1. Abstract Early detection of Alzheimer’s disease (AD) remains limited by the inability of current imaging biomarkers to capture dynamic tissue changes preceding overt neurodegeneration. Conventional approaches primarily reflect molecular burden or structural atrophy and incompletely characterize the intermediate microenvironmental remodeling linking early pathology to clinical progression. Here, we introduce optical conductivity as a novel imaging biomarker characterizing tissue responses to optical-frequency electromagnetic fields. Using an optical wave tomography (OWT) framework, optical conductivity is reconstructed from diffuse optical measurements by integrating absorption-driven energy deposition with microstructure-dependent field redistribution. Applied to an APP/PS1 mouse model across young and old cohorts, optical conductivity revealed a significant age-by-disease interaction (F = 9.20, p = 0.0071). Quantitatively, its value was reduced in young AD animals relative to controls (∼13%; Cohen’s d = 1.93) but showed attenuation or reversal in older animals, indicating a stage-dependent crossover. Consistently, optical conductivity achieved strong classification performance in the young cohort (AUC = 0.80), outperforming absorption and exceeding scattering, with improved performance via multimodal integration (AUC = 0.87). Together, these findings suggest that optical conductivity captures stage-dependent microenvironmental remodeling not detectable by conventional optical parameters, offering potential for early detection and disease staging.

Microwave-induced thermoacoustic imaging, as an emerging biomedical imaging technique, combines the high contrast of microwave imaging with the high spatial resolution of ultrasound imaging. Microwave-induced thermoacoustic microscopy, as an important branch of this technology, retains these advantages while possessing the ability to visualize finer tissue characteristics. However, traditional raster scanning mechanisms introduce interference into microwave field distribution due to mechanical motion, thus necessitating multiple signal average to maintain signal-to-noise ratio. Additionally, the idle time during motor movement results in extended single-scan durations, limiting its practical applications. To address these limitations, this work proposes a rapid imaging system based on one-dimensional galvanometer scanning. The system employs a hybrid galvanometer-translation stage architecture and an optimized scanning strategy to minimize microwave field interference, reduce the number of signal averages and shortens the idle time, ultimately achieving more than a tenfold improvement in imaging speed. A specially designed timing control algorithm ensures the precise synchronization of microwave excitation, galvanometer motion, and ultrasound detection, while the reconstruction algorithm suitable for the optimizing scanning method effectively corrects distortions generated in the scanning process. The system performance is assessed through phantom and ex vivo tissue experiments. Resolution tests show hundred-micrometer resolution along all three axes (332 μm × 324 μm × 79 μm), while contrast and depth imaging experiments confirm its ability to clearly distinguish targets with different conductivities, achieving an effective detection depth of at least 10 mm in tissue. Early tumor mimicking experiments further demonstrate the ability of the system to identify lesion boundaries, preliminarily revealing its potential for rapid tumor margin assessment. This approach maintains the imaging quality of microwave-induced thermoacoustic microscopy while enhancing imaging efficiency and system stability, thereby laying a crucial foundation for advancing the technology from laboratory research to clinical applications.

Gouty arthritis (GA) is a common inflammatory joint disease triggered by the deposition of urate crystals. Early noninvasive and accurate detection is of great clinical importance for diagnosis and treatment. Current clinical imaging modalities play a key role in the detection of GA, but they still possess inherent limitations in evaluating early inflammation, soft-tissue resolution, radiation safety, or examination cost. Microwave-induced thermoacoustic imaging (MITAI), as a novel noninvasive hybrid imaging modality, combines the advantages of high contrast and deep penetration from microwave imaging with the high resolution of ultrasound imaging, enabling in vivo visualization of differences in tissue dielectric properties and anatomical structures. With unique technical merits, this newly developed imaging approach shows potential in identifying tissue lesions and inflammatory changes in human joints. In this study, real-time MITAI was applied to the in vivo detection of GA. The bilateral dorsal foot joints of a patient were imaged, with a healthy volunteer serving as a control. The results showed that, compared with the healthy joint, GA-affected joints exhibited a characteristic abnormal hyperintense sign in the MITAI images. This finding is likely attributed to pathological changes in GA—such as inflammatory edema, crystal deposition, and soft tissue swelling—which alter the dielectric properties and microwave absorption capacity of the tissues. This preliminary study suggests that real-time MITAI has the potential to distinguish gouty arthritis from normal joints, and may be developed as a novel, noninvasive, radiation-free, and low-cost imaging modality for GA detection.This work represents an initial step toward its future clinical application.

To address the issue of incomplete acoustic processes in microwave-induced thermoacoustic imaging (MITAI) simulations, this study proposes coupling structural mechanics (SM) and pressure acoustics (PA) fields inside and outside tissues, respectively, to visualize the entire MITAI physical process. Based on this, we investigate the MITAI artifacts formation mechanism from a mechanical property perspective. This study establishes a dual-imaging target model based on COMSOL, coupling electromagnetic, heat transfer, and SM and PA modules to achieve a complete MITAI process. This method is verified by extracting the signals for image reconstruction. Furthermore, four complex biological tissue models are constructed based on the presence and location of bone, to explore the mechanical impact of bone on surrounding soft tissues. The dual-target results show that the proposed method accurately characterizes the multiphysical phenomena of microwave excitation, energy deposition, thermal expansion, and ultrasound propagation in the MITAI process. The images correctly represent both dielectric functional information and structural features of the model, while also revealing artifacts caused by the signal reflection. The image of the dual-target phantom further validates this finding. Four complex models' results show that although MITAI clearly distinguishes bones and different soft tissues, the bones degrade image quality, with greater degradation occurring in soft tissues closer to the bones. The proposed method can achieve complete MITAI process visualization, overcome incomplete acoustic simulation, and elucidate the MITAI artifacts generation mechanism from mechanical properties. This work holds considerable value for advancing MITAI mechanism research and guiding biological tissue imaging applications.

In microwave-induced thermoacoustic brain imaging, there is acoustic heterogeneity between the skull and brain tissue, as well as between different brain tissues. If a simple uniform medium model is directly used for image reconstruction, sound velocity mismatch and scattering and reflection effects can typically lead to severe distortion of the projected acoustic path and pronounced signal tailing, thereby exacerbating image artifacts and reducing imaging quality. For microwave-induced thermoacoustic brain imaging, existing optimization methods have mostly focused on sound velocity redistribution, while relatively fewer studies have addressed the suppression of scattering and reflection artifacts. Moreover, these methods have not yet been practically applied to in vivo detection experiments. In this study, the issue of artifacts caused by scattering and reflection was investigated, and a tail attenuation algorithm was proposed and developed to suppress tail signals and improve the quality of microwave-induced thermoacoustic brain imaging images. The method proposed in this study is simple and efficient, has computational efficiency. The experiments are conducted on normal live rats using a 1.3 GHz microwave generator to evaluate the performance of the trailing attenuation algorithm. The results showed that the in vivo thermoacoustic brain images processed by the algorithm can effectively reduce the artifacts from the environment and internal tissue, resulting in a 9.2dB increase in signal-to-noise ratio and a 5% increase in contrast. This clearly indicates that the tail decay algorithm can effectively improve the quality of thermoacoustic brain images, and is expected to promote the development of microwaveinduced thermoacoustic imaging technology in brain disease detection and monitoring.

We present a novel metal reflector based illumination strategy for optimized thermoacoustic tomography (TAT) of intracerebral hemorrhage (ICH). The strategy can capture stray microwave energy from the imaging area, redirects it synchronously to brain tissues, disrupts original electric (E)-field boundary conditions, and reshapes the E-field distribution into uniformity. This renders TAT more dependent on inherent dielectric property differences of brain tissues than external E-field variations, thereby enhancing image quality. Numerical simulations were performed to evaluate the reflector dimensions, and it was determined that a half-wavelength diameter reflector provided an optimal performance, achieving minimal E-field intensity deviation, optimal energy absorption density in brain tissue, and superior lesion-to-tissue contrast. Based on the optimal parameters from the simulations, a metal reflector was fabricated and integrated into the TAT system for in vivo detection in normal and ICH rats. Imaging of rat brain was then conducted and demonstrated that the reflector enhanced TAT significantly improved tissue boundary definition, hemorrhage/edema visibility, image sharpness/contrast and intergroup symmetry differences. The TAT image quality is further improved when an image reconstruction algorithm integrating logarithmic depth weighting with multi-scale Retinex is used. Anatomical photographs and HE-stained pathological examinations and MRI scans of the brain tissue imaged proved these TAT imaging findings. The reflector-optimized strategy has low cost and high system compatibility. This study suggests that our reflector-based TAT may offer a new yet practical tool for in vivo dielectric brain research and auxiliary diagnostics of neurological diseases.

OBJECTIVE:To develop and validate an ultrasound-guided low-intensity focused ultrasound (USg-LIFU) platform for subject-specific deep-brain neuromodulation and test whether ventral tegmental area (VTA) targeting modulates alcohol-related behaviour in a mouse model of alcohol use disorder. METHODS:USg-LIFU uses ultrasound tomography (UT)-assisted acoustic velocity mapping and subject-specific finite element method (FEM) simulations to incorporate reconstructed acoustic-property information into FEM-based focal prediction and guide mechanical alignment of the predicted focal zone toward the target. Targeting accuracy is verified ex-vivo using high-intensity focused ultrasound (HIFU) to create thermal lesions in mouse brain tissue. In-vivo neuromodulatory effects were tested in cHAP mice (n=10) that underwent six USg-LIFU sessions targeting the VTA on alternating days and compared with sham controls (n=5). Daily measures were recorded to assess alcohol and water intake, and blood alcohol concentration (BAC) was assessed before, during, and after treatment. RESULTS:Simulations showed that changes in brain acoustic velocity shifted focal position and altered pressure distribution, underscoring the need for subject-specific correction. Ex-vivo studies demonstrated that the optimized focus converged on the predefined intracranial target, with visible thermal damage at the intended location after sonication. In-vivo, most LIFU-treated mice showed reduced ethanol intake and increased water consumption after treatment initiation, whereas sham animals maintained or strengthened alcohol preference. Across the treatment period, mean alcohol intake decreased by approximately 51% in the treated group, consistent with reductions in BAC measurements. CONCLUSION:USg-LIFU achieved accurate subject-specific targeting and produced measurable modulation of alcohol-related behaviour, supporting its potential as an accessible platform for non-invasive deep-brain neuromodulation.

Accurate multimodal registration between magnetic resonance imaging (MRI) and photoacoustic tomography (PAT) is essential for localizing functional hemodynamic signals to specific neuroanatomical structures, such as the hippocampus, in preclinical brain studies. However, the inherently low soft-tissue contrast and high noise characteristics of PAT images render precise alignment challenging. Existing registration methods typically rely on global optimization and lack the structural priors necessary for localizing small and complex regions like the $\mathbf{C}$ shaped hippocampus. To address the challenge of aligning MRI and PAT images under significant modality differences and lowcontrast conditions, this paper proposes a structural priorconstrained method for MRI-PAT brain image registration and segmentation. First, the SAM-Med2D model is employed to segment MRI images and generate hippocampal region masks, which are introduced as structural priors into the registration process to enhance constraints on small target regions. Second, ORB feature matching combined with the RANSAC algorithm is utilized to estimate affine transformation parameters, using the superior sagittal sinus as a reference anchor to achieve initial cross-modal alignment. Building upon this, a structural consistency constraint term is constructed to optimize PAT segmentation results and improve structural overlap. Experimental results on mouse brain MRI-PAT data demonstrate that the proposed method improves alignment accuracy and enhances structural consistency in the hippocampal region. This work provides a robust solution for integrating structural and functional imaging data in small animal studies.

Background: Sleep deprivation (SD) has been increasingly recognized for its detrimental effects on human health, particularly its severe damage to brain tissue. Photoacoustic tomography (PAT), hybrid imaging modalities combining light and sound, enable real-time, high-spatial-resolution imaging to rapidly capture SD-induced alterations in cerebral oxygen metabolism. Objective: To investigate the impact of SD on survival and analyze cerebral oxygen metabolic changes as potential predictors of mortality. Methods: A custom-built photoacoustic tomography system was employed to scan the hippocampal coronal plane of SD mice. Twenty-three mice were randomly assigned to three groups: sustained SD (SSD, n=9), intermittent SD (ISD, n=7), and control (n=7). The SSD group underwent continuous sleep deprivation (20-hour deprivation starting daily at 9:00 AM, 4-hour rest, repeated in 72-hour cycles followed by 24-hour rest/data collection). The ISD group alternated between 72-hour deprivation cycles, 24-hour rest/data collection and 72-hour rest periods. PAT data were collected every 72 hours later to monitor cerebral oxygen metabolism. Results: After 18 deprivation cycles, significant survival differences were observed among groups (Logrank P=0.005), with the SD group showing the highest mortality. PAT data revealed that deceased mice exhibited significantly lower photoacoustic signals for oxygenated hemoglobin (HbO(2)) (P=0.02) and oxygen saturation (sO(2)) (P=0.001) compared to survivors in their final scans, while deoxygenated hemoglobin (HbR) showed no difference (P=0.147). Conclusion: Prolonged sleep deprivation directly accelerates mortality, with cumulative deprivation duration correlating positively with mortality risk. HbO(2) and sO(2) may serve as potential early warning indicators for SD-induced fatal outcomes.

Thyroid cancer has emerged as a globally prevalent malignancy, exhibiting a continuously rising incidence rate. While conventional ultrasound (US) is the current primary screening modality, its reliance on morphological features often results in insufficient specificity and consequent over-biopsies. This study introduces an ultrasound-multispectral photoacoustic (PAUS) dual-modality imaging system. Using a multi-wavelength (760-930 nm) spectral unmixing algorithm, it accurately resolves key molecular components including oxyhemoglobin (HbO), deoxyhemoglobin (HbR), total hemoglobin (HbT), and oxygen saturation (sO(2)). A feature-level fusion algorithm integrates PA and ultrasound images, incorporating spatial registration and an adaptive weighting strategy to achieve complementary voxel-level features across modalities. This facilitates comprehensive in vivo quantitative analysis of thyroid nodules, yielding multi-parametric structural-functional data to aid in benign-malignant differentiation. Clinical in vivo PAUS data were acquired from thyroid nodule cases and validated against conventional US and normal thyroid tissue.

As the largest organ of the human body, the skin serves as the first barrier protecting the body from external harm. Current diagnostic methods for skin diseases primarily rely on visual inspection by physicians and skin biopsies, but these approaches have significant limitations and cannot comprehensively evaluate vascular functional changes in skin lesion areas. This study proposes a non-invasive skin detection method based on Multispectral Photoacoustic Tomography (MSOT), enabling real-time, high-resolution three-dimensional (3D) imaging of the skin to provide high-volume structural and functional information. Using an MSOT system, we imaged melanocytic nevi and acne lesions on volunteers' skin, clearly observing the distribution of melanocytic nevi and presenting their spatial structure within the skin through 3D reconstruction. Additionally, 3D Photoacoustic scanning of acne lesions revealed significantly enhanced Photoacoustic signals around acne sites compared to normal skin, particularly in inflamed regions. This enhancement primarily originates from local vasodilation and increased blood flow. Through multispectral reconstruction, we quantitatively analyzed hemodynamic parameters (HbR, HbO, water content, and sO(2)) in acne areas, identifying significant differences between lesion regions and normal skin. Furthermore, an optical flow algorithm based on blood spatial heterogeneity was employed to measure blood flow velocity in acne lesions, demonstrating that flow velocity in these areas was significantly higher than in normal skin. These results highlight the powerful capabilities of MSOT in detecting and evaluating skin lesions, positioning it as a promising tool for the diagnosis and management of skin diseases.

Objective This study aims to develop a microwave-induced thermoacoustic and ultrasound dual-modality microscopy system that integrates the advantages of both imaging techniques to investigate the dielectric properties of biological tissues at a microscopic level. Methods This paper first discusses a method to enhance system resolution by combining short-pulse microwave excitation with high-frequency point-focused ultrasonic transducer detection. A three-dimensional microwave-induced thermoacoustic microscopic imaging system was constructed based on this approach and further developed into a dual-modality system capable of both thermoacoustic and ultrasonic imaging. The image reconstruction and dual-modality image fusion strategies are also described. Subsequently, experiments were conducted in the following sequence: imaging of copper wires to evaluate the system's spatial resolution along the X/Y/Z axes; imaging of tubes containing 3% and 6% saline solutions and tubes filled with coupling agent/vegetable oil to demonstrate the complementary information provided by the two modalities; imaging of brain tissue and bone-cartilage samples to assess the applicability of the technology; and osteoporosis detection to validate the disease diagnostic capability of the dual-modality system. The microwave-induced thermoacoustic and ultrasound microscopic images of these samples were verified against corresponding photographs or micro-CT images. Results The thermoacoustic and ultrasonic images of the copper wire closely matched the physical photograph. The three-dimensional resolutions of the microwave-induced thermoacoustic and ultrasound imaging systems, as estimated from the copper wire experiment, were 178x178x88 mu m3 and 177x177x42 mu m3, respectively. These measured values align well with theoretical predictions. The dual-modality imaging system successfully combines dielectric property differences captured by thermoacoustic imaging and acoustic impedance variations captured by ultrasound imaging, thereby providing both functional and structural information of the samples. Specifically, the system distinguished between tubes containing saline solutions of different concentrations and those containing vegetable oil, demonstrating strong spatial consistency with physical photographs. The thermoacoustic image contrast among saline solutions corresponded to theoretical dielectric properties, while the ultrasonic contrast between saline and oil reflected their difference in acoustic impedance. The system identified multiple brain tissue structures, including the cortex, hippocampus, superior colliculus, corpus callosum, cingulate cortex, and striatum. The bimodal imaging approach exhibited superior performance, visualizing tissue structures with greater clarity and detail than either modality alone. The brain tissue images were consistent with physical photographs, tissue dielectric properties, and publicly available anatomical atlases. The bimodal system clearly delineated cartilage and epiphyseal lines via thermoacoustic imaging, while ultrasonic imaging revealed bone structures. Thermoacoustic imaging alone differentiated bone sections between normal and osteoporotic groups; however, incorporating prior skeletal contour information from ultrasound significantly enhanced discriminatory power, resulting in intergroup differences with higher statistical significance. The imaging results of bone samples corresponded well with physical photographs, micro-CT images, and theoretical analyses of dielectric properties for cartilage, normal bone, and osteoporotic bone. Conclusion The microwave-induced thermoacoustic and ultrasound dual-modality microscopy system developed in this study demonstrates potential for microscopic detection of complex biological tissues based on dielectric properties. It is expected to provide a new imaging tool for functional assessment of brain tissue and the skeletal system, as well as for studies on disease pathogenesis.

Photoacoustic imaging (PAI) has emerged as a transformative modality for bridging traditional Chinese medicine (TCM) theory and contemporary biomedical research in acupuncture mechanism studies. This review assesses PAI’s capacity to decode acupuncture-induced neuromodulatory and hemodynamic effects, with dual focus on the central nervous system (CNS) responses and acupoint-specific microcirculatory dynamics. Leveraging the photoacoustic effect coupled with ultrasonic detection, PAI enables non-invasive, high-resolution mapping of cerebral hemodynamic parameters, including blood flow, oxygen saturation and hemoglobin concentrations, in real time. Experimental evidence from murine models of cerebral hypoperfusion and ischemic stroke demonstrates acupoint-specific spatiotemporal activation patterns, particularly at Yongquan (KI1) and Yanglingquan (GB34), revealing cortical hemodynamic reorganization and angiogenesis. At the microcirculatory level, PAI identifies functional transitions from quiescent to activated vascular states during disease progression, characterized by altered perfusion dynamics and vascular permeability. While structural metrics (e.g., microvascular density and curvature) show no significant differences in knee osteoarthritis models, functional parameters such as hemoglobin flux and oxygen metabolism emerge as critical biomarkers of acupoint specificity. PAI further enhances treatment precision through standardized acupoint localization, as evidenced by electrostimulation studies at Hegu (LI4) and Zhongwan (CV12). This synthesis highlights PAI’s dual contributions: (1) validating CNS-mediated systemic regulation via acupoint-brain functional correlations, and (2) providing multimodal quantification of microcirculatory dynamics. Future directions emphasize integration of molecular probes for neuroendocrine pathway visualization and multimodal imaging to address unresolved thermal/optical interactions. By synergizing TCM principles with advanced biophotonics, PAI establishes a paradigm for mechanistic acupuncture research and clinical translation.

BACKGROUND:Sarcopenia, a significant geriatric syndrome, faces challenges in accurate diagnosis due to limitations of current imaging techniques. This study explores the novel application of multispectral optoacoustic tomography (MSOT) in evaluating sarcopenia, focusing on quantifying oxygen dynamics and collagen distribution in skeletal muscles. METHODS:We conducted MSOT imaging on the lower limbs of senescence-accelerated mouse prone 8 (SAMP8; n = 14) and senescence-accelerated mouse resistant 1 (SAMR1; n = 8) models, using light wavelengths of 760, 840 and 930 nm. CT, histopathology and immunofluorescence were used for cross-validation. RESULTS:Label-free MSOT imaging directly visualized muscle structure and metabolism with high spatiotemporal resolution. Compared to SAMR1 controls, sarcopenic SAMP8 mice demonstrated 23.8% lower HbO2 levels (SAMP8: 0.0016 ± 0.0003 a.u. vs. SAMR1: 0.0021 ± 0.0005 a.u.; p = 0.018) and reduced metabolic activity in skeletal muscles. SAMP8 mice also revealed 43.2% higher collagen content (SAMP8: 3.451 ± 1.159 a.u. vs. SAMR1: 2.409 ± 0.635 a.u.; p = 0.030) alongside more disordered muscle structure, suggesting increased fibrosis. An inverse correlation was observed between computed tomography (CT) values and MSOT-derived collagen signals (r = -0.789, p < 0.001), whereas no such correlation existed with HbO2, indicating that MSOT provides unique metabolic insights beyond traditional imaging techniques. CONCLUSIONS:This first application of MSOT in sarcopenia research highlights its potential as a noninvasive, real-time tool for early diagnosis, therapeutic evaluation and mechanistic understanding. Its ability to detect metabolic changes not captured by CT underscores its complementary role in comprehensive muscle assessment. Future research should focus on longitudinal studies and clinical translation.

Conventional microwave-induced thermoacoustic imaging (MITAI) suffers from nonuniform electric (E)-field distributions, reducing image contrast and edge-detection capability. This nonuniformity mainly arises from the radiation properties of the excitation antenna and the participation/modulation of biological tissues during microwave propagation. This work proposes an E-field normalization calibration incorporating tissue participation. Reference fields are derived from both a tumor-containing tissue model and a simplified homogeneous medium model without tumors, and normalization factors are constructed to correct the original power-loss density (PLD). A finite-element-based forward model is then established from the PLD distribution to solve for the initial acoustic pressure and acoustic signal propagation. Using detected acoustic pressure signals, MITAI images are reconstructed with the delay-and-sum algorithm for comparative analysis—covering uncalibrated, homogeneous-medium field calibrated, and tumor-inclusive field calibrated versions—and image quality under these conditions is evaluated qualitatively and quantitatively. Results show that both calibration approaches enhance image contrast and improve the delineation of tissue boundaries, while calibration employing the tumor-inclusive reference E-field yields more accurate tumor features and closer agreement with the simulation setup. In addition, a saline-tube phantom experiment further verifies the effectiveness of the proposed electric-field normalization calibration. These findings suggest that subsequent calibration methods should account for tissue-electromagnetic interactions to obtain more realistic E-field distributions within tissues. The proposed method holds considerable promise for practical application, providing a viable strategy and new perspective for high-accuracy MITAI in complex tissue environments.

Prenatal alcohol exposure (PAE) is a leading cause of developmental abnormalities, yet its effects on fetal cardiac development remain understudied. We employed real-time, label-free multispectral photoacoustic tomography (PAT) to noninvasively assess cardiac development in mouse fetuses exposed to chronic alcohol. Using a custom-built PAT system, fetal hearts were imaged from E12 to E16 in alcohol-exposed (3 g/kg ethanol via oral gavage, n  = 9) and control ( n  = 7) CD-1 mice. PAT enabled quantitative measurements of cardiac morphology, oxygen saturation (sO 2 ), and heart rate. Alcohol-exposed fetuses exhibited consistently lower sO 2 and greater heart rate variability, particularly at later gestational stages. While structural growth progressed in both groups, functional impairments became more pronounced with alcohol exposure. These findings suggest PAE alters fetal cardiovascular regulation despite normal anatomical development. This study highlights the utility of PAT as a high-resolution, noninvasive tool for monitoring fetal cardiac health and supports its potential application in developmental biology and prenatal diagnostics.

With the advancement of modernization of traditional Chinese medicine (TCM) modernization and the development of computer and image processing technologies, there is a growing demand for objective and digital tongue diagnosis. While traditional tongue examination methods can partially achieve objective analysis of tongue color and texture, they have difficulty obtaining deep vascular networks and metabolic information of the tongue. To overcome this limitation, this study proposes a novel dynamic tongue monitoring method based on photoacoustic tomography (PAT), combined with multispectral photoacoustic quantitative algorithms, enabling real-time dynamic monitoring of tongue hemodynamics. Through continuous PAT monitoring of tongue changes in diabetic rats, we confirmed significant differences in hemodynamic parameters between diabetic and normal groups, mainly characterized by decreased oxyhemoglobin (HbO), increased deoxyhemoglobin (HbR), and significantly reduced oxygen saturation (sO(2)). By analyzing changes in tongue hemodynamic parameters, this study identifies the abnormal photoacoustic characteristics of diabetic tongues, providing quantitative evidence for TCM tongue diagnosis and promoting the objective and standardized development of TCM diagnostics.