissues
[15][16]. In general, the PA response is excited by pulsed laser radiation with a pulse duration of ~10 ns in the visible and near-infrared spectral range, and the acoustic waves, which appear after the absorption of radiation and are caused by the thermoelastic expansion of the media, are detected by various broadband sensors, analyzed in detail in
[17].
In PAT, images are reconstructed by applying appropriate tomographic algorithms to the acquired time-resolved signals for an accurate reconstruction of imaging. To create a high-quality image, several hundred to tens of thousands of detector positions are usually used. To achieve this high number of sensors, several approaches could be applied. Dean-Ben et al.
[18] developed a volumetric multispectral optoacoustic tomography platform for imaging a neural activation deep in scattering brains. Such a platform contains 512 elements, which allow us to obtain single-shot, 3D tomographic images. The reconstruction of an image by applying a 3D algorithm can be performed by a curved array rotating relative to the studied object, emulating a spherical array with thousands of virtual sensor points
[19]. A PAT image can be reconstructed by a series of 2D sectional images generated by a ring-shaped array that encloses the studied object and is then translated along the ring axis
[20][21]. Additionally, one can independently analyze each 2D slice. The latter allows for more precision, considering the time-dependent processes in a slice. For instance, Paltauf et al. presented an array of extended, line-shaped detectors that generate 2D projections of the photoacoustic sources for photoacoustic tomography
[22][23].
The possibility of the noncontact piezoelectric detection of photoacoustic signals in tissue phantoms was analyzed in
[24] in detail. The authors studied human blood flow in a silicon rubber tube mimicking a blood vessel with an inner diameter of 4 mm. The photoacoustic time traces were recorded by the transducers, which were located at a 7.5 mm distance from the phantom interface, at different frequencies. It was shown that the sensitivity of the air-coupled ultrasound transducers is sufficient to detect PA signals generated by an artificial blood vessel. The latter allows for the integration of PA noncontact piezoelectric detection simultaneously with the X-ray mammographic screening procedure.
Gao et al.
[25] applied the piezoelectric photoacoustic method based on a simplified thermoelastic theory to evaluate the thermal diffusivities of some biological tissues. Particularly, the thermal diffusivities of porcine tissues (skin, fat, muscle, heart, liver, and kidney) with different preparation conditions were assessed. They showed that the thermal diffusivities of the fresh tissues are higher in comparison with those of the dried and specially prepared tissues. This may occur because evaporating the tissues increases the discontinuity in the tissues and fixing the proteins and fats of the tissues increases the thermal resistance. Both these effects decreased the thermal diffusivities of the tissues. The obtained results show that the photoacoustic method with piezoelectric detection is effective for evaluating the effective thermal diffusivities of the tissues with micro-inhomogeneities.
Kolkman et al.
[26] demonstrated the possibility of the application of photoacoustic imaging (PAI) in the detection of blood vessels inside the tissue in a non-invasive way. The biological tissue is irradiated by a short light pulse, which is partly absorbed by the particles, such as red blood cells. The resulting acoustic wave propagates through the tissue and can be detected at the surface by a piezoelectric transducer.
The photoacoustic technique with piezoelectric detection (by using a fast PZT-ceramic acoustic transducer) also allowed for the in vitro and in vivo study of individual cells and nanoparticles in
[27]. The latter offers the possibility for the diagnosis of breast cancer labelled with gold nanoparticles. Together with further hyperthermia treatment, this allows us to develop an efficient theranostic approach for breast cancer.
Among other areas in therapy, the analysis of the glucose level in blood or interstitial fluid is crucial for diabetic diagnosis. Kottmann et al. studied glucose concentration in the human epidermis by the PA method for the first time
[28][29]. They used an external-cavity quantum-cascade laser and a gas–microphone method for the detection of acoustic signals described above. The detection limits of glucose concentration were obtained in the physiological range of 30–500 mg/dL. Nevertheless, for monitoring the glucose levels non-invasively, some improvements are necessary. In
[30], a tunable pulsed laser for a photoacoustic method with piezoelectric detection was used for measurements of glucose concentration. However, a significant disadvantage of such studies is the need to use expensive microphones and quantum cascade lasers.
Bayrakli et al.
[31] developed a PA sensor with an inexpensive piezoelectric film transducer as an acoustic detection module and an amplitude-stabilized external cavity diode laser source. A wavelength equal to 1050 nm of the source was chosen for glucose concentration analysis. This sensor offers the possibility of achieving a detection limit of 900 mg/dL. In
[32], the theory of liquid photoacoustic resonance was developed. The relation between signal characteristics and the concentration of the glucose solution was obtained. The experimental results show that a liquid’s photoacoustic resonance can enhance the signal and the resolution thus achieved is 20 mg/dL. Therefore, the proposed method overcomes the issue of low sensitivity and inaccurate detection in nonresonant cases.