Molybdenum Disulfide-Assisted Spontaneous Formation of Multistacked Gold Nanoparticles for Deep Learning-Integrated Surface-Enhanced Raman Scattering.

Several fabrication methods have been developed for label-free detection in various fields. However, fabricating high-density and highly ordered nanoscale architectures by using soluble processes remains a challenge. Herein, we report a biosensing platform that integrates deep learning with surface-enhanced Raman scattering (SERS), featuring large-area, close-packed three-dimensional (3D) architectures of molybdenum disulfide (MoS2)-assisted gold nanoparticles (AuNPs) for the on-site screening of coronavirus disease (COVID-19) using human tears. Some AuNPs are spontaneously synthesized without a reducing agent because the electrons induced on the semiconductor surface reduce gold ions when the Fermi level of MoS2 and the gold electrolyte reach equilibrium. With the addition of polyvinylpyrrolidone, a two-dimensional large-area MoS2 layer assisted in the formation of close-packed 3D multistacked AuNP structures, resembling electroless plating. This platform, with a convolutional neural network-based deep learning model, achieved outstanding SERS performance at subterascale levels despite the microlevel irradiation power and millisecond-level acquisition time and accurately assessed susceptibility to COVID-19. These results suggest that our platform has the potential for rapid, low-damage, and high-throughput label-free detection of exceedingly low analyte concentrations.

One-Stop Plasmonic Nanocube-Excited SERS Immunoassay Platform of Multiple Cardiac Biomarkers for Rapid Screening and Progressive Tracing of Acute Myocardial Infarction.

Rapid and precise acute myocardial infarction (AMI) diagnosis is essential for preventing patient death. In addition, the complementary roles of creatine kinase muscle brain (CK-MB) and cardiac troponin I (cTnI) cardiac biomarkers in the early and late stages of AMI demand their simultaneous detection, which is difficult to implement using conventional fluorescence and electrochemical technologies. Here, a nanotechnology-based one-stop immuno-surface-enhanced Raman scattering (SERS) detection platform is reported for multiple cardiac indicators for the rapid screening and progressive tracing of AMI events. Optimal SERS is achieved using optical property-based, excitation wavelength-optimized, and high-yield anisotropic plasmonic gold nanocubes. Optimal immunoassay reaction efficiencies are achieved by increasing immobilized antibodies. Multiple simultaneous detection strategies are implemented by incorporating two different Raman reports with narrow wavenumbers corresponding to two indicators and by establishing a computational SERS mapping process to accurately detect their concentrations, irrespective of multiple enzymes in the human serum. The SERS platform precisely estimated AMI onset and progressive timing in human serum and made rapid AMI identification feasible using a portable Raman spectrometer. This integrated platform is hypothesized to significantly contribute to emergency medicine and forensic science by providing timely treatment and observation.

3D plasmonic coral nanoarchitecture paper for label-free human urine sensing and deep learning-assisted cancer screening.

Practical human biofluid sensing requires a sensor device to differentiate patients from the normal group with high sensitivity and specificity. Label-free molecular identification from human biofluids allows direct classification of abnormal samples, providing insights for disease diagnosis and finding of new biomarkers. Here, we introduce a label-free surface-enhanced Raman scattering sensor based on a three-dimensional plasmonic coral nanoarchitecture (3D-PCN), which has strong electromagnetic field enhancement through multiple hot spots. The 3D-PCN was synthesized on a paper substrate via direct one-step gold reduction, forming a coral-like nanoarchitecture with high absorption property for biofluids. This was fabricated as a urine test strip and then integrated with a handheld Raman system to develop an on-site urine diagnostic platform. The developed platform successfully classified the human prostate and pancreatic cancer urines in a label-free method supported by two types of deep learning networks, with high clinical sensitivity and specificity. Our technology has the potential to be utilized not only for urinary cancer diagnosis but also for various human biofluid sensing systems as a future point-of-care testing platform.

Effect of Dentin Desensitizer Containing Novel Bioactive Glass on the Permeability of Dentin.

The objective of this study was to evaluate the effect of novel bioactive glass (BAG)-containing desensitizers on the permeability of dentin. Experimental dentin desensitizers containing 3 wt% BAG with or without acidic functional monomers (10-MDP or 4-META) were prepared. A commercial desensitizer, Seal & Protect (SNP), was used as a control. To evaluate the permeability of dentin, real-time dentinal fluid flow (DFF) rates were measured at four different time points (demineralized, immediately after desensitizer application, after two weeks in simulated body fluid (SBF), and post-ultrasonication). The DFF reduction rate (ΔDFF) was also calculated. The surface changes were analyzed using field emission scanning electron microscopy (FE-SEM). Raman spectroscopy was performed to analyze chemical changes on the dentin surface. The ΔDFF of the desensitizers containing BAG, BAG with 10-MDP, and BAG with 4-META significantly increased after two weeks of SBF storage and post-ultrasonication compared to the SNP at each time point (p < 0.05). Multiple precipitates were observed on the surfaces of the three BAG-containing desensitizers. Raman spectroscopy revealed hydroxyapatite (HAp) peaks on the dentin surfaces treated with the three BAG-containing desensitizers. Novel BAG-containing dentin desensitizers can reduce the DFF rate about 70.84 to 77.09% in the aspect of reduction of DFF through the HAp precipitations after two weeks of SBF storage.

Adiponectin-targeted SERS immunoassay biosensing platform for early detection of gestational diabetes mellitus.

The anisotropic gold nanotriangles (AuNTs) were synthesized by a fast seedless growth process. The high-yield monodispersed AuNT colloids were obtained through a purification process based on depletion-induced interactions. AuNTs were modulated with a localized surface plasmon resonance (LSPR) peak of 638 nm wavelength coherent with the Raman excitation light. However, from finite element computation results, the AuNT clusters showed better performance for the 785 nm laser source due to a red shift in their LSPR properties, hence it was selected for the surface-enhanced Raman scattering (SERS) immunoassay. A self-assembly strategy using a thiol group and ON-OFF strategy in the heat map was performed to ensure the stability of SERS immunoassay platform. The sandwich SERS immunoassay biosensor platform for adiponectin detection demonstrated a wide assay range (10-15-10-6 g/mL), good reliability (R2 = 0.994, clinically relevant range), femto-scale limit of detection (3.0 × 10-16 g/mL), and excellent selectivity without interference from other biomarkers. This showed the possibility of effectively detecting adiponectin levels in the biofluids of pregnant women. Therefore, our technology is the first to quantitatively detect adiponectin based on SERS technology for early detection of gestational diabetes mellitus and has the potential to be used as a clinical biosensor capable of diagnosing various obstetric diseases during early pregnancy.

Biological SERS-active sensor platform based on flexible silk fibroin film and gold nanoislands.

In contrast to conventional surface-enhanced Raman scattering (SERS) platforms implemented on non-biological substrates, silk fibroin has the unique advantages of long-term biosafety and controllable biodegradability for in vitro and in vivo biomedical applications, as well as flexibility and process-compatibility. In this study, a silk fibroin film was developed to fabricate a flexible SERS sensor template with nanogap-rich gold nanoislands. The proposed biological SERS platform presents fairly good enhancements in detection performance such as detection limit, sensitivity, and signal-to-noise ratio. In particular, the sensitivity improvement was by more than 10 times compared to that of the counterpart sample, and an excellent spatial reproducibility of 2.8% was achieved. In addition, the near-field calculation results were consistent with the experimental results, and the effect of surface roughness of the silk substrate was investigated in a quantitative way. It is believed that biological SERS-active sensors could provide the potential for highly sensitive, cost-effective, and easily customizable nanophotonic platforms that include new capabilities for future healthcare devices.

An excitation wavelength-optimized, stable SERS biosensing nanoplatform for analyzing adenoviral and AstraZeneca COVID-19 vaccination efficacy status using tear samples of vaccinated individuals.

We introduce a label-free surface-enhanced Raman scattering (SERS) biosensing platform equipped with metallic nanostructures that can identify the efficacy of Oxford-AstraZeneca (AZD1222) vaccine in vaccinated individuals using non-invasive tear samples. We confirmed the hypothesis that the tears of people who receive the AZD1222 vaccine may be similar to those of adenovirus epidemic keratoconjunctivitis patients since the Oxford-AstraZeneca vaccine is derived from a replication-deficient ChAdOx1 vector of chimpanzee adenovirus. Additionally, we confirmed the potential of the three markers for estimating the vaccination status via analyzing the signals emanating from antibodies or immunoglobulin G by-product using our label-free, SERS biosensing technique with a high reproducibility (<3% relative standard deviation), femtomole-scale limit of detection (1 × 10-14 M), and high SERS response of >108. Therefore, our label-free SERS biosensing nanoplatforms with long-term storage and robust stability will enable rapid and robust monitoring of the vaccine presence in vaccinated individuals.

Effect of Novel Bioactive Glass-Containing Dentin Adhesive on the Permeability of Demineralized Dentin.

This study aimed to evaluate the effect of a novel bioactive glass (BAG)-containing dentin adhesive on the permeability of demineralized dentin. Bioactive glass (85% SiO2, 15% CaO) was fabricated using the sol-gel process, and two experimental dentin adhesives were prepared with 3 wt% silica (silica-containing dentin adhesive; SCA) or BAG (BAG-containing dentin adhesive; BCA). Micro-tensile bond strength (µTBS) test, fracture mode analysis, field-emission scanning electron microscopy (FE-SEM) analysis of adhesive and demineralized dentin, real-time dentinal fluid flow (DFF) rate measurement, and Raman confocal microscopy were performed to compare SCA and BCA. There was no difference in µTBS between the SCA and BCA (p > 0.05). Multiple precipitates were evident on the surface of the BCA, and partial occlusion of dentinal tubules was observed in FE-SEM of BCA-approximated dentin. The DFF rate was reduced by 50.10% after BCA approximation and increased by 6.54% after SCA approximation. Raman confocal spectroscopy revealed an increased intensity of the hydroxyapatite (HA) peak on the dentin surface after BCA application. The novel BAG-containing dentin adhesive showed the potential of both reducing dentin permeability and dentin remineralization.

Label-free breast cancer detection using fiber probe-based Raman spectrochemical biomarker-dominated profiles extracted from a mixture analysis algorithm.

We report the development of a label-free, simple, and high efficiency breast cancer detection platform with multimodal biomarker analytic algorithms on a portable 785 nm Raman setup with an endoscopic Raman-lensed fiber optic probe. We propose a multimodal biomarker extraction algorithm (PCMA) implemented by combining a multivariate statistics principal component analysis (PCA) algorithm and a multivariate curve resolution-alternating least squares (MCR-ALS) computational model for extraction of the biomarker information hidden in Raman spectrochemical data. We show that the six Raman spectrochemical peaks at 1009, 1270, 1305/1443, 1658, and 1750 cm-1 assigned to phenylalanine, amide III in proteins, CH2 deformation in lipids, amide I in proteins, and carbonyl, respectively, can be used as a biomarker for breast cancer diagnosis using the biomarker-dominated PCMA spectrochemical spectra of breast tissues. From 20 human breast tissues, the PCMA-linear discriminant analysis (PCMA-LDA) identification method achieved high classification performance with a sensitivity and specificity >99% along with an improvement of approximately 4.5% compared to the performance without the PCMA mixture analysis algorithm. Our label-free breast cancer detection method has the potential for clinical application to diagnose breast cancer in real-time during surgery.

Wavelength-dependent label-free identification of isolated nontuberculous mycobacteria using surface-enhanced Raman spectroscopy.

We investigated the effect of Raman excitation wavelengths on the surface-enhanced Raman spectroscopy (SERS)-based identification of isolated nontuberculous mycobacteria (NTM). The SERS spectra with 3 commonly used excitation wavelengths, 532, 638, and 785 nm, were compared across 6 representative NTM species that primarily cause human NTM infections in Korea and the United States; these species were identified. The statistical differences among NTM SERS spectra at each Raman excitation wavelength were verified using 1-way analysis of variance, and the 6 NTM species were identified using principal components-linear discriminant analysis with leave-one-out cross validation. The identification accuracies with aromatic amino acid biomarkers were 99.3%, 91.3%, and 90.7% for 532, 638, and 785 nm, respectively. We believe that the proposed SERS protocol with aromatic amino acid biomarkers at the 532-nm Raman excitation wavelength will enable fast and accurate identification of NTM compared to previous identification methods.

Surface Roughness on the Slots and Wings of Various Ceramic Self-Ligating Brackets and their Potential Concern on Biofilm Formation.

OBJECTIVE: The surface roughness of various orthodontic materials could affect biofilm formation and friction. The purpose of this study was to examine the surface roughness and chemical composition of the slots and wings of several ceramic self-ligating brackets. STUDY DESIGN: Four types of ceramic self-ligating brackets were separated into experimental groups (DC, EC, IC, and QK) while a metal self-ligating bracket (EM) was used as the control group. Atomic force microscopy and energy-dispersive x-ray spectroscope were used to examine the surface roughness and chemical composition of each bracket slot and wing. RESULTS: The control group was made of ferrum and chrome while all the experimental groups were comprised of aluminum and oxide. There was a statistically significant difference in the roughness average (Sa) among the various self-ligating brackets (p< 0.001 in slots and p<0.01 in the wing). The slots in the EC group had the lowest Sa, followed by the DC, IC, control, and QK groups. The wings in the IC group had the lowest Sa, followed by the EC, DC, control, and QK groups. CONCLUSIONS: There is a significant difference in the surface roughness of the slots and wings among several types of ceramic self-ligating brackets.

Label-Free Surface-Enhanced Raman Spectroscopy Biosensor for On-Site Breast Cancer Detection Using Human Tears.

Surface-enhanced Raman scattering (SERS) is an ultrasensitive molecular screening technique with greatly enhanced Raman scattering signals from trace amounts of analytes near plasmonic nanostructures. However, research on the development of a sensor that balances signal enhancement, reproducibility, and uniformity has not yet been proposed for practical applications. In this study, we demonstrate the potential of the practical application for detecting or predicting asymptomatic breast cancer from human tears using a portable Raman spectrometer with an identification algorithm based on multivariate statistics. This potentiality was realized through the fabrication of a plasmonic SERS substrate equipped with a well-aligned, gold-decorated, hexagonal-close-packed polystyrene (Au/HCP-PS) nanosphere monolayer that provided femtomole-scale detection, giga-scale enhancement, and <5% relative standard deviation for reliability and reproducibility, regardless of the measuring site. Our results can provide a first step toward developing a noninvasive, real-time screening technology for detecting asymptomatic tumors and preventing tumor recurrence.

Effects of scleral collagen crosslinking with different carbohydrate on chemical bond and ultrastructure of rabbit sclera: Future treatment for myopia progression.

BACKGROUND: Myopia is the most common ocular disorder and is mainly caused by axial elongation of the sclera. If the stiffness of sclera increased, it can inhibit myopia progression. The aim of this study is to compare the effect of the collagen crosslinking with different types and concentrations of carbohydrates on chemical bond and ultrastructural change of rabbit sclera. METHODS: Nine New Zealand white rabbits were treated with five, sequential sub-Tenon injections of 0.15 mL solutions of ribose, sucrose, and glycogen of 0.1, 0.2 and 0.4 M concentration at the right eye over 14 days. Ten weeks after the last injection, the rabbits were sacrificed and chemical bond and ultrastructural changes were compared with those of the untreated left sclera using Raman spectroscopy, atomic force microscopy (AFM), and histology. RESULTS: Raman spectroscopy of the control and cross-linked rabbit sclera tissue revealed different types of collagen interactions. Raman shift of 919 cm-1 (C-C stretching and vibration of the proline ring in collagen) was the highest in ribose, followed by sucrose and glycogen. Total energy intensity was also highest in ribose, followed by sucrose and glycogen, and showed a tendency to increase at higher concentrations. AFM revealed interlocking arrangements of collagen fibrils. The collagen fibril diameter was 105.6 ± 21.2 nm, 109.4 ± 28.8 nm, 113.1 ± 30.8 nm and 137.6 ± 25.3 nm for control group, 0.4 M glycogen, sucrose, and ribose, respectively. Histology indicated increased density of the collagen bundle and no increase in inflammatory cell recruitment compared to control at high concentrations of ribose. CONCLUSIONS: Scleral crosslinking using glycation increased the scleral biomechanical rigidity and these results were particularly pronounced in ribose. Scleral crosslinking using glycation may be a promising method for inhibiting high myopia progression.

Surface analysis of metal clips of ceramic self-ligating brackets.

OBJECTIVE: The aim of this study was to analyze the surface composition, roughness, and relative friction of metal clips from various ceramic self-ligating brackets. METHODS: Six kinds of brackets were examined. The control group (mC) consisted of interactive metal self-ligating brackets while the experimental group (CC, EC, MA, QK, and WA) consisted of interactive ceramic self-ligating brackets. Atomic force microscopy-lateral force microscopy and scanning electron microscopy-energy-dispersive X-ray spectroscopy were used to analyze the surface of each bracket clip. RESULTS: All the clips in the experimental groups were coated with rhodium except for the QK clip. The results showed that the QK clip had the lowest average roughness on the outer surface, followed by the MA, EC, WA, and CC clips. However, the CC clip had the lowest average roughness on the inner surface, followed by the QK, WA, MA, and EC clips. The QK clip also had the lowest relative friction on the outer surface, followed by the MA, EC, CC, and WA clips. Likewise, the CC clip had the lowest relative friction on the inner surface, followed by the QK, WA, MA, and EC clips. CONCLUSIONS: The surface roughness and relative friction of the rhodium-coated clips were generally higher than those of the uncoated clips.

Paper-Based Surface-Enhanced Raman Spectroscopy for Diagnosing Prenatal Diseases in Women.

We report the development of a surface-enhanced Raman spectroscopy sensor chip by decorating gold nanoparticles (AuNPs) on ZnO nanorod (ZnO NR) arrays vertically grown on cellulose paper (C). We show that these chips can enhance the Raman signal by 1.25 × 107 with an excellent reproducibility of <6%. We show that we can measure trace amounts of human amniotic fluids of patients with subclinical intra-amniotic infection (IAI) and preterm delivery (PTD) using the chip in combination with a multivariate statistics-derived machine-learning-trained bioclassification method. We can detect the presence of prenatal diseases and identify the types of diseases from amniotic fluids with >92% clinical sensitivity and specificity. Our technology has the potential to be used for the early detection of prenatal diseases and can be adapted for point-of-care applications.

A label-free cellulose SERS biosensor chip with improvement of nanoparticle-enhanced LSPR effects for early diagnosis of subarachnoid hemorrhage-induced complications.

It is very difficult to predict some complications after subarachnoid hemorrhage (SAH), despite rapid advances in medical science. Herein, we introduce a label-free cellulose surface-enhanced Raman spectroscopy (SERS) biosensor chip with pH-functionalized, gold nanoparticle (AuNP)-enhanced localized surface plasmon resonance (LSPR) effects for identification of SAH-induced cerebral vasospasm and hydrocephalus caused by cerebrospinal fluid (CSF). The SERS biosensor chip was implemented by the synthesis reaction of the AuNPs, which were charged positively through pH level adjustment, onto a negatively-charged cellulose substrate with ξ = -30.7 mV. The zeta potential, nanostructural properties, nanocrystallinity, and computational calculation-based electric field distributions of the cellulose-originated AuNPs were optimized to maximize LSPR phenomena and then characterized. Additionally, the performance of the SERS biosensor was compared under two representative excitation laser sources in the visible region (532 nm) and near-infrared region (785 nm). The Raman activities of our SERS biosensor chip were evaluated by trace small molecules (crystal violet, 2 µL), and the biosensor achieved an enhancement factor of 3.29 × 109 for the analytic concept with an excellent reproducibility of 8.5% relative standard deviation and a detection limit of 0.74 pM. Furthermore, the experimental results revealed that the five proposed SERS-based biomarkers could provide important information for identifying and predicting SAH-induced cerebral vasospasm and hydrocephalus complications (91.1% reliability and 19.3% reproducibility). Therefore, this facile and effective principle of our SERS biosensor chip may inspire the basis and strategies for the development of sensing platforms to predict critical complications in various neurosurgical diagnoses.

Label-free monitoring of inflammatory tissue conditions using a carrageenan-induced acute inflammation rat model.

Although the confirmation of inflammatory changes within tissues at the onset of various diseases is critical for the early detection of disease and selection of appropriate treatment, most therapies are based on complex and time-consuming diagnostic procedures. Raman spectroscopy has the ability to provide non-invasive, real-time, chemical bonding analysis through the inelastic scattering of photons. In this study, we evaluate the feasibility of Raman spectroscopy as a new, easy, fast, and accurate diagnostic method to support diagnostic decisions. The molecular changes in carrageenan-induced acute inflammation rat tissues were assessed by Raman spectroscopy. Volumes of 0 (control), 100, 150, and 200 µL of 1% carrageenan were administered to rat hind paws to control the degree of inflammation. The prominent peaks at [1,062, 1,131] cm-1 and [2,847, 2,881] cm-1 were selected as characteristic measurements corresponding to the C-C stretching vibrational modes and the symmetric and asymmetric C-H (CH2 ) stretching vibrational modes, respectively. Principal component analysis of the inflammatory Raman spectra enabled graphical representation of the degree of inflammation through principal component loading profiles of inflammatory tissues on a two-dimensional plot. Therefore, Raman spectroscopy with multivariate statistical analysis represents a promising method for detecting biomolecular responses based on different types of inflammatory tissues.

Low-Cost Label-Free Biosensing Bimetallic Cellulose Strip with SILAR-Synthesized Silver Core-Gold Shell Nanoparticle Structures.

We introduce a label-free biosensing cellulose strip sensor with surface-enhanced Raman spectroscopy (SERS)-encoded bimetallic core@shell nanoparticles. Bimetallic nanoparticles consisting of a synthesis of core Ag nanoparticles (AgNP) and a synthesis of shell gold nanoparticles (AuNPs) were fabricated on a cellulose substrate by two-stage successive ionic layer absorption and reaction (SILAR) techniques. The bimetallic nanoparticle-enhanced localized surface plasmon resonance (LSPR) effects were theoretically verified by computational calculations with finite element models of optimized bimetallic nanoparticles interacting with an incident laser source. Well-dispersed raspberry-like bimetallic nanoparticles with highly polycrystalline structure were confirmed through X-ray and electron analyses despite ionic reaction synthesis. The stability against silver oxidation and high sensitivity with superior SERS enhancement factor (EF) of the low-cost SERS-encoded cellulose strip, which achieved 3.98 × 108 SERS-EF, 6.1%-RSD reproducibility, and <10%-degraded sustainability, implicated the possibility of practical applications in high analytical screening methods, such as single-molecule detection. The remarkable sensitivity and selectivity of this bimetallic biosensing strip in determining aquatic toxicities for prohibited drugs, such as aniline, sodium azide, and malachite green, as well as monitoring the breast cancer progression for urine, confirmed its potential as a low-cost label-free point-of-care test chip for the early diagnosis of human diseases.

Rapid label-free identification of Klebsiella pneumoniae antibiotic resistant strains by the drop-coating deposition surface-enhanced Raman scattering method.

Although many methodologies have been developed to identify unknown bacteria, bacterial identification in clinical microbiology remains a complex and time-consuming procedure. To address this problem, we developed a label-free method for rapidly identifying clinically relevant multilocus sequencing typing-verified quinolone-resistant Klebsiella pneumoniae strains. We also applied the method to identify three strains from colony samples, ATCC70063 (control), ST11 and ST15; these are the prevalent quinolone-resistant K. pneumoniae strains in East Asia. The colonies were identified using a drop-coating deposition surface-enhanced Raman scattering (DCD-SERS) procedure coupled with a multivariate statistical method. Our workflow exhibited an enhancement factor of 11.3×106 to Raman intensities, high reproducibility (relative standard deviation of 7.4%), and a sensitive limit of detection (100 pM rhodamine 6G), with a correlation coefficient of 0.98. All quinolone-resistant K. pneumoniae strains showed similar spectral Raman shifts (high correlations) regardless of bacterial type, as well as different Raman vibrational modes compared to Escherichia coli strains. Our proposed DCD-SERS procedure coupled with the multivariate statistics-based identification method achieved excellent performance in discriminating similar microbes from one another and also in subtyping of K. pneumoniae strains. Therefore, our label-free DCD-SERS procedure coupled with the computational decision supporting method is a potentially useful method for the rapid identification of clinically relevant K. pneumoniae strains.

Highly Reproducible Au-Decorated ZnO Nanorod Array on a Graphite Sensor for Classification of Human Aqueous Humors.

Gold-decorated, vertically grown ZnO nanorods (NRs) on a flexible graphite sheet (Au/ZnONRs/G) were developed for surface-enhanced Raman scattering (SERS)-based biosensing to identify trace amounts of human aqueous humors. This Au/ZnONRs/G SERS-functionalized sensor was fabricated via two steps: hydrothermal synthesis-induced growth of ZnO NRs on graphite sheets for nanostructure fabrication, followed by e-beam evaporator-induced gold metallization on ZnONRs/G for SERS functionalization. The thickness of the Au layer and the height of the ZnO NRs for enhancing SERS performance were adjusted to maximize Raman intensity, and the optimized Au/ZnONRs/G nanostructures were verified by the electric finite element computational models to maximize the electric fields. The proposed Au/ZnONRs/G SERS sensor showed an enhancement factor of 2.3 × 106 via rhodamine 6G Raman probe and excellent reproducibility (relative standard deviation of <10%) via Raman mapping of a SERS active area with a square of 100 × 100 µm2. To evaluate the actual bioapplicability of point-of-care-testing (POCT) analysis in clinics, SERS data acquisition was performed with an integration time of 1 s from a 1 µL analytic droplet of the sample. The performance of this Au/ZnONRs/G sensor was evaluated using human aqueous humors with cataract and two oxidative stress-induced eye diseases, age-related macular degeneration, and diabetic macular edema. These three eye diseases could be identified without any labeling or modification using the Au/ZnONRs/G SERS sensor and the computational algorithm incorporating a support vector machine and multivariate statistical prediction. Therefore, these findings indicate that our label-free, highly reproducible and flexible Au/ZnONRs/G SERS-functionalized sensor supported by a multivariate statistics-derived bioclassification method has great potential in POCT applications for identifying eye diseases.