Cutaneous microvasculature not only provides structural support but also plays a key role in defense and immune responses. Diabetes mellitus, however, can damage the skin microvascular system, thereby compromising the skin’s barrier protective function. At present, a variety of techniques have been employed for the assessment of skin microvessels, including laser Doppler flowmetry (13), nailfold capillaroscopy (14, 15), dermatoscopy, Doppler ultrasound, laser speckle contrast imaging (LSCI), optical coherence tomography, and spatial frequency domain imaging (16). Among these, LSCI assesses blood flow velocity by analyzing dynamic fluctuations of laser speckle patterns. While it offers advantages such as full-field imaging and high temporal-spatial resolution, it remains susceptible to tissue motion and illumination distance, with additional limitations including limited penetration depth. Consequently, it is primarily suitable for evaluating cutaneous microcirculatory dysfunction (17). These techniques primarily focus on observing the structure and morphology of microvessels but lack imaging assessments related to the metabolic aspects of diabetic microvascular lesions, failing to reveal the associations between microvessels and metabolism, blood oxygenation, inflammation, and other factors. Nonetheless, some current research has begun to explore this aspect. PAI has garnered attention due to its non-invasive nature, high selectivity, and deep penetration, offering new possibilities for the exploration of skin microvessels (Figure 2).

PAI can correlate with vascular parameters and metabolic indicators, thereby assessing microvascular lesions and inflammatory states in the skin. The study by Yao et al. (18) utilized photoacoustic microscopy (PAM), which is capable of not only imaging the skin of healthy mice but also calculating vascular morphometric parameters such as vessel diameter distribution, microvascular density, vascular tortuosity, and fractal dimension, demonstrating the applicability of PAI in skin microvessel monitoring. Favazza et al. (9) used PAM to obtain vascular images of the skin from the forearms and palms of healthy volunteers, confirming the potential of PAI in characterizing the structure and function of human skin microvascular networks. Krumholz et al. (19) further focused on type 1 diabetic mice and used PAM to monitor the effects of varying blood glucose levels on vascular morphology and function. Their research linked changes in blood glucose metabolism to skin vascular imaging, confirming that PAI can effectively assess microvascular damage induced by disorders of glucose metabolism.

In addition to photoacoustic microscopy, photoacoustic mesoscopy can also be applied to skin microvascular imaging. The study by Karlas et al. (20) demonstrated that photoacoustic mesoscopy can reflect the characteristic changes in the skin of individuals with diabetes. By using a raster-scanning photoacoustic mesoscope to extract images from the leg skin of healthy individuals and diabetic patients, the researchers found that the skin characteristics changed at different stages of diabetes, primarily in the dermis and the epidermis/subpapillary vascular plexus layers. Photoacoustic mesoscopy offers high-contrast imaging of vascular structures and high-resolution three-dimensional visualization of the epidermis and dermis, as well as directly associating glucose metabolism with skin imaging. Moreover, skin microvascular lesions often occur in the early stages of diabetes, possibly preceding microvascular complications in other organs and also preceding macrovascular complications and marked hyperglycemia (blood glucose levels ≥200 mg dL^-1). Therefore, photoacoustic mesoscopy is of great importance for the prevention and control of early microvascular lesions in diabetes.

Diabetic retinopathy (DR) is the most common retinal vascular disease and one of the leading causes of blindness in individuals over the age of 40. The fundamental pathological process of DR involves damage to microvascular cells, leading to microvascular dilation, microaneurysms, leakage, and subsequently to microvascular occlusion and the formation of non-perfused areas. This results in retinal ischemia and hypoxia, ultimately progressing to proliferative lesions. Therefore, it is particularly important to adopt appropriate diagnostic methods and to manage DR early in its course.

Currently, clinical ophthalmic imaging predominantly utilizes optical imaging techniques. However, these technologies often face challenges such as invasiveness or depth limitations, preventing effective non-invasive visualization of deep structures like the choroid. These optical imaging modalities include optical coherence tomography, optical coherence tomography angiography, scanning laser ophthalmoscopy, adaptive optics, fundus autofluorescence, and molecular imaging. Although some of these imaging technologies have evolved to achieve high spatial resolution, they still suffer from limited penetration. Moreover, most imaging techniques primarily provide anatomical information, with relatively little physiological information (21).

PAI is capable of providing anatomical and functional features of the eye, particularly excelling in visualizing blood vessels. The photoacoustic funduscope is a newly developed instrument based on PAI for examining the ocular microcirculation. Compared to traditional light-based imaging systems, it offers greater tissue penetration depth. This technology cleverly utilizes hemoglobin and melanin, the two dominant natural chromophores in retinal optical absorption, allowing for the acquisition of structural and functional information without the need for exogenous contrast agents (5). Moreover, PAM can achieve an axial resolution as high as 2.5 μm and a lateral resolution as high as 0.4-0.7 μm, enabling the diagnosis and monitoring of minute intracular structures and cells (22). However, the resolution decreases with increasing imaging depth.

Multimodal imaging has demonstrated great potential in the visualization of diabetic retinopathy. By integrating the advantages of different imaging modalities, PAI can more accurately assess the structure and function of the retina. This technology aids in a deeper understanding of the pathogenesis and pathological processes of the disease, providing a scientific basis for future therapeutic strategies. Nguyen et al. (23) combined photoacoustic microscopy (PAM) with spectral-domain optical coherence tomography (SD-OCT) to observe rabbit retinal neovascularization (RNV). This system clearly characterized the location and morphology of individual RNV, capable of displaying high-resolution hemoglobin optical absorption as well as retinal and choroidal vascular imaging. The research by Du (24) further expanded this field. They established a dual-modal imaging system combining PAM and SD-OCT for in vivo fundus imaging of rats with choroidal neovascularization, and incorporated nanoprobes as exogenous contrast agents to enhance imaging, thereby more comprehensively obtaining information on choroidal vasculature. These studies have validated the feasibility of photoacoustic and multimodal imaging systems for in vivo ocular imaging, expanded the diagnostic approaches for diabetic retinopathy, and provided new insights and tools for the diagnosis and treatment of related diseases.

Diabetes can lead to impairments in the structure, function, and metabolic capacity of skeletal muscle, clinically manifesting as reduced muscle mass, muscle weakness, and decreased exercise capacity, a condition referred to as diabetic myopathy (25). The microvascular lesions in skeletal muscle are a significant factor contributing to changes in muscle mass and quality in patients with diabetes. Impaired function of the skeletal muscle microvasculature results in insufficient nutritional support for skeletal muscle cells, thereby affecting the normal metabolism and function of muscle cells. However, there are currently limited methods for evaluating skeletal muscle microvascular perfusion, each with its own limitations. For instance, contrast-enhanced ultrasound has been used in studies to assess skeletal muscle microcirculation in healthy individuals and patients with diabetes (26, 27), but low-power contrast-enhanced ultrasound imaging produces a relatively low signal-to-noise ratio, and while higher power can yield a stronger signal, it may lead to microbubble destruction (28); MR perfusion imaging (29) is time-consuming and costly, precluding its use as a routine examination; 1-methylxanthine, a substrate for xanthine oxidase, can be measured by liquid chromatography to quantify microvascular perfusion, but its application is limited due to the low activity of xanthine oxidase in the human skeletal muscle microcirculation (30, 31).

In contrast, PAI offers advantages in skeletal muscle microvascular imaging with high contrast, high penetration, allowing for noninvasive functional, metabolic, and histological imaging applications. Loai et al. (32) utilized a high-frequency ultrasound photoacoustic imaging system to image a type 2 diabetic rat model and identified a significant reduction in leg skeletal muscle perfusion, aiding in the understanding of the manifestations of diabetic skeletal muscle microvascular dysfunction. Furthermore, the research group (33) discovered that early reductions in leg skeletal muscle perfusion in diabetic cardiomyopathy were associated with gender-related microvascular abnormalities, with female diabetic rats exhibiting a significantly delayed decrease in skeletal muscle tissue oxygen saturation compared to male diabetic rats. This indicates that photoacoustic imaging technology can be used to assess microvascular perfusion in diabetic myopathy and to reveal gender differences. However, the researchers noted that the sensitivity of using PAI to indirectly measure perfusion may not be very high, and significant changes in oxygen metabolism, possibly indicative of more severe microvascular complications, may be required to meet the sensitivity threshold for photoacoustic imaging, which could result in the missed detection of perfusion impairment in the legs of female diabetic rats at an early stage.

Despite its limitations, the potential of photoacoustic imaging technology in assessing microvascular perfusion in diabetic myopathy remains noteworthy, and it is anticipated that future technological innovations will enhance its sensitivity and clinical applicability.

Existing methods for assessing local peripheral vascular function have certain limitations: Ultrasound imaging is non-invasive and convenient, but lacks sensitivity for small vessels; computed tomography angiography requires the use of nephrotoxic contrast agents, which is problematic as diabetes patients often have kidney damage; intravascular ultrasound and optical coherence tomography are invasive and their application is limited to large arteries; magnetic resonance angiography can provide details of the circulation in the upper and lower extremities, but does not offer direct perfusion information and still requires the use of nephrotoxic contrast agents; diffuse optical tomography can provide dynamic changes of deep vessels, but has relatively low resolution; laser Doppler imaging is simple to use and can monitor perfusion of foot vessels, but is limited to superficial tissues. In the realm of emerging imaging technologies, diffuse speckle contrast analysis (DSCA), an advanced development based on speckle techniques in recent years, has overcome the depth limitations of traditional LSCI. It enables blood flow measurements in deep tissues and can extract tissue perfusion indices for quantification, showing potential in the assessment of diabetic foot perfusion. Nevertheless, DSCA remains sensitive to motion artifacts and requires further optimization of probe mechanical stability (34). In parallel, NIR-II imaging utilizes the second near-infrared window (approximately 900–1880 nm) to achieve high signal-to-background ratio and high-resolution vascular visualization, with high sensitivity to perfusion changes in early-stage pathologies. Studies have demonstrated (35) that NIR-II imaging can acutely detect microcirculatory perfusion impairments caused by arterial plaques even during the preclinical stages of peripheral arterial disease. In research by Wang et al. (36), NIR-II imaging clearly revealed vascular truncation planes, collateral circulation development, and quantitatively evaluated perfusion recovery at different time points. While NIR-II imaging provides a novel perspective for early warning and intervention of diabetic vascular complications, its dependence on contrast agents currently limits broad clinical application.

PAI, as a novel non-destructive imaging technology, possesses the capability to monitor and assess peripheral vasculature. Joongho et al. (37) utilized photoacoustic microscopy to non-invasively monitor morphological changes in the rat limb vessels during rapid increases in blood glucose levels and identified their correlation with acute hyperglycemia. Jinge Yang et al. (38) extended the research to human studies, employing photoacoustic tomography to monitor vascular dynamics during lower limb vascular stimulation, demonstrating hemodynamic responses observed within the vessels and assessing the magnitude and time parameters of reactive hyperemia following arterial occlusion, confirming the decline in vascular function with aging. Wu Man’s study (39) further included observations in diabetic patients. In this study, PAT was used to image the finger vessels of two groups of volunteers (healthy and diabetic) and to monitor hemodynamic changes before and after vascular occlusion, uncovering imaging evidence of diminished vascular function in diabetic patients. These studies indicate that PAI is feasible in assessing limb vascular function and provides a new tool for the prevention and control of diabetic limb vascular complications.

Because of the characteristics and advantages of photoacoustic imaging in peripheral vascular evaluation, researchers have used it specifically for the evaluation of diabetic foot, and most of them are photoacoustic and multimodal imaging. In the study by Choi et al. (40), a photoacoustic/ultrasound imaging system integrated with a foot scanner was able to clearly visualize the structures of blood vessels, bones, and skin without the use of contrast agents, and reliably quantify foot vascular parameters non-invasively, providing functional information such as total hemoglobin concentration, hemoglobin oxygen saturation, vascular density, and vessel depth (Figure 3). Jinge Yang et al.’s research (11) included observations of diabetic foot, where they used a photoacoustic tomography system based on a concave ultrasound sensor array to image the dorsum of the foot in both diabetic and healthy subjects. They observed lower levels of oxygen saturation in diabetic patients compared to healthy subjects, and their peripheral blood exhibited a distinct hemodynamic response.

The aforementioned studies indicate that photoacoustic imaging technology has certain applicability in detecting vascular dysfunction in diabetic foot and assessing therapeutic efficacy. However, these studies also highlight some limitations, such as the limited light penetration due to the presence of bone structures, and the potential for non-uniform sensitivity in the PAT obtained from the laser irradiation method on the side of the sensor array. Future research could visualize the plantar aspect by adding another imaging probe at the bottom. Additionally, the issue of non-uniform sensitivity could be addressed by using PA coupling pads with light diffusion or quantitative PAT methods, further optimizing the application of photoacoustic imaging technology in the diagnosis of diabetic foot.

Chronic wounds, including diabetic ulcers and pressure ulcers, represent a significant challenge in the healthcare field. Pressure ulcers can be categorized according to the depth and severity of the injury into stages of erythema, inflammatory infiltration, shallow ulceration, and necrotic ulceration. The necrosis of ulcers can be attributed to reduced blood flow under sustained pressure, leading to vascular dysfunction of the tissue. Both pressure ulcers and diabetic foot ulcers result in tissue vascular dysfunction, but this dysregulation is often difficult to detect in the early stages.

Currently, imaging modalities for chronic wound examination include ultrasound, computed tomography, magnetic resonance imaging, optical coherence tomography, and others. However, some of these imaging modalities are challenging in assessing early-stage damage in chronic wounds. The advantage of PAI lies in its capability to detect early lesions, non-invasively observing photoacoustic intensity changes associated with early tissue damage, thus preventing further ulcer progression and epithelial barrier disruption. In a murine pressure ulcer model, Hariri et al. (41) evaluated alterations in photoacoustic signals and injury depth. The results indicated that photoacoustic imaging can differentiate between different stages of pressure ulcers and monitor the healing process. Nevertheless, the study also highlighted limitations of photoacoustic imaging, such as difficulties in imaging through bone and the possibility that photoacoustic signals from deep tissues may not fully represent the severity of the ulcer.

Remarkably, PAI also plays a role in assessing the efficacy of treatments for chronic wounds. For instance, hyperbaric oxygen therapy is a common treatment for chronic wounds aimed at improving blood and tissue oxygenation, promoting angiogenesis, and accelerating wound healing. In a study monitoring chronic wound healing under hyperbaric oxygen therapy, Mantri et al. (42) employed photoacoustic imaging (PAI) to track changes in hemoglobin oxygen saturation and perfusion, generating 3D maps of forearm vascular oxygenation that validated the correlation between PA signals and hemodynamic parameters. Additionally, there was a significant increase in radial artery oxygen saturation after hyperbaric oxygen therapy, indicating that PAI can be used to evaluate the efficacy of hyperbaric oxygen therapy for chronic wounds. Nevertheless, limitations of photoacoustic imaging due to skin thickness and pigmentation (melanin concentration) should be noted. Variations in skin thickness and pigmentation may affect the penetration of light into tissue and the accuracy of measurement results. Subjects with darker skin may experience increased light absorption at the skin surface, leading to reduced sO₂ readings, decreased penetration depth, and blurred imaging of the underlying vascular system.

In summary, photoacoustic imaging holds significant application value in the diagnosis and treatment monitoring of chronic wounds, but further research is needed to overcome its limitations and to enhance its accuracy in clinical applications. With the continuous development and optimization of the technology, photoacoustic imaging is expected to become an important tool for assessing chronic wounds in the future.