Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy最新文献

{"title":"Transfer of silver nanoparticles (AgNPs) to dichloromethane for single-step and in-situ synthesis of PDMS-AgNPs nanocomposites for SERS analysis","authors":"Valentina Aguilar-Melo ,&nbsp;Rodrigo Jesús Rojas-Carenzo ,&nbsp;Ildefonso Esteban Pech-Pech ,&nbsp;Ángel Adrián Bacelis-Jiménez ,&nbsp;Alejandro Ávila-Ortega ,&nbsp;William Alejandro Talavera-Pech","doi":"10.1016/j.saa.2026.127524","DOIUrl":"10.1016/j.saa.2026.127524","url":null,"abstract":"<div><div>Silver nanoparticles (AgNPs) deposited in polydimethylsiloxane (PDMS) films has recently received significant attention for the fabrication of SERS substrates, but, conventional deposition of AgNPs on PDMS requires complex, additional post-synthesis steps. This work presents a novel, single-step, and in-situ synthesis of flexible PDMS films with incorporated AgNPs for Raman-SERS applications. The method involves an AgNPs phase transfer from aqueous colloid to a dichloromethane (DCM) organic phase, which is then used as the solvent during PDMS synthesis. UV–Vis spectroscopy results show a maximum transference using a minimum volume of DCM and centrifugation of the colloid. Given its historical and cultural relevance as a colorant in diverse artefacts and ancient codices made of cellulose paper, the common dye, annatto, impregnated in paper, was used as an analyte model to evaluate the SERS performance of PDMS-AgNPs films in Raman spectroscopy, using 785 and 532 nm lasers, along with different concentrations of aqueous annatto extracts (0.01, 0.1, 0.125, 0.25, 0.5, and 1% <em>w</em>/<em>V</em>). The results demonstrated a significant enhancement in the intensity of the characteristic peaks of bixin (the main component of annatto), especially in extracts of low concentration (0.125 and 0.25% <em>w</em>/<em>V</em>) when the PDMS-AgNPs nanocomposite films synthesized with the AgNPs transferred using a minimum volume of DCM and centrifugation were used.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"352 ","pages":"Article 127524"},"PeriodicalIF":4.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146076491","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting water quality parameters using proximal spectral sensing technology and adaptive ensemble regression","authors":"Yubo Zhao ,&nbsp;Jie Zhan ,&nbsp;Jinling Chen ,&nbsp;Zeyuan Zhang ,&nbsp;Yitong Wu ,&nbsp;Tao Yu","doi":"10.1016/j.saa.2026.127554","DOIUrl":"10.1016/j.saa.2026.127554","url":null,"abstract":"<div><div>To address the limitations of conventional and remote-sensing-based water quality monitoring, this study developed a proximal spectral sensing system incorporating 12 configurable spectral bands and coupled it with an adaptive ensemble regression framework to achieve continuous, non-contact retrieval of turbidity (Tur), chlorophyll-a (Chla), chemical oxygen demand (COD), and dissolved oxygen (DO). The proposed approach integrated an abnormal spectral curves removal framework, feature augmentation, and Recursive Feature Elimination with Cross-Validation (RFECV), and constructed a dynamic model pool consisting of heterogeneous base learners, including Random Forest Regression (RFR), Gradient Boosting Regression (GBR), Gaussian Process Regression (GPR), and k-Nearest Neighbor Regression (KNNR), together with two ensemble strategies, namely weighted averaging and stacking. The results demonstrated that the optimal models achieve high predictive accuracy on the test set, with coefficients of determination (<span><math><msup><mi>R</mi><mn>2</mn></msup></math></span>) of 0.988 for Tur, 0.814 for Chla, 0.882 for COD, and 0.833 for DO. Except for COD, the mean absolute percentage errors (MAPE) of the remaining three parameters were all below 10%. Ensemble strategies showed clear advantages for non-optically active parameters, whereas well-tuned tree-based base learners performed better for optically active parameters. In addition, the models maintained strong robustness against Gaussian noise and performed stably with limited training data, effectively tracking both sudden fluctuations and long-term trends in water quality. Overall, this study provides a stable and low-cost solution for high-frequency, continuous surface water quality monitoring.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"352 ","pages":"Article 127554"},"PeriodicalIF":4.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146159638","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Label-free serum SERS combined with RFE-GBDT algorithm for non-invasive screening of liver cancer","authors":"Jingjing Gao ,&nbsp;Tianyi Lv ,&nbsp;Xianqiong Gong ,&nbsp;Xingen Gao ,&nbsp;Wei Qiao ,&nbsp;Fuqiang Wang ,&nbsp;Junzheng Wu ,&nbsp;Juqiang Lin","doi":"10.1016/j.saa.2026.127552","DOIUrl":"10.1016/j.saa.2026.127552","url":null,"abstract":"<div><div>Early diagnosis of liver cancer is crucial for developing clinical treatment strategies and improving patient survival rates. However, current diagnostic methods are often invasive, complex, and time-consuming, making them unsuitable for early screening in practical settings. Therefore, there is an urgent need to develop efficient and convenient non-invasive diagnostic techniques. This study presents a non-invasive optical diagnostic approach based on surface-enhanced Raman spectroscopy (SERS) and a deep learning algorithm for liver cancer staging identification and auxiliary screening. We systematically collected high-quality SERS spectral data from serum samples of patients with different stages of liver cancer (T1, T2, T3), hepatitis B (HBV), and healthy controls (Normal). Recursive feature elimination (RFE) was employed for feature selection, eliminating redundant spectral bands and retaining features highly relevant to classification, which significantly enhanced the model's discriminative ability. The selected features were input then into a gradient boosting decision tree (GBDT) model. Through residual iterative optimization, the model effectively captured nonlinear feature interactions, and key spectral bands were interpreted using the local interpretable model-agnostic explanations (LIME) algorithm. Compared to other commonly used classifiers such as logistic regression (LR) and random forest (RF), the RFE-GBDT model demonstrated superior performance in liver cancer staging tasks, achieving an accuracy of 92.68% in the five-class classification. The results indicate that the integration of SERS technology with the RFE-GBDT algorithm holds promise as an efficient and non-invasive auxiliary tool for the early diagnosis of liver cancer.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"352 ","pages":"Article 127552"},"PeriodicalIF":4.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146151597","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ag nanoclusters as nanozyme for detection of Hg2+ based on aggregation enhanced peroxidase-like activity","authors":"Cuifeng Jiang,&nbsp;Yong Zhu,&nbsp;Wanting Han","doi":"10.1016/j.saa.2026.127562","DOIUrl":"10.1016/j.saa.2026.127562","url":null,"abstract":"<div><div>In this study, lysine protected silver nanoclusters (Lys-AgNCs) were prepared and the mimetic property was systematically investigated. The nanoclusters displayed oxidase-like and peroxidase-like activity for the catalytic oxidation of 3, 3′, 5, 5′-tetramethylbenzidine (TMB). Interestingly, the nanozyme property could be tuned by adjusting the pH or temperature. In the low pH (3.6–4.2), oxidase-like activity is predominant. In the high pH(4.2–5.4), peroxidase- like activity is the main activity. In the lower temperature (20–40 °C), oxidase-like activity is predominant. Importantly, the Lys-AgNCs showed stronger affinity for TMB. The activation energy of Lys-AgNCs was lower than bare Ag nanoclusters, indicating a faster reaction rate. In addition, a detection method for Hg²⁺ was designed based on the aggregation enhanced peroxidase-like activity of lysine-Ag nanocluster with limit of detection (LOD) of 0.004 μM. The sensing assay exhibited excellent selectivity owing to the specific interaction between lysine and Hg²⁺ and can be used in tap water with an acceptable recovery rate. Considering the cost-effective of this method, it is proposed to be used in practical application.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"352 ","pages":"Article 127562"},"PeriodicalIF":4.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146151653","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Accurate classification of ependymomas and medulloblastomas using Raman spectroscopy and pilot transcriptomic profiling","authors":"Yuki Kawamoto ,&nbsp;Yoshiko Okita ,&nbsp;Kenta Temma ,&nbsp;Toshiki Kubo ,&nbsp;Yasuaki Kumamoto ,&nbsp;Yasuhiro Fujii ,&nbsp;Keisuke Nishimoto ,&nbsp;Reina Utsugi ,&nbsp;Chisato Yokota ,&nbsp;Ryuichi Hirayama ,&nbsp;Noriyuki Kijima ,&nbsp;Khoo Hui Ming ,&nbsp;Naoki Tani ,&nbsp;Satoru Oshino ,&nbsp;Naoki Kagawa ,&nbsp;Daisuke Motooka ,&nbsp;Katsumasa Fujita ,&nbsp;Haruhiko Kishima","doi":"10.1016/j.saa.2026.127532","DOIUrl":"10.1016/j.saa.2026.127532","url":null,"abstract":"<div><div>Raman spectroscopy enabled accurate discrimination of posterior fossa ependymomas and medulloblastomas in both frozen and formalin-fixed, paraffin-embedded (FFPE) specimens in this retrospective study. We acquired Raman spectra (532 nm excitation) from frozen and FFPE tissues to evaluate a principal component analysis–support vector machine classifier by using fivefold cross-validation. We also performed a pilot spatial transcriptomics analysis on three FFPE sections by using the 10× Genomics Xenium In Situ v2 FFPE workflow. In total, 34 specimens (21 frozen and 13 FFPE) were analyzed, and the classification models achieved &gt;90% accuracy in distinguishing ependymomas from medulloblastomas under spectrum-level fivefold cross-validation, suggesting discriminative biochemical differences, whereas patient-level performance requires further validation in larger cohorts. Ependymomas had higher lipid-associated Raman bands (1084, 1128, and 1654 cm⁻¹), whereas medulloblastomas exhibited higher deoxyhemoglobin-related Raman bands (1356, 1548, and 1604 cm⁻¹). Compared with normal controls, tumor tissues had increased carotenoid-related Raman bands and reduced lipid-associated Raman bands. The observed spectral differences are consistent with differences in lipid-associated composition and the heme/oxygenation-related tissue context, and pilot transcriptomic profiling provided qualitative biological context related to lipid metabolism and angiogenesis. These findings support Raman spectroscopy as a label-free spectroscopic technique that may complement conventional diagnostics and aid surgical and therapeutic decision-making. Larger, prospective studies are warranted to further evaluate the clinical generalizability and intraoperative translation of our results.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"352 ","pages":"Article 127532"},"PeriodicalIF":4.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146138222","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Application of FTIR and PCA-LR metabolites recognition for bergamot essential oil authentication","authors":"Laura Manin ,&nbsp;Giuseppe Oliva ,&nbsp;Maria Giovanna Bianco ,&nbsp;Md. Maruf Hossain ,&nbsp;Srećko Valić ,&nbsp;Syed K. Islam ,&nbsp;Filippo Laganà ,&nbsp;Antonino S. Fiorillo ,&nbsp;Salvatore A. Pullano","doi":"10.1016/j.saa.2026.127561","DOIUrl":"10.1016/j.saa.2026.127561","url":null,"abstract":"<div><div>A rapid method using FTIR-ATR spectroscopy combined with PCA and a classification model was applied to distinguish between natural, synthetic, and adulterated Bergamot essential oil (BEO). Synthetic BEOs are often composed of specific alcohols such as ethanol and dipropylene glycol (DPG), which are used to dilute synthetic metabolites like limonene, linalyl acetate, and linalool. Synthetic BEOs exhibited a distinct peak at 1340 cm⁻¹, linked to C<img>H bending of alcohols or methyl group deformation in artificial esters like linalyl acetate, a peak that is absent in natural BEOs. Additionally, an absorption band between 3600 and 3100 cm⁻¹ indicated the presence of DPG and synthetic ethanol, a byproduct of synthetic linalyl acetate. These findings were validated by comparison with NMR spectroscopy for metabolite recognition. A logistic regression (LR) model using PCA was applied to 369 samples, achieving an overall accuracy of 0.976 ± 0.016 through five-fold cross-validation (CV).</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"352 ","pages":"Article 127561"},"PeriodicalIF":4.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146159668","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mercaptophenylboronic acid-modified double-layer PEI-Au magnetic probes: an ultrasensitive SERS biosensor for broad-spectrum bacterial capture and specific detection in clinical urine samples","authors":"Siyun Cheng ,&nbsp;Jin Chen ,&nbsp;Xiaodan Cheng ,&nbsp;Shihui Zhu ,&nbsp;Han Shen ,&nbsp;Ping Yang ,&nbsp;Rui Xiao","doi":"10.1016/j.saa.2026.127544","DOIUrl":"10.1016/j.saa.2026.127544","url":null,"abstract":"<div><div>Accurate identification of uropathogens is critical for guiding effective antibiotic treatment. In this study, we developed a surface-enhanced Raman spectroscopy (SERS) biosensor for rapid and sensitive detection of urinary pathogens. The system combines 4-mercaptophenylboronic acid (MPBA)-modified magnetic gold-shell beads (MSAuS@MPBA) for broad-spectrum bacterial capture with 5,5′-dithiobis(2-nitrobenzoic acid) (DTNB)- and antibody-functionalized gold nanoflowers (AuNFs@Ab) as SERS tags. A double-layer polyethyleneimine (PEI)-assisted 3 nm gold seed design was introduced to fabricate hierarchical rough surfaces on MSAuS@MPBA, significantly enhancing SERS signal generation through localized electromagnetic hotspots. The MSAuS@MPBA achieved capture efficiencies exceeding 87.52% for <em>Escherichia coli</em>, <em>Pseudomonas aeruginosa</em>, and <em>Staphylococcus aureus</em> (<em>S. aureus)</em>. Using <em>S. aureus</em> as a model pathogen, the sandwich-type biosensor (MSAuS@MPBA/bacteria/AuNFs@Ab) exhibited a limit of detection (LOD) of 7 CFU mL⁻¹, a wide linear range (10¹–10⁶ CFU mL⁻¹, R² = 0.991), excellent specificity against five interfering urinary tract infection (UTI) bacteria, repeatability (RSD &lt; 6.25%), and robust anti-matrix interference in urine and blood. Clinical validation using urine from UTI patients demonstrated high recovery rates (89.11–116.32%), confirming the sensor's practical applicability. This streamlined (∼50 min), ultrasensitive platform provides a clinically translatable method for point-of-care detection of urinary pathogens.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"352 ","pages":"Article 127544"},"PeriodicalIF":4.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146159705","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Label-free Raman spectroscopy combined with artificial intelligence for functional subtyping of human sperm and prediction of embryo development outcomes","authors":"Boying Gao ,&nbsp;Ning Zhang ,&nbsp;Dexi Bi ,&nbsp;Haijing Zhao ,&nbsp;Yuqin Zhu ,&nbsp;Wenfeng Zeng ,&nbsp;Yan Nie ,&nbsp;Jianing Chen ,&nbsp;Nengyong Ouyang","doi":"10.1016/j.saa.2026.127498","DOIUrl":"10.1016/j.saa.2026.127498","url":null,"abstract":"<div><div>Raman spectroscopy delivers label-free, non-destructive, single-cell measurements and directly senses intrinsic biochemical signals of nucleic acids, proteins, and lipids. Conventional semen analysis offers limited prognostic power for assisted reproductive technology (ART). We developed a single-cell, label-free Raman workflow integrated with artificial intelligence (AI) to assess sperm functional status and predict outcomes of ART. Spectra from sperm of thirty-one ART patients were processed to extract diagnostically informative bands spanning nucleic acids, proteins, lipids, and glycogen. Spectral analysis resolved two robust spectral subpopulations. The subpopulation exhibiting higher chromatin integrity, more favorable protein conformation, and better membrane lipid dynamics molecular features was associated with better outcomes, higher two-pronuclear (2PN) fertilization rate (76.08% vs 50.17%, <em>P</em> = 0.044), high-quality embryo rate (40.53% vs 17.57%, <em>P</em> = 0.010), and blastocyst formation (59% vs 9%, <em>P</em> &lt; 0.001). Among evaluated classifiers, a gated recurrent unit (GRU) model showed the best predictive performance (accuracy 94%, sensitivity 94%, AUC 0.98). This Raman–AI assay enables objective, non-invasive stratification of sperm quality and clinically relevant prognosis, offering an analytically rigorous tool to guide precise sperm selection in ART.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"352 ","pages":"Article 127498"},"PeriodicalIF":4.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146076350","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Chitosan–collagen–G-90 film enhances wound closure and collagen remodeling: FTIR imaging and PCA-kNN analysis","authors":"Fatemeh Elmi ,&nbsp;Maryam Mitra Elmi ,&nbsp;Lisa Vaccari ,&nbsp;Diana E. Bedolla","doi":"10.1016/j.saa.2026.127480","DOIUrl":"10.1016/j.saa.2026.127480","url":null,"abstract":"<div><div>Chitosan-based films containing different bioactive components were developed to assess their effect on wound healing in a dermal rat skin tissue. Four formulations pure chitosan, chitosan–collagen (ACS), chitosan conjugated with glycolipoprotein extract from earth worm, namely chitosan–G-90, and a mixture of three constituents, chitosan–ACS–G-90 were compared with an untreated control. 14 days post-injury, wound closure was greatest in the chitosan–ACS–G-90 group. Ratio mapping (1325–1351 cm⁻¹/1190–1357 cm⁻¹) confirmed that the chitosan–ACS–G-90 film provided the most consistent improvement in collagen organization, highlighting its superior role in matrix maturation. Vector-normalized, second-derivative spectra from the protein (1710–1475 cm⁻¹) and fingerprint (1800–900 cm⁻¹) regions revealed treatment-dependent variations in amide I/II profiles, collagen triple-helix markers (1338 cm⁻¹, 1203 cm⁻¹), and distinct absorptions in the carbohydrate associated region (∼1150–1000 cm⁻¹). These bands while often attributed to glycosaminoglycans or other ECM polysaccharides may also include overlapping contributions from collagen side chains and nucleic acid phosphate vibrations (DNA/RNA), reflecting the inherently complex biochemical composition of dermal ECM. Principal Component Analysis (PCA) clearly separated the chitosan–ACS–G-90 group from untreated dermis, with loadings associated with cross-link–related (∼1690 cm⁻¹) and amide I/II bands (∼1658 cm⁻¹, 1566–1543 cm⁻¹). Classification of PCA scores using k-Nearest Neighbors (kNN) achieved high accuracy (AUC = 0.993). These results show that the chitosan–ACS–G-90 accelerates wound closure and enhances collagen remodeling.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"352 ","pages":"Article 127480"},"PeriodicalIF":4.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146076362","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Diagnosis of Systemic Lupus Erythematosus (SLE) and Lupus Nephritis (LN) based on a Raman-Metabolomics multimodal re-annotated segment hierarchical fusion model","authors":"Wentao Qin ,&nbsp;Ziyang Zhang ,&nbsp;Xinya Chen ,&nbsp;Wu Xu ,&nbsp;Qianxi Xu ,&nbsp;Chen Chen","doi":"10.1016/j.saa.2025.127355","DOIUrl":"10.1016/j.saa.2025.127355","url":null,"abstract":"<div><div>Significant advances have been achieved in the field of Computer-Aided Diagnosis (CAD) for Systemic Lupus Erythematosus (SLE) and Lupus Nephritis (LN), particularly in the field of multimodal fusion diagnosis. The integration of Raman spectroscopy and metabolomics presents a promising direction for disease diagnosis; however, current Metabolomics-Raman fusion models still encounter challenges such as inadequate interpretability and inefficient utilization of inter-modal interaction information. To address these issues, this study proposes a dedicated multimodal fusion framework for Metabolomics-Raman data, encompassing a Metabolite-Raman re-annotation segmentation method and a Multimodal Segment Hierarchical Fusion Model (MSHFM) constructed based on this method. Through the screening of differential metabolites and the analysis of their superclasses, a mapping relationship was constructed that directly links molecular vibration modes (as identified by Raman spectral wavenumbers) to metabolic superclasses. This approach facilitates metabolically meaningful spectral segmentation, overcoming the limitations of traditional physical partitioning. The MSHFM excavates the correlations between spectral segments and differential metabolites through the Cross-Attention Pre-fusion Module (CAPM), and realizes hierarchical fusion and feature enhancement via the Sample-level Adaptive Multimodal Segment Fusion Module (Slam-SFM). The classification accuracies on the SLE-LN and HC-SLE datasets reach 93.33% and 100%, respectively. We identified key biomarkers such as uric acid and prostaglandin E1 by analyzing the Metabolomics-Raman association patterns using the attention mechanism, which provides an interpretable new method and potential biomarkers for the diagnosis and mechanistic research of SLE and LN.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"352 ","pages":"Article 127355"},"PeriodicalIF":4.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146115507","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}