Photoacoustic/ultrasound dual-modal imaging of human nails: A pilot study

1. Introduction

Nails including both fingernails and toenails play a crucial role in cosmetic appearance and protection. Nail disorders are common and impact health and quality of life, encompassing a wide range, including fungal infections, structural abnormalities, and pigmentary changes. The nail unit, consisting of the nail plate, adjacent soft tissues, blood vessels, and nerves, is characterized by the nail plate being a layered keratin structure that overlies the nail matrix and bed. The nail matrix serves as the primary site for nail growth, whereas the nail bed provides essential support to the nail plate. The complex vascular and nerve supply ensures adequate blood flow and sensory function.¹

The nail bed, a specialized structure within the integumentary system, plays a pivotal role in the broader context of skin blood circulation.¹ This nail bed–skin relationship is underpinned by the nail bed’s unique vascular architecture, which is intricately linked to the cutaneous microcirculation. The nail bed’s capillary loops, extending from the dermal papillae, are not only crucial for the nourishment and growth of the nail, but also serve as a window into systemic vascular health. These capillaries, characterized by their high density and distinct morphology, facilitate a direct and efficient exchange of nutrients and oxygen between the bloodstream and the nail tissue. Furthermore, the nail bed’s vascular network is highly responsive to systemic changes, making it an invaluable site for noninvasive monitoring of circulatory dynamics. Alterations in nail bed perfusion can reflect broader changes in cutaneous and systemic circulation, often preceding clinical manifestations of various systemic diseases. Impaired blood flow to the arteries can significantly impact the finger pulp and nail area. In their uncontrolled study, Samman and Strickland examined nail abnormalities in 41 patients showing signs of peripheral vascular disease.² They noted that symptoms such as nail separation (onycholysis), horizontal nail ridges (Beau’s lines), fragile and thin nails, and yellowing of the nails were all linked to ischemia when other causes were ruled out. Thus, the study of nail bed circulation offers a unique vantage point for understanding the complexities of skin blood flow and its implications for overall health.

Common diagnostic methods for nail disorders include visual examination, ultrasound (US), and magnetic resonance imaging (MRI), each with its advantages and limitations. Visual examination is the most straightforward approach, offering ease and immediacy. However, its major limitation is the lack of in-depth information. It primarily relies on surface observation and cannot provide insights into the deeper structures of the nail or detect subtle changes beneath the nail plate. US imaging provides more detailed insights than visual examination, revealing the structure of the nail plate and bed. While it offers greater depth than visual inspection, its resolution is limited. This means that finer details may not be captured effectively, potentially missing minute pathological changes. MRI is known for its high-resolution imaging capabilities. It offers detailed views of the nail unit’s anatomy and pathology, making it a powerful tool for diagnosis. However, MRI’s primary drawbacks are its high cost and impracticality for routine screening. The procedure is time-consuming and requires specialized equipment, which limits its use to more complex cases where detailed imaging is necessary.³

Optical-resolution photoacoustic microscopy (OR-PAM) represents a noninvasive imaging method that integrates optical and ultrasonic technologies to provide high-resolution images of biological tissues.⁴ Its application in studying the human nail’s microvasculature has garnered significant interest due to the detailed visualization it offers of capillary loops and blood flow dynamics. Recent studies have utilized OR-PAM to monitor the microcirculation in the human nail bed.^5,6 These investigations have notably enabled the precise measurement of hemodynamic parameters, including blood flow rate, hemoglobin (Hb) concentration, and oxygen saturation (SO₂). Furthermore, they have provided insights into the variations in SO₂ and red blood cell flow rates across various segments of the nail capillary loop. A significant development in OR-PAM technology is its ability to achieve high-resolution imaging, enabling the detailed examination of single capillary. This advancement allows researchers to observe and quantify changes in microvessel diameter, detect abnormal bleeding, and assess reductions in capillary density. Additionally, the hydration of the nail enhances the quality of PAM images, which significantly improves the contrast and clarity of the vascular structures. These findings have profound implications for understanding vascular pathophysiology in microcirculation diseases. The capacity for visualizing and quantifying alterations in the microvascular structure of the nail bed may contribute to the early detection and surveillance of conditions such as Raynaud’s phenomenon and systemic scleroderma.⁵ Despite the significant advancements of OR-PAM in studying the nail microvasculature, there are notable limitations to consider. Typically, OR-PAM exhibits a penetration depth of approximately up to 1mm.⁷ This limitation precludes the technique from accessing the deeper layers of the nail bed, thereby impeding a thorough evaluation of the blood circulation in these more profound regions. Consequently, these two studies reveal the potential of optical imaging for the evaluation of human nails, but highlight the need for research into an optical high-resolution imaging method that surpasses the penetration depth of OR-PAM imaging. This would be more advantageous for the detection and assessment of the pathophysiological states in nail diseases. Photoacoustic tomography (PAT), a novel hybrid imaging technique, has been increasingly recognized for its ability to provide crucial structural and functional insights.⁸ One of the most notable advantages of PAT is its significant penetration depth compared to purely optical imaging techniques.^9,10,11 Typically, PAT can penetrate several centimeters into biological tissues, depending on the wavelength of the laser and the specific tissue properties. This depth of penetration allows for the visualization of structures and processes that lie beyond the superficial layers, which is a substantial advancement over traditional optical imaging methods.¹² In terms of spatial resolution, PAT excels by offering resolutions that range from sub-micrometer to a few micrometers, contingent upon the imaging setup and the frequency of the US detector used.^13,14,15 This high spatial resolution enables the detailed visualization of microvascular structures and cellular arrangements, which are crucial for understanding various biological processes and disease mechanisms.^16,17 Currently, due to its combination of excellent penetration depth and spatial resolution, PAT finds extensive applications across a broad spectrum of biomedical research and clinical diagnostics.¹⁸ Its ability to provide detailed images of vascular, tissue oxygenation, and molecular composition makes it an invaluable tool in areas such as oncology, neurology, and cardiovascular studies.^{17,19,20,21,22} Furthermore, PAT has demonstrated significant potential and applicability in the field of dermatology.²³ Its capability to image skin tissue with high resolution and depth has opened new avenues for diagnosing and understanding skin-related diseases, including skin cancers and vascular lesions.^{24,25,26,27,28} A groundbreaking computing method for 4D spectral-spatial imaging in photoacoustic dermoscopy has emerged, offering enhanced quantitative analysis by accounting for the intricate properties of multilayered skin.²⁶ This is complemented by the development of a miniaturized photoacoustic probe, which provides high-resolution, deep penetration imaging of subcutaneous microvessels, representing a significant advancement in the diagnosis and management of vascular diseases.²⁸ Additionally, the photoacoustic microscopic biopsy technique has introduced a new paradigm for high-resolution, noninvasive imaging of deep skin structures, offering unprecedented insights into dermal angiopathy biopsies.²⁷ These outcomes have been achieved through the detection of natural chromophores like Hb, water (H₂O), lipids, and collagen (Col).^29,30 Therefore, the noninvasive nature of PAT, combined with its capability of providing functional data including blood oxygenation levels and metabolic rates, makes it an ideal tool for skin research.³¹

However, optical imaging lacks sufficient capability for anatomical structure localization, yet its integration with other imaging modalities in multi-modal imaging compensates for this limitation. The development of dual-modality PAT/US imaging represents a significant advancement in biomedical imaging, enhancing diagnostic capabilities, and precision in anatomical and functional assessments.³² This technique combines the high-contrast, molecular sensitivity of PAT with the deep tissue penetration and real-time structural imaging of US.³² Such integration offers a detailed view of tissue morphology and physiology, proving especially valuable in oncology, cardiology, and neurology.^33,34,35 In recent years, there has been growing interest in the research application of PAT/US in dermatology, facilitating dual-modality imaging for skin disease diagnosis.³⁶ This includes the investigation of primary skin diseases^37,38 as well as skin lesions secondary to systemic diseases.³⁹ As shown previously, PAT combined with US as a dual-modality imaging technique can accurately determine the depth of melanoma in vivo, offering a valuable noninvasive tool for guiding surgical interventions in partially biopsied skin tumors.³⁷ In subsequent research, Fedorov et al. confirmed that a novel system for simultaneous and co-localized US and PAT effectively measures the size and depth of skin lesions noninvasively.³⁸ In research on secondary skin diseases, Eisenbrey et al. provided a focused analysis on the oxygenation levels in nail beds associated with Raynaud’s phenomenon using PAT/US imaging.³⁹ However, their investigation centered primarily on the quantification of oxygen content, lacking a multiparameter approach necessary for a comprehensive assessment of the pathology of the nail and nail bed.

Consequently, the previous PAT/US studies have confirmed that this dual-modality method can be used for skin disease research. This study is dedicated to assessing the feasibility of employing PAT/US imaging for human nail evaluation. This study aims to utilize this dual-modality technology to more comprehensively evaluate the anatomical structure of the nail bed and its microvascular circulation changes. By overcoming the limitations of previous research, we provide a new imaging technique that offers a more detailed and integrated reflection of the pathophysiological changes associated with nail diseases. To the best of our knowledge, it is the first to explore the PAT/US dual-modality imaging of human nails under different hemodynamic influences, with a particular focus on the quantitative analysis of Hb and Col content in the nail bed.

2. Materials and Methods

2.1. Human subjects

In this pilot clinical study, a total of five volunteers were recruited from outpatient departments at the Fifth People’s Hospital of Chengdu (FPHC). The inclusion and exclusion criteria were as follows: (1) Inclusion criteria included healthy individuals with no history of nail or finger-related diseases and (2) Exclusion criteria comprised the following: (a) Individuals with systemic diseases affecting peripheral circulation or nail morphology, such as diabetes, psoriasis, rheumatoid arthritis, and other related diseases, (b) Nail polish users, (c) Active smoking, (d) Individuals occupationally exposed to cold conditions or intense mechanical vibrations, and (e) Individuals with uncontrolled systemic arterial hypertension or proximal arterial disease. The FPHC’s institutional ethics committee approved the study protocols, and all participants provided informed consent before the procedures.

2.2. Imaging system

Our research group developed an imaging system dedicated to performing PAT imaging,^16,40 facilitating real-time observation of human nails. The imaging system utilizes laser light with wavelengths ranging from 680nm to 950nm, generated by optical parametric oscillators. These oscillators are powered by an Nd:YAG laser, operating at a frequency of 20Hz (Surelite, Continuum, CA, USA). Optical fibers are employed to precisely sdirect the laser light towards the imaging target. PA signals are captured by a concave array transducer spanning 180^∘ with a radius of 50mm, operating at a central frequency of 5MHz. The transducer relays signals to 64 data acquisition channels, each utilizing a 12-bit sampling process at a frequency of 50MHz. A multiplexer is employed to achieve a time resolution of 100ms. The system achieves a spatial resolution of approximately 150$\mu$m and can image to a depth of at least 25mm. To mitigate finger movements and enhance the signal-to-noise ratio, the signal was averaged 10 times, a technique validated by previous studies.¹² US imaging was performed using a linear probe (Model S60 Pro, SonoScape Medical System, Shenzhen, China). The transducer employed for this purpose was the L12-A, which operates in B mode across a frequency range of 3.0–17.0MHz. We have specified that the system achieves an axial resolution of 0.5mm and a lateral resolution of 1.0mm. These specifications underscore the transducer’s high-resolution imaging capabilities, making it particularly well-suited for visualizing superficial structures, such as the nail. Additionally, in Color Doppler Flow Imaging (CDFI) mode, the system operates at a pulse repetition frequency of 1.5kHz.

2.3. PAT/US nail imaging

Volunteers were required to abstain from consuming any food, caffeine, or medication for at least 1h while resting in a quiet room maintained at 25^∘C before the experiment. Their health status was assessed and confirmed through their medical history and a physical examination. The flowchart of the experimental protocol is as shown in Fig. 1.

The nails were imaged along the transverse and longitudinal planes using PAT/US imaging, respectively. Each volunteer’s hand was held in a fixed and comfortable position, and the imaging scan focused on the nail. Both PAT and US scans were performed by a registered sonographer (L.X.), who possesses over eight years of experience in US diagnostics. Laser protective eyewear is mandatory for all participants present in the room during the scanning process. PAT/US imaging was performed on the nails of all volunteers by the following steps.

2.3.1. Image acquisition of the median sagittal longitudinal section

(A) US Imaging The examination began with the collection of longitudinal median sagittal section images of the distal phalanges of each volunteer’s dominant hand (all right hands), which was positioned with the dorsal side on the imaging table. The CDFI was adjusted to maximize the display of microvessels while ensuring the elimination of Doppler motion artifacts. After applying the acoustic coupling gel, US imaging was conducted to capture the blood flow signals in and around the nail bed. For each section, three representative CDFI images were collected. Subsequently, the US imaging positions on the skin surface at the head and tail sides of the respective section were marked using a marking pen.

(B) PAT Imaging The dorsal side of the right hand remained positioned on a specially designed bracket, maintaining a stable hand posture. The corresponding longitudinal PAT imaging at the previously marked skin positions under US imaging indicates the locations for laser irradiation. After applying acoustically coupled gel, sagittal scans were performed on the nail bed (via the nail), nail fold, and distal finger segment of the right hand of the volunteers. PAT imaging was performed using PA signals excited by the laser pump at 840nm wavelengths.

2.3.2. Image acquisition of the cross section

Subsequent to the longitudinal imaging, we employed a $90^{\circ }$ rotation of the transducer and optical fiber to capture cross-sectional PAT images of the nail. During the real-time PAT imaging process, we adjusted the horizontal position of the transducer and fiber assembly, enabling continuous imaging from the fingertip to the distal nail matrix. We selected a wavelength of 840nm for irradiation and activated the laser. In order to provide a comprehensive view of the entire nail unit, a series of six consecutive cross-sectional images of the thumb nail were captured.

Later, multispectral PAT imaging was performed using PA signals excited by the laser pump at 760nm, 840nm, and 910nm wavelengths, and functional tissue parameters such as oxygenated hemoglobin (HbO₂), deoxygenated hemoglobin (HbR), SO₂, and Col were measured. Subsequently, we employed the tourniquet test and cold stimulus experiment to assess alterations of these tissue functional parameters, thereby simulating them under various physiological conditions. Furthermore, we conducted a comparative analysis of the PAD alterations before and after conducting these two tests. In this context, PAD within the nail bed is delineated as the cumulative intensity of PA signals encompassed within the arc-shaped region of interest (ROI) of the nail bed, normalized by the ROI’s area.

(A) Tourniquet test The steps of the tourniquet test are as follows: Initially, the volunteer remained seated in a relaxed state, the stimulation was then induced by pressing the right upper arm with an air cuff (shown in Fig. 1). In order to trigger arterial occlusion, we inflated the air cuff above 20mmHg of the volunteers’ systolic blood pressure for 2min to reach the arterial occlusion pressure. The successful occlusion of the artery was verified by a pulse oxygen apparatus to verify that the fingertip could not detect the pulse of the artery. Consequently, an air cuff pressure ranging from 132mmHg to 140mmHg was applied according to the volunteers’ systolic blood pressure, and after collecting US (in CDFI mode) and multispectral PAT data, it was then released for rest. The protocol is carefully designed to avoid over-compression of the artery, causing discomfort and even pain, as described in the references.⁴¹ We recorded both PAT and US images at baseline data and 2min after the tourniquet pressure, storing all measurements for analysis.

(B) Cold stimulus test After completing the tourniquet test, volunteers rested for at least 30min before initiating the cold stimulus test. During the cold stimulus test, US (in CDFI mode) and multispectral PAT imaging were conducted. The procedure for the cold stimulus test was as follows: An ice pack was continuously placed on the volunteer’s wrist to induce a cold stimulus (as illustrated in Fig. 1), maintained for 10min. While the ice pack remained on the wrist, the PAT/US imaging of the nail was performed. After 10min, we removed the ice pack from the volunteer’s wrist. We recorded both PAT and US images at baseline (Group A), 10min after the test (Group B), and again 5min after removing the ice pack during the recovery period (Group C), storing all measurements for analysis.

2.4. Image reconstruction and data measurement

During the acquirement, we mainly focused on the nail bed, cuticle, and surrounding soft tissue. The data were captured in real-time using an onboard image reconstruction algorithm, ensuring immediate processing. Subsequently, all the raw PAT data were stored and transferred to a dedicated computer for image reconstruction, facilitating further offline analysis. The raw multispectral PAT data underwent post-processing using Matlab software. Furthermore, a maximum amplitude projection image was created using a multispectral reconstruction algorithm devised by Huabei Jiang.⁴² This approach enabled the examination of HbO₂, HbR, SO₂, and Col content. The quantification of the three-wavelength blood oxygenation was determined as

2.5. Statistical analysis

The quantitative data were expressed as mean ± SD. To compare the differences between two groups across various parameters, an independent T-test was employed. The assumptions of normality and equal variance were verified for the T-test application. Data distribution and inter-group comparisons were visually analyzed using box plots, generated through the OriginPro software (version 2022). A P-value of less than 0.05, in a two-tailed test, was considered statistically significant.

3. Results

In our study, PAT/US dual-modality imaging was conducted on five cases, comprising two males and three females, with a median age of $34.6\pm 4.6$ years (ranging from 30 to 41 years). All five volunteers (Body Mass Index, BMI range: 18.1–22.2Kg/m²; mean±SD: $20.2\pm 1.6$Kg/m²) underwent PAT/US imaging. Two sonographers (W.Y. and L.X.) independently assessed all PAT/US images, consistently rating them as displaying good image quality, regardless of the volunteers’ BMI levels, in all five cases.

We employed a wavelength of 840nm for irradiating the nail area because this wavelength corresponds to the optical absorption peak of Hb, which is more advantageous for displaying the microvascular network of the nail bed. In our multispectral PAT imaging study, we selected wavelengths of 760nm, 840nm, and 910nm, as these wavelengths correspond to significant optical absorption characteristics of Hb and Col. This selection is based on their absorption peaks in biological tissues, thereby optimizing the imaging of microstructures, particularly in revealing the microvascular network of nail bed, which holds significant importance.

In the comparative analysis of PAT and corresponding US images of the fingertip’s median sagittal plane, we can observe the distribution of PA signals from the fingertip to the nail root. This includes the proximal nail fold, nail plate, nail bed, nail matrix, and distal nail fold, which correspond precisely to the anatomical structures of the nail unit identified in the US images (Fig. 2). The PAT images (wavelength 840nm) demonstrated the nail as an arcuate band of pronounced PA signal, mirroring the arcuate strong echo in US imaging. However, significant disparities are observed between PAT and US in depicting nail bed perfusion. US imaging reveals sparse, punctate CDFI signals in the nail bed, in contrast to the richer, reticulated PA signals evident in PAT, indicative of dense cuticle capillary loops within the nail bed.⁴³ Notably, in the thumb’s nail bed, PAT delineates parallel linear PA signals, congruent with the anatomical vascular arrangement in the nail bed, and characterized by the alignment of small blood vessels along a common axis.

In the cross-sectional PAT imaging of the nail, the real-time PAT images offered a detailed visualization of the entire distal end of the finger. These images sequentially displayed the aspects from dorsal to ventral, encompassing the nail, nail bed, distal phalanx, and ventral skin. Notably, the nail plate exhibited a distinct arc-shaped acoustic signal, indicative of its dense, keratinized structure, as elucidated in the foundational nail anatomy work by de Berker.¹ Beneath the nail plate, the nail bed showed densely packed high-intensity acoustic signals, corresponding to its rich microvascular network. This aspect is crucial for understanding nail physiology and pathologies. Furthermore, the PAT images revealed varying acoustic signal intensities in the periosteum and internal structure of the distal phalanx, highlighting the heterogeneous nature of these tissues. The six cross-sectional images provided a complete representation of the nail, sequentially displaying the proximal nail fold, nail plate (distal, middle, proximal), distal nail fold, and nail matrix (Fig. 3). The nail matrix, known for its nail-producing capability, is of significant interest in nail imaging due to its role in the microcirculation. We can observe the distribution of PA signals at the nail matrix, providing valuable information about the microvascular distribution in this area, which is crucial for assessing the microcirculation at the nail matrix. Additionally, the nail matrix contains melanocytes, typically in a dormant state but capable of activation, synthesizing melanin, and transferring it to surrounding keratinocytes. This process can lead to nail pigmentation or melanonychia, characterized by black streaks on the nail plate.⁴⁴ Due to PAT’s sensitivity to melanin absorption, it can offer high sensitivity imaging of activated melanocytes in the nail matrix. This capability positions PAT as a potential new modality for detecting nail diseases involving melanocyte activation.

In this experimental protocol, the tourniquet test⁴¹ and cold stimulus test⁴⁵ are employed to modulate the perfusion of the distal phalanges and nail beds, with the objective of real-time detection of these hemodynamic changes via PAT/US imaging. The results of the experiment are discussed in the following sections.

3.1. The tourniquet test

Before the tourniquet test, a dual-layered arcuate structure of the nail exhibiting strong echogenicity is discernible, along with the hypoechoic nail bed situated beneath the nail in the transverse section of US imaging. In the CDFI mode, punctate arteriovenous blood flow signals are observable. Meanwhile, in PAT imaging, the PA signal in the central region of the nail bed is stronger than on the sides, corresponding to the blood flow distribution pattern displayed by the CDFI mode in US. We used the c-d horizontal line in Fig. 3 for regional imaging to ensure that the light irradiation range was always at the center level of the nail bed during the imaging experiment.

The arterial occlusion is achieved by applying a pressure that exceeds the systolic blood pressure of the subjects by 20 mmHg. The effectiveness of this arterial blockade is confirmed when pulse oximetry fails to detect arterial pulsations at the fingertips, indicating successful arterial occlusion. Additionally, the duration of the tourniquet application is limited to 2min. This constraint is based on observations that extending the air cuff pressure beyond this period leads to significant forearm swelling and pain, difficulty in maintaining a steady hand position, as well as dorsal hand congestion and skin erythema. Given the PAT imaging is predicated on the absorption of light by chromophores, changes in skin pigmentation could potentially introduce measurement inaccuracies in the PAT data. Therefore, the PAT acquisition time is set at 2min to mitigate these potential errors. Throughout this study, we observed no significant adverse effects. Notably, all five participants reported experiencing a transient numbness and a mild sensation of swelling and pain in the wrist. These symptoms emerged within 31–87s following the attainment of the artery occlusion pressure by the air cuff, and then subsided promptly after the pressure was alleviated.

Two minutes post the tourniquet test, there was a noticeable reduction in both arterial and venous blood flow signals in the nail bed in US imaging. However, PAT imaging demonstrated a significant increase in the nail bed in PA signals following the tourniquet test. The difference in blood flow signals between PAT and US can be attributed to the fact that US imaging (CDFI mode) is based on the Doppler effect to visualize the movement of red blood cells. When the velocity of venous blood flow decreases, it results in the inability of venous blood flow to be displayed in CDFI mode. During the tourniquet test, as the venous return is obstructed, congestion gradually developed in the volunteer’s hand, with subcutaneous veins on the dorsum of the hand dilating, leading to stasis of blood flow in the veins at the fingertips and nail bed. This led to a significant increase in PA signals in the PAT images. Therefore, due to the different imaging principles, this phenomenon indicates that PAT imaging can more accurately reflect the changes in the blood circulation of the nail bed. Figure 4 presents comparative images depicting changes in PA signals for various parameters (Col, HbO₂, HbR, and SO₂), along with alterations in CDFI signals in US imaging before and after the tourniquet test.

The quantitative analysis results of multispectral PAT are as follows: Intra-group comparison between baseline and 2min post-tourniquet test showed Col-PAD ($p=0.390$), HbO₂-PAD ($p=0.858$), HbR-PAD ($p=0.025$), and SO₂-PAD ($p=0.070$). This indicates that there is statistical difference in HbR-PAD, whereas the variations in Col-PAD, SO₂-PAD, and HbO₂-PAD do not demonstrate statistical significance. The data for Col-PAD, HbO₂-PAD, and HbR-PAD are relative values, while SO₂-PAD is an absolute value. The quantitative data of functional parameters from the tourniquet test are presented in Table 1.

In the analysis of multispectral PAT reconstruction data, the quantitative Col results for the nail bed indicate no distributional change before and after the tourniquet test. This suggests that PAT can effectively capture Col distribution within the nail bed and perform quantitative analysis of its content. Compared to the surrounding tissue, the PA signal of Col in the nail bed is significantly stronger. This is related to the unique anatomical characteristic of the nail bed being rich in collagen. The structure and function of the nail bed are critical to nail health, with Col within the nail bed playing a pivotal role. As described by de Berker¹ in their study on nail anatomy, the dermis of the nail bed contains minimal fat and is firmly connected to the underlying periosteum through Col, without the presence of sebaceous glands or hair follicles. This finding underscores the importance of collagen in maintaining the structural integrity of the nail bed and its tight connection to the underlying periosteum, thereby providing necessary support for healthy nail growth. Therefore, PAT holds promise as a new method for assessing the extent of nail bed damage and monitoring the repair process.

Additionally, following the tourniquet test, despite successful occlusion of distal arterial blood supply, quantitative analysis revealed no significant decrease in HbO₂ and SO₂ levels, while HbR values increased. This is attributed to the high occlusion pressure, which completely obstructs venous blood flow before arterial occlusion, leading to venous return impairment, increased venous pressure in the fingertip, and consequently, an augmented blood volume in the nail bed’s veins, reflected in elevated HbR values. Simultaneously, arterial blood in the fingertip does not fully deoxygenate within 2min, hence the lack of a notable decrease in HbO₂. This indicates that PAT is highly sensitive to changes in blood circulation in the nail bed.

3.2. The cold stimulus test

Before the cold stimulus test, strip-like blood flow signals were visible at the nail bed under US imaging (CDFI mode), and the multispectral PAT reconstruction images showed the distribution of PA signals of Col, HbO₂, HbR and SO₂ at the nail bed. After the 10min cold stimulus test, US imaging displayed a redistribution of blood flow signals, and PAT images revealed a decrease in the signal intensity of HbO₂ and HbR in the central region of the nail bed (Group B). Following the 15min cold stimulus test (following a 10min cold stimulus and a 5min recovery period after removing the ice pack), US images showed an increase in blood signals (Group C). Correspondingly, the multispectral PAT images displayed stronger PA signals, characterized by a significant increase in the PA signal intensity of HbO₂ and HbR (Fig. 5).

The inter-group comparison results are as follows (Fig. 6): There was a statistically significant difference between Groups A and B in both HbO₂-PAD and HbR-PAD. Between Groups B and C, significant differences were observed in SO₂-PAD, HbO₂-PAD, and HbR-PAD. Additionally, between Groups A and C, significant differences were noted in SO₂-PAD and HbR-PAD. The analysis of these inter-group comparisons confirmed that 10min after the cold stimulus (Group B), there was a decrease in both HbO₂ and HbR in the nail bed compared to before the cold stimulus. This suggests that initial vasoconstriction occurred following cold exposure. When fingers are exposed to a cold environment, arterial constriction happens first as a protective response to minimize heat loss. This constriction typically occurs rapidly, within seconds to minutes of exposure to cold. Five minutes after the removal of the cold stimulus (Group C), compared to before the cold stimulus (Group A), there was a decrease in SO₂ and an increase in HbR. Compared to the condition at 10min of continuous cold stimulus (Group B), there was a decrease in SO₂, and increases in both HbO₂ and HbR. Under sustained cold stimulus, intermittent vasodilation may occur, a phenomenon known as ‘Cold-Induced Vasodilation’. This dilation usually begins minutes after the onset of cold exposure and acts as a protective mechanism against frostbite. The response of the nail bed veins may not be as pronounced as that of the arteries, but they also participate in regulating blood flow. Under cold stimulus, veins might slightly dilate to facilitate blood return. The compensatory arteriovenous dilation during the recovery phase after cold exposure facilitates renewed blood flow.⁴⁶ Therefore, the increases in HbO₂ and HbR observed in Group C during the recovery period can be attributed to this mechanism. Moreover, consistent with the results of the tourniquet experiment, there was no significant change in the Col content of the nail bed before and after the cold stimulation experiment, suggesting that short-term variations in hemodynamics do not affect the distribution of Col within the nail bed.

Thus, the cold stimulus test findings suggest that PAT is capable of detecting changes in the microcirculation of blood in the nail bed, consistent with the result reported by Eisenbrey et al.³⁹ Unlike the previous study, this study extends the investigation by conducting multi-parameter quantitative analysis, especially in simulating different physiological states to observe the variations in the content of HbO₂, HbR, and Col within the nail bed, providing a more comprehensive understanding of the changes in nail bed hemodynamics.

4. Discussion

The promising results from our pilot clinical study underscore the substantial potential of PAT/US dual-modality imaging as a real-time, noninvasive diagnostic tool for visualizing human nails. This technology stands out for its unique ability to provide a comprehensive analysis of Hb and collagen Col content in the nail bed, which are crucial for assessing nail bed health and monitoring the repair process.

The integration of PAT/US imaging into clinical settings could revolutionize the approach to diagnosing and monitoring nail disorders. Its ability to offer real-time, quantitative analysis of critical biomarkers within the nail bed — such as HbO₂, HbR, and SO₂ levels — provides clinicians with valuable insights into the microcirculatory dynamics and structural integrity of the nail bed. This is particularly relevant for early detection of pathological changes and for evaluating the efficacy of therapeutic interventions. Moreover, the dual-modality approach enhances diagnostic accuracy by combining the strengths of both PAT and US imaging. While US imaging excels in visualizing the anatomical structure of the nail and surrounding tissues, PAT adds functional information by mapping the distribution of oxygenated and deoxygenated Hb, as well as collagen content. This synergy allows for a more detailed assessment of nail bed health than could be achieved with either modality alone. The advantages of PAT/US dual-modality imaging extend beyond its diagnostic capabilities. Its noninvasive nature and the absence of ionizing radiation make it a safer alternative to traditional imaging techniques, reducing patient exposure to potential risks. Additionally, the potential for real-time imaging facilitates immediate clinical decision-making, enhancing patient care by allowing for timely interventions.

Integrating the promising aspects of PAT/US dual-modality imaging into routine clinical practice offers transformative potential for noninvasive diagnostic techniques in dermatology. However, realizing this potential hinges on overcoming several challenges, including the standardization of imaging protocols, training clinicians in the nuanced interpretation of PAT/US images, and further validation of the technology across larger and more diverse patient cohorts. Our study, while pioneering, acknowledges limitations that highlight the necessity for future advancements to optimize the utility of PAT/US imaging. The integration process is impeded by the current inability to display quantitative parameters such as HbO₂, HbR, and SO₂ content within nails in real-time, a relatively small sample size that limits comprehensive statistical analysis, and the separate execution of PAT and US imaging, which could lead to inconsistencies in acquisition sections between these modalities.

Therefore, future advancements are required to further enhance the PAT/US dual-modality system in several aspects: (1) Real-time display: Implementing real-time PAT/US imaging on monitors in a split-screen format to facilitate immediate clinical decision-making; (2) Fast multispectral parameter processing: Enhancing the system to process fast multispectral parameters and conduct a more comprehensive quantitative analysis of nails, offering real-time functional information; (3) High temporal resolution: Achieving high temporal resolution to better capture the dynamic physiological processes within the nail bed; (4) Three-dimensional imaging: Developing three-dimensional imaging capabilities for detailed vessel information of the nail bed, providing a more comprehensive view of nail bed microcirculation; (5) Expanded clinical studies: Conducting further studies in expanded cohorts of healthy individuals and patients with nail disorders to validate the efficacy and reliability of PAT/US dual-modality imaging across a broader spectrum of clinical scenarios.

5. Conclusions

In conclusion, PAT/US dual-modality imaging exhibits substantial potential as a real-time diagnostic tool for visualizing human nails. It shows potential for evaluating patients with nail damage, focusing on the early detection of changes in nail bed health and the dynamic monitoring of the nail bed repair process. This study is expected to contribute to the advancement of noninvasive diagnostic techniques in dermatology, enhancing our understanding of the changes in the nail bed under different hemodynamic conditions.

Acknowledgments

We thank Liyan Bian for the meticulous proofreading of this paper. This research was supported by the program of Chengdu Fifth people’s hospital Fund, No. KYJJ 2021-29, the Xinglin Scholars research program, No. YYZX2021037, the Chengdu Medical Research Project, Nos. 2022055 and 2023022, Chongqing Education Commission, Youth Fund (No. KJQN202000607) and Chongqing post-doctoral research project (special funding project, No. 2021XM3040).

Conflict of Interest

The authors declare that there are no conflicts of interest relevant to this paper.

ORCID

Yanting Wen  https://orcid.org/0000-0002-1049-5110

Yijie Huang  https://orcid.org/0009-0009-0373-8733

Lin Huang  https://orcid.org/0000-0003-3616-7315

Dan Wu  https://orcid.org/0000-0002-4870-5304

Renbin Zhong  https://orcid.org/0000-0003-0179-4538

Shixie Jiang  https://orcid.org/0009-0003-6236-8775

Yudi Zhang  https://orcid.org/0009-0002-4820-1103

Ting Liu  https://orcid.org/0000-0003-2349-0842

Xiaotian Liu  https://orcid.org/0009-0009-2237-5405

Huabei Jiang  https://orcid.org/0000-0003-0008-0074