Journal of Biophotonics最新文献

Accurate, label-free, non-destructive discrimination between head and neck squamous cell carcinoma (HNSCC) and keratoacanthoma (KA) remains challenging due to their overlapping morphology. We introduce a real-time, end-to-end hyperspectral imaging (HSI) workflow applied to 80 formalin-fixed, paraffin-embedded sections, each sampled with five 50 × 50-pixel ROIs and captured across 400–1000 nm to produce 128-band reflectance cubes. After reflectance calibration, Savitzky–Golay smoothing, and first-derivative preprocessing, a compact one-dimensional convolutional neural network achieved 87% accuracy, 93% sensitivity, 77% specificity, and AUC = 0.85 on a held-out test set. Spectral difference analysis revealed key biomarkers at the hemoglobin Q-band (630 nm) and OH overtone (917.5 nm), corresponding to vascular and extracellular matrix variations. This resource-efficient photonic platform enables rapid, automated “optical biopsy” without exogenous stains, offering scalable adjunctive diagnostics and a clear pathway toward intraoperative tissue classification.

This study validated the feasibility of visible spectroscopy in rapidly detecting Urinary Microalbumin (UALB). Based on 127 clinical urine samples, spectra ranging from 400 to 750 nm were collected using a microspectrometer. The successive projections algorithm (SPA) was used to screen for nine wavelengths highly correlated with UALB, and the spectral index (SI) method was fused to construct an m-SPA-SI strategy. The m-SPA-SI was combined with a random forest (RF) model for discriminant analysis. The results showed that the m-SPA-SI₆-RF model for low-dimensional spectra exhibited the best performance, with a training accuracy of 100% and improved testing accuracy and sensitivity of 97.37% and 0.9333, respectively. The number of wavelengths was reduced from 457 to 9. Research has confirmed that the combination of low-dimensional spectroscopy and a wavelength selection algorithm can achieve efficient discrimination of UALB, providing a new method for portable screening of diseases such as chronic kidney disease.

Atopic dermatitis (AD), a chronic inflammatory skin disorder, faces diagnostic limitations due to subjective clinical assessments. This study introduces an integrated photoacoustic imaging (PAI)/optical coherence tomography (OCT) system for noninvasive, quantitative evaluation of MC903-induced AD-like dermatitis in mice and Crisaborole efficacy. Multimodal imaging demonstrated Crisaborole's suppression of pathological angiogenesis (50.7% fewer vascular junctions, 23% reduced density, 47% shorter vessel length) and tissue remodeling (20% epidermal thinning, 23% signal decline), validated by histopathology. The system captured in vivo dynamic vascular network restructuring and epidermal alterations during AD progression, revealing temporal links between angiogenesis, hyperplasia, and inflammation. PAI/OCT synergistically provides complementary functional-structural biomarkers, advancing precision diagnosis and therapy in AD models. This work pioneers real-time monitoring of AD pathophysiology, supporting clinical translation through quantitative multiparametric characterization for tailored therapeutic strategies.

Given the demand for ablation treatment using picosecond lasers in medical cosmetology, researchers analyze the degree of thermal damage to skin tissue by combining theoretical modeling with experimental research. The effects of laser scanning speed, number of scans (30–90) and scanning paths on the heat-affected zone, ablation depth, and thermal damage were investigated using one-way analysis of variance. The results show that increasing the scanning speed can reduce the heat-affected zone; the number of repeated scans is positively correlated with the ablation depth; and the concentric circle scanning path can reduce the accumulation of thermal damage compared with linear scanning. On this basis, the optimal combination of process parameters is obtained, with good macroscopic morphology and lower thermal damage than continuous lasers.

Keloid is a common benign fibroproliferative skin disorder impairing appearance and life quality. Currently, the diagnosis of keloids primarily relies on clinicians' clinical experience, but the extent of their infiltration is difficult to accurately determine by visual inspection alone. Keloid formation is often associated with excessive fibroblast proliferation, during which cells exhibit bioenergetic patterns resembling those of cancer cells, primarily relying on glycolysis for ATP production. This study investigated the metabolic differences between keloid and normal skin tissues. Using FLIM, we analyzed the autofluorescence of the endogenous fluorophores NAD(P)H and FAD to evaluate the metabolic states of skin tissues. The FLIRR was calculated to quantify these metabolic changes. These metabolic discrepancies can provide crucial guidance for the diagnosis of keloids. The combination of FLIM and FLIRR enables noninvasive differentiation between keloid and normal tissues, offering the potential for more accurate diagnosis and treatment.

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 (sO₂), and heart rate. Alcohol-exposed fetuses exhibited consistently lower sO₂ 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.

Handgrip pressure is a crucial biomechanical indicator, reflecting limb strength and vertebral bone fracture risk. This study introduces a non-invasive Smart Grip strength measurement system utilizing a Fiber Bragg Grating (FBG) sensor. The device quantifies grip pressure exerted by volunteers in various postures through FBG wavelength shift analysis. Demonstrating a sensitivity of 1.32 pm/με, the system measures grip pressures ranging from 6 to 131 kPa across different hand positions. Tested on 20 volunteers aged 18–25, the maximum recorded grip pressure is 131.9 kPa in the arm-dangled position. The FBG-based Smart Grip system provides a precise and efficient method for handgrip strength assessment, valuable for clinical rehabilitation. The novelty of this work lies in the integration of a single FBG-based Smart Grip system for dual applications such as (1) real-time posture-dependent grip strength monitoring and (2) progressive assessment of bone fracture healing.

We investigated changes in the pulse waveform recorded by photoplethysmography (PPG) during a local heating test in patients with diabetes. The following groups of subjects were studied: healthy individuals (n = 15), patients with Type 2 diabetes accompanied by diabetic retinopathy (n = 14), patients with Type 2 diabetes and diabetic foot syndrome (n = 16). Measurements were taken from the big toes of both limbs while the skin was heated from 32°C to 42°C. We found statistically significant changes in PPG waveform after heating between all three groups. The main change is that the dicrotic notch becomes more pronounced during the heating period. This effect is strongest in Group 1, slightly less pronounced in Group 2, and practically absent in Group 3. Thus, monitoring of the PPG waveform during the heating test may be an effective means of assessing microvascular lesions in diabetes.

A three-dimensional convolutional neural network (3D-CNN) was developed for the analysis of volumetric optical coherence tomography (OCT) images to enhance endoscopic guidance during percutaneous nephrostomy. The model was performance-benchmarked using a 10-fold nested cross-validation procedure and achieved an average test accuracy of 90.57% across a dataset of 10 porcine kidneys. This performance significantly exceeded that of 2D-CNN models that attained average test accuracies ranging from 85.63% to 88.22% using 1, 10, or 100 radial sections extracted from the 3D OCT volumes. The 3D-CNN (~12 million parameters) was benchmarked against three state-of-the-art volumetric architectures: the 3D Vision Transformer (3D-ViT, ~45 million parameters), 3D-DenseNet121 (~12 million parameters), and the Multi-plane and Multi-slice Transformer (M3T, ~29 million parameters). While these models achieved comparable inferencing accuracy, the 3D-CNN exhibited lower inference latency (33 ms) than 3D-ViT (86 ms), 3D-DenseNet121 (58 ms), and M3T (93 ms), representing a critical advantage for real-time surgical guidance applications. These results demonstrate the 3D-CNN's capability as a powerful and practical tool for computer-aided diagnosis in OCT-guided surgical interventions.

Microscopic imaging techniques pursue high-resolution, large depth of field (DoF) imaging but are limited by hardware, especially the strong focusing of objective lenses. Optical-resolution photoacoustic microscopy (OR-PAM) has a narrow DoF due to the intense laser focusing needed for high-resolution imaging. To address this, we propose a novel volumetric information fusion method using a three-dimensional siamese multi-level features convolutional neural network (3DSMFCNN) for cost-effective, large-DoF imaging. Initially, an initial decision map (IDM) is produced by performing focus region identification on multi-focus 3D photoacoustic data with the pre-trained 3DSMFCNN. The IDM is then refined through consistency verification and Gaussian filtering to generate the final decision map (FDM). A DoF-enhanced photoacoustic image is obtained by voxel-weighted averaging based on the FDM. Experiments with multi-focus 3D simulated fibers, blood vessels, and real data demonstrate that the method significantly extends the DoF of OR-PAM without sacrificing lateral resolution, which confirms its effectiveness, robustness, and applicability.