Infrared spectral quantification is challenged by high-dimensional inputs, overlapping absorption bands, nonlinear matrix effects, and limited interpretability. This study developed a SHapley Additive exPlanations (SHAP)-guided radial basis function Kolmogorov-Arnold network (RBF-KAN) framework for accurate, compact, and interpretable spectral regression. KAN was first benchmarked against representative nonlinear models on two public near-infrared olive datasets, demonstrating favorable generalization with a compact architecture. Nine KAN edge-function forms were then compared under identical conditions, and RBF-KAN achieved the strongest independent prediction performance, supporting localized nonlinear representation for complex spectra. Although the full-spectrum models were already accurate, SHAP revealed that predictive information was concentrated within limited local spectral regions. Based on the SHAP rankings, the Top-40 and Top-30 variables were retained to reconstruct the olive-fruit and olive-oil models, respectively, while the corresponding hyperparameters were optimized by five-fold cross-validation within the calibration set. The resulting models substantially reduced spectral dimensionality and further improved independent prediction performance. The framework was subsequently extended to whole-blood FTIR-ATR glucose quantification. The Top-23 model achieved a prediction-set Rp[2] of 0.9951 and an RMSEP of 0.1744 mmol/L. Bland-Altman analysis indicated low overall bias, and all 32 independent prediction samples fell within Zone A of the Clarke error grid. These results demonstrate that SHAP-guided RBF-KAN provides an accurate, compact, and interpretable strategy for infrared spectral quantification across food and biomedical matrices.

Streptococcus mutans is one of the main pathogens responsible for dental caries, necessitating an urgent search for new effective strategies against this bacterium. Hydrolysable tannins, including 1,2,3,4,6-penta-O-galloyl-β-d-glucose (PGG), 1,2-di-O-galloyl-4,6-valoneoyl-β-d-glucose (dGVG), and 2-O-bis-digalloyl-4,6-valoneoyl-β-d-glucose (b-dGVG), exhibit antibacterial activity against S. mutans but their direct molecular interactions with bacterial lipids remain unclear. In this study, we investigated the binding mechanism of these tannins to liposomes prepared from lipids isolated directly from S. mutans using a biophysical approach. Fluorescence quenching and anisotropy measurements demonstrated that the tannins used interact with lipids and preferentially form complexes localized at the hydrophilic parts of the liposomes, with binding constants (Kb) in the range of 10[3] to nearly 10[5]. Dynamic light scattering and zeta potential analyses showed that valoneoyl-containing tannins induce a shift towards more negative surface charges (e.g., down to -23.42 mV for b-dGVG) and minor aggregation. In contrast, the more flexible PGG primarily adsorbs onto the surface without significantly disrupting lipid packing. FT-IR spectra of dried lipid-tannin films, analyzed by spectral subtraction and Fourier self-deconvolution, indicated that hydrogen bonding between the tannins and lipid ester carbonyl groups is possible. Our results demonstrate that the tannins used interact with lipids, but with lower affinity than that observed for whole bacterial cells. This indicates that structural complexity of the intact bacterium, including the cell wall, surface polysaccharides, and membrane proteins, likely plays a major role in mediating its full antibacterial activity against S. mutans. These findings highlight the importance of targeting both lipid and protein membrane domains, offering a valuable molecular basis for designing novel, tannin-based anti-caries therapeutics.

Coronary artery bypass grafting (CABG) is a critical intervention for obstructive coronary artery disease, with the graft patency significantly impacting long-term outcomes. This study evaluates the potential of infrared (IR) spectroscopic imaging combined with machine learning to assess biochemical markers for determining the patency in internal mammary artery (IMA), radial artery (RA), and saphenous vein (SV) conduits. Tissue samples of the grafts from CABG patients were analysed using IR spectroscopic imaging. Hyperspectral images were pre-processed and subjected to unsupervised hierarchical cluster analysis (UHCA) to identify regions with high triglyceride content. Spectral signatures of triglycerides, proteins, nucleic acids, and carbohydrates were extracted, and ratiometric analysis revealed significant variations in glycosylation and protein-to-lipid ratios among graft types (p < 0.005). The IMA exhibited lower glycosylation and protein-to-lipid ratios compared to the RA and SV. Principal component analysis followed by linear discriminant analysis (PCA-LDA) achieved clear differentiation of graft types, with second-derivative spectra enhancing class separation. These findings highlight IR spectroscopy's potential to provide detailed biochemical insights for pre-operative graft selection, potentially improving CABG outcomes by identifying conduits with optimal patency characteristics. This preliminary study evaluates the feasibility of IR spectroscopy for assessing the quality of autologous bypass grafts prior to CABG. Further studies involving larger patient cohorts and evaluation of potential gender-related differences would help strengthen and validate these findings.

The development of multifunctional purely organic room-temperature phosphorescence materials remains a formidable challenge. Herein, a pyridine-bromobenzene-based luminogen named NNPA was successfully prepared, integrating both aggregation-induced emission (AIE) activity and pH-dependent fluorescence emission. More importantly, the fluorescence emission maxima of NNPA exhibits a monotonic red shift with increasing water content in THF-H2O mixtures, enabling precise water quantification over a broad range (0-90%). Remarkable acid-triggered fluorescence emission shifts are demonstrated, transitioning from 383 to 520 nm in solution and 390 to 465 nm in the crystalline state. Meanwhile, the ultralong phosphorescence lifetime undergoes a monotonic attenuation with increasing acid concentration, allowing for time-resolved, naked-eye sensing of acid vapors. Crucially, by incorporating Rhodamine B via melt-blending, a red-shifted room-temperature phosphorescence system reaching 655 nm was constructed through a Förster resonance energy transfer (FRET) mechanism, effectively overcoming the inherent wavelength barrier of organic afterglow. Leveraging alcohol solubility and processability of NNPA, eco-friendly flexible afterglow papers were fabricated, enabling advanced information encryption including static covert codes, dynamic time-resolved decryption, and 3D stereoscopic anti-counterfeiting. This work not only elucidates the intrinsic mechanism of multifunctional NNPA but also paves the way for next-generation smart photonic materials.

It is essential to develop a simple method for detecting formaldehyde (FA) because of its harmful effects on human health. Herein, we present a novel metal-coordinated supramolecular polymer chemosensor utilizing a complex between the tetraphenylethylene (TPE) derivative (T4J) and Cu[2+] for FA detection in both gaseous and aqueous phases. In this system, Cu(NO3)2 acts as both a linker and a promoter, enabling selective FA detection by T4J-Cu through an aggregation-induced emission (AIE) mechanism. This configuration affords a clear fluorescence turn-on response that can be recognized by the naked eye with a calculated limit of detection (LOD) of 17.5 nM. A practical FA detection kit was fabricated by immobilizing T4J-Cu onto test paper substrates, allowing field-deployable analysis of FA-contaminated environments with preserved sensitivity. Meanwhile, at the same concentration, T4J-Cu exhibited high adsorption efficiencies for 1 × 10[-2] and 1 × 10[-3] M FA, achieving 85% and 88%, respectively, making it a candidate adsorbent for FA. The proposed metal-coordinated supramolecular polymer strategy, coupled with Cu[2+]-accelerated hydrazone formation and the reaction-triggered AIE luminescence enhancement, establishes a novel paradigm for FA sensing. This methodology exhibits significant potential for expansion to diverse volatile organic compound (VOC) monitoring applications, particularly in scenarios requiring rapid on-site detection capabilities.

Developing multifunctional phosphors suitable for both anti-counterfeiting and optical thermometry applications remains challenge. Herein, a series of NaYTi2O6: Pr[3+] phosphors were synthesized, and their luminescence properties, temperature sensitivity and potential applications were systematically investigated. Under ultraviolet excitation, the phosphors exhibited prominent red emission at 614 nm ([1]D2 → [3]H4) along with weaker emissions centered at 690 nm ([3]P0 → [3]F3). Notably, red persistent luminescence (PersL) can also be observed from the NaYTi2O6: Pr[3+] phosphor. Based on thermoluminescence analysis, a feasible persistent afterglow mechanism of the phosphors was presented and discussed. Furthermore, luminescent patterns are fabricated by utilizing these phosphors as inks for anti-counterfeiting applications. In addition, the thermometric mode based on the normalized integrated intensity over 580-650 nm delivers a maximum relative sensitivity (Sr) of 2.77%K[-1]. Further FIR thermometry constructed from the Pr[3+ 1]D2 → [3]H4 and [3]P0 → [3]F3 transitions achieves a higher maximum Sr of 4.02%K[-1]. These results indicate that NaYTi2O6: Pr[3+] phosphor with persistent luminescence has potential applications in anti-counterfeiting and optical thermometry.

To investigate the influence of molecular structure on the optoelectronic properties of unsymmetrical squaraine dyes, this study conducted a systematic theoretical simulation analysis of two dye molecules, SD1 and SD3, and comprehensively evaluated their potential as photosensitizers in dye-sensitized solar cells (DSSCs). Meanwhile, using the SD3 molecule with superior optoelectronic properties as the parent molecule, we introduced conjugated π-bridges (-DTS and -DTT) and an anchoring group (cyanoacrylic acid (-CA)) to design four dye molecules (SD3-DTS, SD3-DTT, SD3-CA, and SD3-DTS-CA), and further investigated the mechanisms by which these two design strategies modulate the optoelectronic properties of the dyes. Both modification strategies effectively modulated the energy level structure, absorption spectra, electronic structure parameters, charge transport, and dye regeneration performance of the dye molecules. The light-harvesting efficiency (LHE) of all designed molecules was significantly enhanced, and SD3-CA and SD3-DTS-CA extended the light-harvesting ranges to the near-infrared region. Calculations indicate that SD3-DTS-CA exhibits excellent performance parameters, such as the smallest chemical hardness (h), electron reorganization energy (λe) and the largest LHE. Ultimately, SD3-DTS-CA demonstrated the optimal optoelectronic performance, including the short-circuit current density (JSC) of 24.218 mA cm[-2], open-circuit voltage (VOC) of 0.543 V and power conversion efficiency (PCE) of 10.712%, owing to the synergistic effect of -DTS and -CA. Furthermore, the PCEs of all designed molecules improved significantly compared with SD3, indicating that this modification strategy can reliably guide performance optimization of DSSC dye sensitizers.

Citrus Huanglongbing (HLB), caused by Candidatus Liberibacter asiaticus, threatens global citrus production, while early diagnosis is hindered by a prolonged asymptomatic period and laboratory-intensive methods. This study establishes a non-destructive framework integrating portable Raman spectroscopy, reference compound validation, optimized machine learning, and model interpretability for early HLB detection. A total of 1025 Thai pomelo leaves yielding 3075 spectra were classified into four categories: healthy, early HLB, symptomatic HLB, and nutrient deficiency. Class-specific spectral variations occurred in carotenoid, chlorophyll a, carbohydrate, and polyphenol regions. Measurements of pure reference standards supported key molecular assignments beyond literature-based interpretation. UMAP followed by K-means clustering achieved a purity of 91.72% ± 0.96%, indicating spectral separability among the four classes. For supervised classification, 72 preprocessing workflows were evaluated across six classifiers, followed by hyperparameter optimization using four search strategies. The optimized support vector machine (SVM) achieved the highest accuracy of 95.93% ± 1.42% and an F1-score of 0.9597 ± 0.0146. SHAP analysis linked classification decisions to these molecular assignments, identifying pigment-associated features as major contributors to overall classification and a cellulose-associated feature at 1096.8 cm[-1] as the leading contributor to early HLB identification. By connecting molecular assignments, classification performance, and model interpretation, this framework provides a biologically grounded approach to early HLB detection with potential for field deployment.

Acute liver injury (ALI) is linked to oxidative stress, with hydrogen peroxide H2O2 as a key mediator. We report a near-infrared fluorescent probe, PhCy-H2O2, based on an ICT mechanism. The probe shows high sensitivity (detection limit in the low micromolar range) and excellent selectivity, producing a significant fluorescence turn-on at 796 nm upon H2O2 recognition. This reaction has the advantages of high sensitivity, high selectivity, fast reaction and stability under physiological pH conditions. In both cellular and mice models, the probe clearly visualized dose-dependent H2O2 generation induced by LPS and APAP, with fluorescence signals effectively attenuated by GSH and NAC, respectively, confirming the observed signals are associated with oxidative stress. Moreover, the hepatic fluorescence intensities correlated well with the levels of MDA and TNF-α, and the signal changes faithfully reflected the protective effects of antioxidant interventions, indicating that the probe reliably reports the severity of oxidative injury in ALI. Thus, PhCy-H2O2 offers a reliable tool for studying oxidative stress in ALI and evaluating hepatoprotective agents.

Dorsal Root Entry Zoneotomy (DREZotomy) requires extreme surgical precision to eliminate hyperactive nociceptive pathways without compromising adjacent functional spinal tracts. Traditional electrothermal tools often induce unpredictable collateral thermal damage, posing significant risks of postoperative neurological deficits. In this study, we investigated the efficacy and underlying mechanisms of a 355 nm ultraviolet (UV) nanosecond laser for the highly controlled ablation of lipid-rich spinal cord nerve tissue. By systematically evaluating spatiotemporal parameters, we characterized the critical transitions between photochemical and photothermal tissue interactions. Morphological analyses demonstrated that while excessive energy densities trigger plasma-induced optical shielding, an optimal fluence window maximizes material removal efficiency. Furthermore, our temporal evaluation revealed that high-frequency pulse delivery exacerbates collateral thermal injury due to rapid heat accumulation when the inter-pulse interval falls below the tissue's inherent thermal relaxation time. Under optimized spatiotemporal conditions, Raman spectroscopy validated a predominant cold photochemical ablation regime, with the immediate residual tissue exhibiting remarkably preserved biomolecular signatures, including intact protein secondary structures and myelin lipids. Ultimately, this UV laser approach restricts the lateral zone of thermal damage to strictly less than 20 μm. These findings highlight the 355 nm nanosecond laser as a highly predictable, minimally invasive modality, offering a critical safety margin for delicate neurosurgical interventions.

Tetracycline antibiotics (TCs) pose threats to the environmental and health due to their overuse. Therefore, developing an efficient method for synchronous detecting multiple residues is crucial for guaranteeing health and safety. However, conventional sensors struggle to discriminate structurally similar TCs. In this study, a three-emission single-sensing element hydrogen-bonded organic framework composite (Eu-PFC-1@HOF) was designed to construct the fluorescent sensor array. Through the synergistic effects of inner filter effect (IFE) and photoinduced electron transfer (PET), the fluorescence signals of Eu-PFC-1@HOF were effectively quenched by TCs. Based on the different responses of various TCs to Eu- PFC-1@HOF, and incorporated with a pH-regulated strategy, a six-channel response matrix was established for multidimensional fingerprinting of TCs. Combined with linear discriminant analysis (LDA), 100% accurate identification of six TCs was achieved by the fluorescence fingerprint spectra, along with excellent concentration resolution and quantitative capability. This work integrates the multi-signal output of a single-sensing element with pattern recognition, providing an innovative strategy for high-throughput screening of trace antibiotics in complex matrices.

Phosphate ions are essential for biological and environmental processes; however, their excessive accumulation in water bodies can lead to eutrophication, and their entry into the human body through the food chain can cause metabolic disorders and other health issues. Therefore, developing a rapid, sensitive, and precise detection method is crucial for controlling drinking water pollution and monitoring human health. To address this critical issue, this paper proposes a single mode-tapered multimode-single mode (STMS) fiber sensor for trace phosphate ion detection. Experimental results show that the proposed fiber sensor exhibits good spectral stability, with a specific detection time for phosphate ions of less than 10 min. In tap water sample testing, the STMS microfiber sensor exhibits a low limit of detection (LoD) of 0.25 pM and good selectivity, with minimal cross-reactivity to common anions in water, including cl-, SO₄2-, CO₃2-, and NO₃-. Furthermore, results for the detection of phosphate ion in serum samples also demonstrated a low LoD of 0.6 pM, which is far below the normal physiological phosphate concentration range in adult serum or plasma (0.81 to 1.45 mmol/L) as defined by the National Institutes of Health. These findings suggest that the propo-sed microfiber sensor has potential for drinking water monitoring and clinical diagnostics.

The primary charge separation in Photosystem II (PSII) is governed by a complex interplay between the chromophore cofactors and the electrostatic protein matrix. While the protein environment is known to impose a functional asymmetry favoring the active branch the D1, a quantitative understanding of its dielectric influence on spectroscopic properties remains a challenge. This study employs a state-of-the-art computational approach, combining time-dependent density functional theory (TDDFT) with screened range-separated hybrid (SRSH) functionals, to investigate the excited state energies and character of a PSII reaction center. We use a simplified tetramer model as PD1PD2ChlD1ChlD2 and compare two implicit solvation methods: an isotropic dielectric model via polarizable continuum model (PCM) and a novel anisotropic model PCM(r) via Poisson equation solver (PEqS) that accounts for the heterogeneous dielectric microenvironments of the D1 and D2 branches. Our results demonstrate that the isotropic model captures general environmental screening trends but fails to reproduce the correct energetic ordering of key chromophores, predicting a blue shift for ChlD1 that contradicts the experimental red shift. In contrast, the anisotropic SRSH-PCM(r) model, with a higher dielectric constant assigned to the D1 branch (ɛ1= 3.5-5.0) and a lower one for the D2 branch (ɛ2= 1.5), not only successfully yields bright state energies with correct ordering and their localization but also an electrochromic shift (deviated only 0.03 eV) in qualitative agreement with experimental absorption data. This work establishes that dielectric anisotropy is not merely a minor correction but a critical factor for accurately modeling the spectroscopic properties of PSII, providing a more realistic framework for understanding the role of the protein environment.

Early and reliable assessment of neonatal conditions is important in neonatal intensive care units, where non-invasive imaging approaches may provide complementary information for clinical evaluation. Hyperspectral imaging provides a non-invasive means of capturing spatially resolved spectral information from neonatal tissue; however, its high dimensionality and the limited availability of clinical samples make robust model development challenging. This study proposes the Multiscale Contextual Involution-Inception Network (MC-I[2]Net), a leakage-controlled framework for neonatal HSI classification. The model combines spatially adaptive involution operators, Inception-based multiscale feature extraction, and contextual feature fusion to learn compact and discriminative spectral-spatial representations. A balanced six-class neonatal hyperspectral imaging dataset containing 240 samples was evaluated, including coarctation of the aorta, intracranial hemorrhage, necrotizing enterocolitis, pneumothorax, respiratory distress syndrome, and healthy control groups. Fixed signal-conditioning operations were applied using predefined processing rules, whereas PCA was treated as a data-adaptive transformation. PCA was fitted exclusively on each training fold, and the learned transformation was subsequently applied to the corresponding held-out test fold. All models were assessed using the same 5-fold cross-validation protocol repeated 10 times. MC-I[2]Net achieved an overall accuracy of 94.38 ± 4.60% and a Macro-F1 score of 94.35 ± 4.57%, with only small numerical differences in OA relative to MC-3DCNN, ViT-HSI, and Swin-HSI. The model used fewer trainable parameters than MC-3DCNN, ViT-HSI, and Swin-HSI and substantially fewer FLOPs than MC-3DCNN, supporting a favorable performance-complexity profile rather than a large predictive-performance margin. An additional anatomically controlled analysis restricted to PTX and RDS samples acquired from the same thoracic region retained an OA of 95.75 ± 6.49% and a Macro-F1 score of 95.66 ± 6.68%. Sequential nested cross-validation, with PCA dimensionality and involution hyperparameters selected using training data only, yielded a pooled outer out-of-fold OA of 95.00% and a Macro-F1 score of 94.98%. Ablation, sensitivity, calibration, nested-validation, runtime, and saliency analyses provided complementary evidence regarding architectural behavior, robustness, computational characteristics, and preliminary explainability. Overall, these findings support MC-I[2]Net as a proof-of-concept framework for neonatal HSI classification and provide a methodological basis for future prospective, multicenter, and cross-device evaluation.

Room temperature phosphorescent carbon dots (RTP CDs) have attracted significant attention due to their long-lived emission and potential applications in security and forensic science. In this work, zinc-doped carbon dots (CDs@ZnCl2) were synthesized via a simple thermal method, and a boron oxide matrix was introduced to construct a rigid environment that stabilizes triplet excitons and enhances phosphorescence emission. The obtained CDs@ZnCl2-BO composite exhibits strong fluorescence and long-lived room temperature phosphorescence with good environmental stability. The rigid matrix effectively suppresses non-radiative transitions and protects the triplet excitons from oxygen and water quenching, resulting in enhanced afterglow emission. Due to their dual-mode luminescence properties, the prepared materials were successfully applied in latent fingerprint imaging, where clear ridge details were observed with high contrast under phosphorescence mode after UV light removal. In addition, the materials were used as security ink for anti-counterfeiting and information encryption, demonstrating multi-level security patterns visible under UV light and afterglow emission. This work provides a simple strategy for designing RTP carbon dot materials for forensic detection and advanced anti-counterfeiting applications.