Journal of biophotonics最新文献

Vulvar intraepithelial neoplasia (VIN) is increasing in prevalence, yet screening options remain limited and existing diagnostic methods show low accuracy. This study evaluates infrared thermal imaging as an alternative screening approach for VIN detection. We analyzed thermal images from 51 patients with histopathologically confirmed VIN, captured using a FLIR A400 thermal camera. Temperature distributions of healthy vulvar tissue were first characterized to establish baseline values. Thermal features of VIN lesions were then extracted and optimized using principal component analysis (PCA). Three machine learning models-support vector machine (SVM), random forest (RF), and linear discriminant analysis (LDA)-were trained and evaluated for VIN diagnosis. SVM demonstrated the best performance with an F1 score of 75% and accuracy of 74.19%. These findings suggest that machine learning-based infrared thermography shows promise as a non-invasive screening tool for VIN detection.

Fibrosis is a pathological scarring process that disrupts tissue architecture, and is characterized by excessive extracellular matrix (ECM) deposition, leading to tissue stiffening and impaired organ function. Accurate quantification and spatial mapping of fibrotic tissue mechanics are critical for diagnosis, monitoring disease progression, and evaluating therapeutic responses. Here, we employ Brillouin microspectroscopy, a non-invasive, label-free optical technique, to quantify the mechanical properties of human fibrotic tissue in in situ. We show that Brillouin spectroscopy distinguishes fibrotic tissue from healthy tissue on the basis of localized differences in the complex longitudinal modulus and enables real-time monitoring of dynamic alterations in viscoelastic properties during fibrogenesis. To our knowledge, this is the first demonstration of Brillouin spectroscopy for in situ characterization of fibrosis and wound healing in a human model. These findings underscore Brillouin microspectroscopy's potential application as a promising diagnostic and monitoring tool for fibrotic diseases.

This study identifies fluorescence excitation-emission pairs correlated with atherosclerotic pathology in an ex vivo human aorta. Fluorescence spectroscopy, wide-field fluorescence imaging, gross pathologic and histologic evaluation of ex vivo cadaveric human aorta are performed. A matrix of Pearson correlation coefficients are determined for the relationship between relevant histologic features and fluorescence intensity for 427 excitation-emission pairs. A multiple linear regression analysis indicates that tryptophan and elastin fluorescence intensity predicts 58% of the variance in intima thickness (R-squared = 0.588, F(2, 18) = 12.8, p = 0.0003), and 48% of the variance in media thickness (R-squared = 0.483, F(2, 18) = 8.42, p = 0.002). Excluding lesions identified as necrotic lipid cores on histology from analysis, a combination of tyrosine, tryptophan, collagen and elastin autofluorescence predicted 86.0% of the variance in intima thickness (R-squared = 0.8598, F(4, 13) = 19.9, p = 1.87 × 10^-5) and 51.8% of the variance in media thickness (R-squared = 0.518, F(4, 13) = 3.49, p = 0.0382).

This study employed Spatial Frequency Domain Imaging (SFDI) at a wavelength of 530 nm and a spatial frequency of 0.4 mm^-1 to measure the reduced scattering coefficient (μ_s') of artificially demineralized dental tissues across various demineralization durations. Scanning electron microscope and polarized light microscope were utilized to detect surface morphology and demineralization depth. Results show that μ_s' of dental tissue increases nonlinearly with demineralization time. μ_s' of human dental tissue increased from 0.403 to 0.897 mm^-1 in 72 h, while that of bovine dental tissue increased from 0.604 to 1.420 mm^-1. The demineralization depth of bovine dental tissue increased from 0 to 169.83 μm. Artificial caries models with various demineralization durations showed a positive nonlinear correlation between μ_s' and depth of demineralization, with a correlation coefficient of 0.9673. It demonstrates SFDI's capability for nondestructive detection of early caries.

Accurate identification of driver mutations such as ALK, EGFR, and KRAS in lung adenocarcinoma is essential for guiding personalized therapies, yet standard genomic assays are invasive and may alter tissue integrity. In this study, we introduce a non-destructive genotyping approach that combines visible-to-near-infrared hyperspectral imaging (400-1000 nm) of unstained pathological sections with a dual-branch deep-learning fusion framework and gradient-boosting classification. The imaging system captures rich spectral-spatial signatures, which are processed by a fusion network that synergistically extracts global contextual features and local textural details. These fused representations are then classified by an optimized XGBoost model. Evaluation on 90 clinical specimens yielded class-specific accuracies between 83.5% and 90.2%, and area under the ROC curve values from 0.83 to 0.91. Our results demonstrate that hyperspectral imaging coupled with deep-learning fusion enables rapid, tumor genotyping, offering a promising tool for real-time clinical diagnostics in the field of biomedical photonics.

Multispectral photoacoustic microscopy (PAM) using stimulated Raman scattering (SRS) has been employed to measure oxygen saturation (sO₂) in biological tissue. However, laser-scanning photoacoustic microscopy (LS-PAM) inherently suffers from low detection sensitivity due to the use of a flat transducer and non-coaxial alignment of the transducer with the optical scan. Although wide-field-of-view LS-PAM has been implemented, it typically results in coarser lateral resolution and hence lower sensitivity than existing LS-PAM systems. Here, we present a wide-field multispectral LS-PAM system for measuring sO₂ in biological tissue. Instead of relying on two discrete wavelengths, our method employs two wavelength groups-a isosbestic group (532 nm and 545 nm) and a deoxyhemoglobin-dominant group (545 nm and 558 nm). We demonstrate that using these groups improves the signal-to-noise ratio (SNR) of the detected signals, leading to more accurate sO₂ measurements. The performance of this system is validated through both phantom and in vivo studies.

When illuminated with green light, tissue shows negligible autofluorescence in comparison to urinary stones. In automatically controlled lithotripsy, this property is utilized to prevent the laser from being triggered if the fiber is mispositioned: the fluorescence signal is compared to a set threshold before each pulse. However, previous studies have shown that tissue damage cannot be completely ruled out. We are investigating this phenomenon and its impact on fluorescence guidance. An experiment with porcine calyx (with the automatic control switched off) shows that single Ho:YAG laser pulses are sufficient to coagulate tissue, resulting in an increase in autofluorescence. During lithotripsy of fluorescent artificial stones embedded in renal cortex, thermal damage occurs despite automatic laser control. Maximum fluorescence values measured on those tissue places were above the control's set threshold for laser emission. Therefore, an increase in autofluorescence in the event of denaturation must be considered when using automatically controlled lithotripsy.

Alzheimer's disease (AD) and vascular dementia (VaD) are two common forms of dementia. Differentiating between them is challenging due to the lack of clear clinical and auxiliary test differences. In this study, we developed a novel diagnostic method combining near-infrared spectroscopy with a convolutional neural network and self-attention mechanism (CNN-SAM). The CNN-SAM model, which integrates the self-attention mechanism to highlight important spectral features, outperformed other models with 99.3% accuracy. Data pre-processing, feature extraction, and parameter optimization further enhanced the model's performance. Visualization using the self-attention mechanism revealed key spectral bands at 1364 and 1484 nm as crucial for distinguishing AD and VaD. This approach offers a rapid, non-invasive, and accurate method for the diagnosis of AD and VaD, potentially advancing clinical practice.

In this study, we present our previously fabricated star-shaped magnetic-plasmonic Au@Fe₃O₄ nanostars as promising agents for photothermal therapy. The nanostars exhibit a photothermal conversion efficiency of ~60% at a concentration of 25 μg/mL under 808 nm laser irradiation. WST-1 analysis revealed that Au@Fe₃O₄ nanostars moderately reduced the viability of human hepatocarcinoma Huh7 cells after 24 h exposure at concentrations of 1, 5, and 10 μg/mL, accompanied by notable morphological alterations. Flow cytometry demonstrated that treatment with 5 μg/mL nanostars followed by 20 min of laser irradiation resulted in 79% elimination of cancer cells. Furthermore, photothermal therapy increased cellular granularity, with highly granulated cells comprising 23% of the population compared to 4% in untreated controls. The viability of these highly granulated cells decreased to 17% post-treatment. Interestingly, photothermal therapy reduced the proportion of apoptotic cells among Huh7 subpopulations relative to the overall number of dead cells.