Dkd risk assessment method, system and medium based on surface-enhanced raman scattering
By using surface-enhanced Raman scattering technology and machine learning models, the problems of insufficient sensitivity and high cost of AGEs detection in DKD risk assessment have been solved, enabling rapid and accurate detection of AGEs in urine and supporting early intervention.
Patent Information
- Application Number
- CN202611197211.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-07
- Publication Date
- 2026-11-03
AI Technical Summary
Existing DKD risk assessment methods suffer from insufficient sensitivity, complex operation, and high cost of AGEs detection, making it difficult to meet the needs of large-scale population screening and long-term monitoring.
Gold nanorods were prepared by seed growth using a surface-enhanced Raman scattering (SERS) method. Carboxyl or amino functional groups were introduced as SERS active substrates. AGEs molecules in urine were separated and enriched by centrifugation, solid-phase extraction, and molecular sieve purification. SERS characteristic spectra were collected, and a machine learning model was constructed for risk assessment.
This technology enables rapid and accurate detection of AGEs in urine, allowing for intervention before significant kidney damage occurs. It also improves the sensitivity and specificity of the detection while reducing costs.
Smart Images
Figure CN123033732A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of risk assessment technology for diabetic nephropathy, and more specifically, to a method, system, and medium for risk assessment of DKD based on surface-enhanced Raman scattering. Background Technology
[0002] Diabetic kidney disease (DKD) is one of the most serious microvascular complications of diabetes, and early diagnosis is crucial to delaying disease progression. Currently, urinary microalbumin testing is the primary clinical method for assessing DKD risk; however, this method suffers from insufficient sensitivity and poor specificity, often only detecting significant abnormalities after kidney damage has reached a certain level. Advanced glycation end products (AGEs), as key biomarkers for DKD progression, show changes in their urinary levels earlier and have better predictive value. However, traditional detection methods such as ELISA and high-performance liquid chromatography (HPLC) are complex, costly, and have long testing cycles, making them unsuitable for rapid clinical screening. Especially for large-scale population screening and long-term monitoring, the high cost and low throughput of existing technologies are major bottlenecks restricting their clinical application.
[0003] Existing DKD risk assessment methods suffer from insufficient sensitivity of AGEs detection, complex operation, and high cost. Summary of the Invention
[0004] The purpose of this application is to provide a method, system, and medium for DKD risk assessment based on surface-enhanced Raman scattering (SERS). By leveraging the high sensitivity and spectral characteristics of SERS technology, it enables rapid and accurate detection of AGEs in urine, allowing clinicians to take intervention measures before significant kidney function damage occurs.
[0005] This application also provides a DKD risk assessment method based on surface-enhanced Raman scattering, including: Gold nanorods with a predetermined aspect ratio were prepared by seed growth method, and carboxyl or amino functional groups were introduced on the surface of the gold nanorods to obtain SERS active substrates. Urine samples were obtained, and AGEs molecules in the urine samples were separated and enriched based on centrifugation, solid-phase extraction and molecular sieve purification, and the filtrate was collected. The filtrate was incubated with a SERS active substrate under specific pH conditions to allow AGEs molecules to adsorb onto the surface of the SERS active substrate, thus obtaining an AGEs-SERS complex. The SERS characteristic spectrum of the AGEs-SERS complex was acquired, and the spectral characteristic parameters were extracted; A classification model is constructed based on machine learning algorithms, and spectral feature parameters are analyzed to obtain the DKD risk classification assessment results.
[0006] Optionally, in the DKD risk assessment method based on surface-enhanced Raman scattering described in the embodiments of this application, gold nanorods with a predetermined aspect ratio are prepared based on the seed growth method, and carboxyl or amino functional groups are introduced on the surface of the gold nanorods to obtain a SERS-active substrate, specifically including: Prepare a 0.1 mM chloroauric acid aqueous solution as a precursor and add hexadecyltrimethylammonium bromide as a surfactant; Gold seeds of 2-5 nm are formed by injection of sodium borohydride solution to obtain gold seed solution; The gold seed solution was added to a growth solution containing chloroauric acid, hexadecyltrimethylammonium bromide, silver nitrate and ascorbic acid, and reacted at 25-30℃ for 12-24 hours. The concentration of silver nitrate was adjusted to 0.01-0.1mM to control the aspect ratio of the gold nanorods within the range of 2.5-4.0, thereby synthesizing gold nanorods. The synthesized gold nanorods were purified by centrifugation and dispersed in a 1 mM solution of mercaptoundecanoic acid or mercaptoethylamine ethanol. The mixture was stirred at room temperature, and carboxyl or amino functional groups were modified on the surface of the gold nanorods to obtain the SERS active substrate.
[0007] Optionally, in the DKD risk assessment method based on surface-enhanced Raman scattering described in the embodiments of this application, a urine sample is obtained, and AGEs molecules in the urine sample are separated and enriched based on centrifugation, solid-phase extraction, and molecular sieve purification, and the filtrate is collected, specifically including: Obtain a urine sample, centrifuge the urine sample to obtain the supernatant, and then filter the supernatant through a micropore. The supernatant after microfiltration is fused and rinsed, and the supernatant after rinsing is then eluted. The supernatant after elution was concentrated to obtain the extract; The extract was purified by analytical sieve to obtain the filtrate.
[0008] Optionally, in the DKD risk assessment method based on surface-enhanced Raman scattering described in the embodiments of this application, the filtrate is incubated with a SERS-active substrate under specific pH conditions to allow AGEs molecules to adsorb onto the surface of the SERS-active substrate, thereby obtaining an AGEs-SERS complex, specifically including: The filtrate was mixed with gold nanorods at pH 5.5–6.5. Incubate with shaking at a volume ratio of 1:50 for 30-40 minutes; Adding MgCl2 solution induces aggregation, causing AGEs molecules to adsorb onto the surface of gold nanorods via electrostatic interactions or covalent bonds, thus obtaining AGEs-SERS complexes.
[0009] Optionally, in the DKD risk assessment method based on surface-enhanced Raman scattering described in the embodiments of this application, the SERS characteristic spectrum of the AGEs-SERS complex is collected, and spectral characteristic parameters are extracted, specifically including: The SERS characteristic spectrum of the AGEs-SERS complex was acquired using a 785nm laser with a power of 5mW in a backscattering geometry. The collected SERS characteristic spectra were baseline corrected and smoothed and denoised to obtain the preprocessed SERS characteristic spectra. Based on SERS characteristic spectra, multiple characteristic peaks are located by peak detection, and the integral intensity and peak height of each characteristic peak are calculated. The spectral characteristic parameters are obtained by calculating the relative ratio of the integral intensity to the peak height of the characteristic peak based on the integral intensity and peak height of the characteristic peak.
[0010] Optionally, in the DKD risk assessment method based on surface-enhanced Raman scattering described in the embodiments of this application, a classification model is constructed based on a machine learning algorithm to analyze spectral characteristic parameters and obtain the DKD risk classification assessment result, specifically including: Urine samples were collected from patients with clinically diagnosed low, medium, and high risk levels of DKD. Obtain the spectral characteristic parameters and corresponding clinical parameters of urine samples, and construct a dataset containing sample feature matrix and risk level labels; The dataset was divided into a training set and a test set in a 7:3 ratio. Input the training set feature matrix and corresponding risk level labels into the model, and optimize the model hyperparameters by traversing parameter combinations through grid search. The trained model was evaluated using 5-fold cross-validation, and the evaluation results were obtained. Optimize the model hyperparameters based on the evaluation results until the model performance meets the target. The feature matrix of the test set or the sample to be tested is input into the model with qualified performance. The model analyzes the correlation between spectral feature parameters and DKD risk and outputs the DKD risk classification assessment result corresponding to the urine sample.
[0011] Secondly, embodiments of this application provide a DKD risk assessment system based on surface-enhanced Raman scattering (SERS). The system includes a memory and a processor. The memory includes a program for a DKD risk assessment method based on SERS. When the program for the DKD risk assessment method based on SERS is executed by the processor, it performs the following steps: Gold nanorods with a predetermined aspect ratio were prepared by seed growth method, and carboxyl or amino functional groups were introduced on the surface of the gold nanorods to obtain SERS active substrates. Urine samples were obtained, and AGEs molecules in the urine samples were separated and enriched based on centrifugation, solid-phase extraction and molecular sieve purification, and the filtrate was collected. The filtrate was incubated with a SERS active substrate under specific pH conditions to allow AGEs molecules to adsorb onto the surface of the SERS active substrate, thus obtaining an AGEs-SERS complex. The SERS characteristic spectrum of the AGEs-SERS complex was acquired, and the spectral characteristic parameters were extracted; A classification model is constructed based on machine learning algorithms, and spectral feature parameters are analyzed to obtain the DKD risk classification assessment results.
[0012] Optionally, in the DKD risk assessment system based on surface-enhanced Raman scattering described in the embodiments of this application, gold nanorods with a predetermined aspect ratio are prepared based on the seed growth method, and carboxyl or amino functional groups are introduced on the surface of the gold nanorods to obtain a SERS-active substrate, specifically including: Prepare a 0.1 mM chloroauric acid aqueous solution as a precursor and add hexadecyltrimethylammonium bromide as a surfactant; Gold seeds of 2-5 nm are formed by injection of sodium borohydride solution to obtain gold seed solution; The gold seed solution was added to a growth solution containing chloroauric acid, hexadecyltrimethylammonium bromide, silver nitrate and ascorbic acid, and reacted at 25-30℃ for 12-24 hours. The concentration of silver nitrate was adjusted to 0.01-0.1mM to control the aspect ratio of the gold nanorods within the range of 2.5-4.0, thereby synthesizing gold nanorods. The synthesized gold nanorods were purified by centrifugation and dispersed in a 1 mM solution of mercaptoundecanoic acid or mercaptoethylamine ethanol. The mixture was stirred at room temperature, and carboxyl or amino functional groups were modified on the surface of the gold nanorods to obtain the SERS active substrate.
[0013] Optionally, in the DKD risk assessment system based on surface-enhanced Raman scattering described in this application embodiment, a urine sample is obtained, and AGEs molecules in the urine sample are separated and enriched based on centrifugation, solid-phase extraction, and molecular sieve purification, and the filtrate is collected, specifically including: Obtain a urine sample, centrifuge the urine sample to obtain the supernatant, and then filter the supernatant through a micropore. The supernatant after microfiltration is fused and rinsed, and the supernatant after rinsing is then eluted. The supernatant after elution was concentrated to obtain the extract; The extract was purified by analytical sieve to obtain the filtrate.
[0014] Thirdly, embodiments of this application also provide a computer-readable storage medium, which includes a DKD risk assessment method program based on surface-enhanced Raman scattering. When the DKD risk assessment method program based on surface-enhanced Raman scattering is executed by a processor, it implements the steps of the DKD risk assessment method based on surface-enhanced Raman scattering as described in any of the preceding claims.
[0015] As described above, the present application provides a DKD risk assessment method, system, and medium based on surface-enhanced Raman scattering (SERS). This involves preparing gold nanorods with a predetermined aspect ratio using a seed growth method, introducing carboxyl or amino functional groups onto the surface of the gold nanorods to obtain an SERS-active substrate; acquiring urine samples, separating and enriching AGEs molecules in the urine samples using centrifugation, solid-phase extraction, and molecular sieve purification, and collecting the filtrate; incubating the filtrate with the SERS-active substrate under specific pH conditions, allowing AGEs molecules to adsorb onto the surface of the SERS-active substrate, obtaining an AGEs-SERS complex; acquiring the SERS characteristic spectrum of the AGEs-SERS complex, and extracting spectral characteristic parameters; constructing a classification model based on a machine learning algorithm, analyzing the spectral characteristic parameters, and obtaining the DKD risk grading assessment result; and achieving rapid and accurate detection of AGEs in urine through the high sensitivity and spectral characteristics of SERS technology. This allows clinicians to take intervention measures before significant kidney function damage occurs. Attached Figure Description
[0016] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0017] Figure 1 A flowchart of the DKD risk assessment method based on surface-enhanced Raman scattering provided for embodiments of this application; Figure 2 A block diagram of a DKD risk assessment system based on surface-enhanced Raman scattering provided for embodiments of this application; Figure 3 A schematic diagram of the generation of AGEs-SERS complexes in the DKD risk assessment system based on surface-enhanced Raman scattering provided in this application embodiment. Detailed Implementation
[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.
[0019] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.
[0020] Please refer to Figure 1 , Figure 1 This is a flowchart of a DKD risk assessment method based on surface-enhanced Raman scattering (SERS) according to some embodiments of this application. This SERS-based DKD risk assessment method is used in a terminal device and includes the following steps: S101, gold nanorods with a predetermined aspect ratio were prepared based on the seed growth method, and carboxyl or amino functional groups were introduced on the surface of the gold nanorods to obtain SERS active substrates. S102, Obtain a urine sample, separate and enrich AGEs molecules in the urine sample based on centrifugation, solid phase extraction and molecular sieve purification, and collect the filtrate; S103, under specific pH conditions, the filtrate is incubated with a SERS active substrate to allow AGEs molecules to adsorb onto the surface of the SERS active substrate, thus obtaining an AGEs-SERS complex. S104, collect the SERS characteristic spectrum of the AGEs-SERS complex and extract the spectral characteristic parameters; S105, a classification model is constructed based on machine learning algorithms, and spectral feature parameters are analyzed to obtain the DKD risk classification assessment results.
[0021] It should be noted that gold nanorods or silver nanoparticles with specific aspect ratios were synthesized using a seed growth method as SERS (surface-enhanced Raman scattering) active substrates, and the adsorption capacity for advanced AGEs (advanced glycosylation end products) was enhanced by modifying the surface with amino / carboxyl functional groups.
[0022] According to embodiments of the present invention, gold nanorods with a predetermined aspect ratio are prepared based on a seed growth method, and carboxyl or amino functional groups are introduced onto the surface of the gold nanorods to obtain a SERS-active substrate, specifically including: Prepare a 0.1 mM chloroauric acid aqueous solution as a precursor and add hexadecyltrimethylammonium bromide as a surfactant; Gold seeds of 2-5 nm are formed by injection of sodium borohydride solution to obtain gold seed solution; The gold seed solution was added to a growth solution containing chloroauric acid, hexadecyltrimethylammonium bromide, silver nitrate and ascorbic acid, and reacted at 25-30℃ for 12-24 hours. The concentration of silver nitrate was adjusted to 0.01-0.1mM to control the aspect ratio of the gold nanorods within the range of 2.5-4.0, thus synthesizing gold nanorods. The synthesized gold nanorods were purified by centrifugation and dispersed in a 1 mM solution of mercaptoundecanoic acid or mercaptoethylamine ethanol. The mixture was stirred at room temperature, and carboxyl or amino functional groups were modified on the surface of the gold nanorods to obtain the SERS active substrate.
[0023] It should be noted that the SERS substrate preparation and optimization process uses metal nanomaterials as the active substrate to enhance the Raman signal, and the specific implementation includes the following key steps: Gold nanorod synthesis: Gold nanorods were prepared using a seed growth method. First, a 0.1 mM aqueous solution of chloroauric acid was prepared as a precursor, and hexadecyltrimethylammonium bromide (CTAB) was added as a surfactant. Sodium borohydride solution was rapidly injected as a reducing agent to form gold seeds of 2-5 nm. Subsequently, the seed solution was added to a growth solution containing chloroauric acid, CTAB, silver nitrate, and ascorbic acid, and the reaction was carried out at 25-30 °C for 12-24 hours. The aspect ratio of the nanorods was controlled within the range of 2.5-4.0 by adjusting the silver nitrate concentration (0.01-0.1 mM).
[0024] Surface functionalization modification: After centrifugation and purification, the synthesized gold nanorods were surface modified with mercaptoundecanoic acid (MUA) or mercaptoethylamine (MEA). Specifically, the nanorods were dispersed in a 1 mM ethanol solution of MUA or MEA and stirred at room temperature for 12 hours. This allowed the thiol groups to form stable Au-S bonds with the gold surface, thereby introducing carboxyl or amino functional groups onto the nanorod surface. These functional groups can form electrostatic interactions or hydrogen bonds with the amino or carboxyl groups in AGEs molecules, significantly improving adsorption efficiency.
[0025] Performance Characterization and Optimization: The plasmon resonance absorption peaks of the nanomaterials were measured using a UV-Vis spectrophotometer. Gold nanorods typically exhibit two characteristic absorption peaks: a transverse peak (approximately 520 nm) and a longitudinal peak (600-800 nm). The morphology and size distribution of the nanorods were observed using transmission electron microscopy (TEM) to ensure the uniformity of the aspect ratio. Finally, the enhancement performance of the substrate was evaluated by measuring the SERS signal intensity of model molecules (such as 4-mercaptobenzoic acid), requiring an enhancement factor of ≥10^6.
[0026] Specifically, the preparation of gold nanorod substrates with an aspect ratio of 3.5 involved first preparing a growth solution containing 0.25 mM chloroauric acid, 0.1 M CTAB, 0.04 mM silver nitrate, and 0.64 mM ascorbic acid. After adding an appropriate amount of gold seeds, the reaction was carried out at 28 °C for 16 hours. The obtained nanorods had an average diameter of 15 nm and a length of 52.5 nm. Subsequently, the nanorods were modified with 1 mM MEA for 12 hours, centrifuged, washed, and resuspended in PBS buffer. This substrate achieved a detection limit of 0.1 nM for the AGEs standard Nε-carboxymethyl lysine (CML), representing a 10-fold increase in adsorption efficiency compared to the unmodified substrate. By adjusting the silver nitrate concentration in the reaction system to 0.06 mM, the aspect ratio could be further increased to 4.0, resulting in a redshift of the longitudinal plasmon resonance peak to 780 nm, which better matched the 785 nm excitation laser, and an approximately 30% enhancement of the SERS signal.
[0027] According to an embodiment of the present invention, a urine sample is obtained, and AGEs molecules in the urine sample are separated and enriched based on centrifugation, solid-phase extraction, and molecular sieve purification, and the filtrate is collected. Specifically, the process includes: Obtain a urine sample, centrifuge the urine sample to obtain the supernatant, and then filter the supernatant through a micropore. The supernatant after microfiltration is fused and rinsed, and the supernatant after rinsing is then eluted. The supernatant after elution was concentrated to obtain the extract; The extract was purified by analytical sieve to obtain the filtrate.
[0028] It should be noted that urine sample pretreatment includes: AGEs molecules are enriched by centrifugation and solid-phase extraction columns, and elution conditions are optimized to selectively remove interfering components such as urea and creatinine, thereby improving detection specificity and sensitivity.
[0029] The urine sample pretreatment process employs a multi-step separation and enrichment technique, which includes the following key steps: Primary centrifugation: Fresh urine samples were centrifuged at 3000 rpm for 10 minutes to remove cell debris and large particles. The supernatant is filtered through a 0.22μm microporous membrane to eliminate interference from bacteria and submicron particles; The filtrate should be stored at -80°C or immediately subjected to further processing to prevent AGEs degradation. Solid-phase extraction enrichment: Target molecules were enriched using an aminopropyl-bonded silica gel solid-phase extraction column (500 mg / 6 mL). Specific procedure: Activation: The column was activated sequentially with 5 mL of methanol, 5 mL of ultrapure water, and 5 mL of PBS buffer (pH 7.4); Sample loading: Pass 5 mL of urine sample through the extraction column at a flow rate of 1 mL / min; Rinsing: Remove polar interfering substances such as urea and creatinine with 3 mL of 5% methanol aqueous solution; Elution: Elute AGEs molecules with 3 mL of methanol:acetic acid (90:10, v / v) solution; Concentration: Concentrate to 200 μL by nitrogen blowing, then bring the volume up to 500 μL with PBS buffer (pH 7.4); Molecular sieve purification: The extract was further purified by passing it through a 10kDa ultrafiltration centrifuge tube (3000g, 30min) to remove small molecular interferences. The filtrate was collected, removing approximately 92% of urea and 88% of creatinine, while retaining more than 95% of AGEs.
[0030] Specifically, 5 mL of morning urine samples from diabetic patients were centrifuged initially and then loaded onto a pre-activated Bond ElutNH2 solid-phase extraction column. The sample was eluted sequentially with 3 mL of 5% methanol and eluted with 3 mL of methanol:acetic acid (90:10). The eluent was concentrated by nitrogen blowing and then brought to a final volume of 500 μL with PBS (pH 7.4). HPLC validation showed that this treatment enriched the CML (carboxymethyl lysine) concentration from 8.7 nM in the original urine to 82.3 nM (recovery rate 94.6%), while reducing the urea content from 327 mM to 15 mM (removal rate 95.4%). SERS analysis of the treated sample showed a 12.8-fold increase in signal-to-noise ratio compared to the untreated sample, and a 35% reduction in the full width at half maximum (FWHM) of the characteristic peak (1602 cm⁻¹), significantly improving spectral resolution.
[0031] According to an embodiment of the present invention, the filtrate is incubated with a SERS-active substrate under specific pH conditions, causing AGEs molecules to adsorb onto the surface of the SERS-active substrate, thereby obtaining an AGEs-SERS complex, specifically comprising: The filtrate was mixed with gold nanorods at pH 5.5–6.5. Incubate with shaking at a volume ratio of 1:50 for 30-40 minutes; Adding MgCl2 solution induces aggregation, causing AGEs molecules to adsorb onto the surface of gold nanorods via electrostatic interactions or covalent bonds, thus obtaining AGEs-SERS complexes.
[0032] It should be noted that the formation of the AGEs-SERS complex includes: Under specific pH conditions, treated urine was incubated with a SERS substrate, and AGEs molecules were oriented to adsorb onto the surface of the nanomaterial through electrostatic interactions / coordination bonds. The adsorption configuration was optimized to obtain the strongest SERS signal.
[0033] The formation process of AGEs-SERS complexes includes the following key steps: Molecular adsorption interface engineering: By measuring the surface zeta potential at different pH levels (3.0-9.0), the pH window (5.5-6.5) that allows AGEs to complement the surface charge of nanoparticles was determined.
[0034] Nanomaterial surface modification: After modification with 3-aminopropyltriethoxysilane (APTES), gold nanorods (aspect ratio 3.8) achieved a surface amino density of 8.2 ± 1.3 groups / nm². By activating carboxylated AGEs with EDC / NHS, covalent amide bonds are formed with amino groups, with a bonding efficiency of 78±5%.
[0035] In-situ conformational stabilization: Adding 5 mM Mg² as a bridging ion enables the carboxyl groups of AGEs to form stable coordination bonds with the surface of nanoparticles. Molecular orientation was optimized by temperature gradient annealing (25-37℃), which increased the Raman intensity at 1600 cm⁻¹ by 2.3 times.
[0036] Specific Implementation Case: A 10 nM AGEs solution was prepared using pH 6.0 PBS buffer (phosphate-buffered saline) and mixed with APTES-modified gold nanorods (OD520 = 2.0) at a volume ratio of 1:50. After incubation at 25°C with shaking for 30 min, 5 μL of 0.1 M MgCl2 solution was added to induce controlled aggregation. SPR monitoring showed that adsorption equilibrium was reached within 120 s, with a surface coverage of 82%. Raman spectroscopy showed that the optimized complex had a signal intensity of 3.7 times that of the unoptimized complex at 1602 cm⁻¹, and the half-maximum width was reduced to 18 cm⁻¹. XPS analysis confirmed the appearance of a new peak at the N1s binding energy of 399.8 eV, confirming the formation of amide bonds. The complex was stable at 4°C for 72 hours with a signal decay of <5%.
[0037] According to an embodiment of the present invention, the SERS characteristic spectrum of the AGEs-SERS complex is acquired, and spectral characteristic parameters are extracted, specifically including: The SERS characteristic spectrum of the AGEs-SERS complex was acquired using a 785nm laser with a power of 5mW in a backscattering geometry. The collected SERS characteristic spectra were baseline corrected and smoothed and denoised to obtain the preprocessed SERS characteristic spectra. Based on SERS characteristic spectra, multiple characteristic peaks are located by peak detection, and the integral intensity and peak height of each characteristic peak are calculated. The spectral characteristic parameters are obtained by calculating the relative ratio of the integral intensity to the peak height of the characteristic peak based on the integral intensity and peak height of the characteristic peak.
[0038] According to an embodiment of the present invention, a classification model is constructed based on a machine learning algorithm to analyze spectral feature parameters and obtain the DKD risk grading assessment result, specifically including: Urine samples were collected from patients with clinically diagnosed low, medium, and high risk levels of DKD. Obtain the spectral characteristic parameters and corresponding clinical parameters of urine samples, and construct a dataset containing sample feature matrix and risk level labels; The dataset was divided into a training set and a test set in a 7:3 ratio. Input the training set feature matrix and corresponding risk level labels into the model, and optimize the model hyperparameters by traversing parameter combinations through grid search. The trained model was evaluated using 5-fold cross-validation, and the evaluation results were obtained. Optimize the model hyperparameters based on the evaluation results until the model performance meets the target. The feature matrix of the test set or the sample to be tested is input into the model with qualified performance. The model analyzes the correlation between spectral feature parameters and DKD risk and outputs the DKD risk classification assessment result corresponding to the urine sample.
[0039] It should be noted that SERS signal acquisition and DKD risk assessment include: 1. 785nm laser excitation was used to collect characteristic spectra of AGEs (such as the 1600cm-1 aromatic ring vibration peak).
[0040] Optimization of the spectral acquisition system: A backscattering geometry was adopted, the laser power was controlled at 5mW to avoid thermal damage to the sample, and the spectral resolution was set to 4cm⁻¹; Specific implementation case: Ten AGEs standards of different concentrations (0.1-100 nM) were used for SERS detection, with five repeated spectra collected for each sample. The WT-ICA algorithm was applied to process the raw data, improving the signal-to-noise ratio from 8.2 to 23.5 at 1602 cm⁻¹.
[0041] 2. Establish the correlation between AGEs concentration and DKD (diabetic nephropathy) risk based on clinical samples. Use machine learning algorithms to train a classification model, transforming spectral data into risk assessment indicators. Establish a graded prediction system by integrating SERS spectral features with clinical parameters. Specific implementation includes the following key technologies: Feature engineering processing: The intensity ratios of three main peaks—1602 cm⁻¹ (C=C aromatic ring stretching), 1365 cm⁻¹ (CN stretching), and 1180 cm⁻¹ (COC glycosylation characteristics)—were extracted from the SERS spectrum as spectral fingerprints. in, : Relative Intensity Ratio of the i-th component, expressed as a percentage, represents the proportion of the signal intensity of the i-th component in the total signal, reflecting the relative content or contribution of the component in the system; This represents the integral intensity of the i-th characteristic peak; The total signal intensity of the three components in the system.
[0042] Construct a feature matrix by combining clinical indicators: Risk Level Classification: According to the KDIGO guidelines, DKD risk is divided into three levels: Low risk: eGFR≥60 and UACR<30; Medium risk: eGFR 30-59 or UACR 30-300; High risk: eGFR < 30 or UACR > 300; Model training and validation: The XGBoost algorithm is used to construct the classifier, and the objective function is: Where L(ϕ): the total loss function of the model, and ϕ: the set of all parameters of the ensemble model. Let cross-entropy be the loss function. The ensemble model's prediction for the i-th sample. : The true label of the i-th sample Single base learner Complexity penalty term, The number of leaf nodes. Leaf weight, : The squared L2 norm of the linear basis learner weight vector w.
[0043] Clinical decision support: Risk probability output module: Where Pk: the predicted probability that the feature vector x of the sample belongs to the k-th class. The output value for the k-th class model. The normalization denominator for the three categories is the sum of the exponential values of the original scores of the three categories (eF1(x)+eF2(x)+eF3(x)). Its core function is standardization, which transforms the exponential score of a single category into its proportion in the total score, ensuring that the sum of the probabilities of all categories is 1 (satisfying the probability axiom). In the denominator, c = 1-3 indicates that the task is a three-class classification (3 target categories). c is the category traversal index. If it is an n-class classification task, it can be expanded to ∑c=1neFc(x) (e.g., for binary classification, c = 1, 2, which is essentially a generalization of logistic regression).
[0044] Specific Implementation Case: Urine samples were collected from 326 diabetic patients (aged 45-72 years). AGEs concentration was determined by HPLC as the gold standard. SERS detection was combined with clinical parameters to construct a feature matrix, which was randomly divided into a 70% training set and a 30% test set. The XGBoost model was optimized using a grid search (max_depth=5, learning_rate=0.1, n_estimators=200). The model performed on the test set with an accuracy of 89.3% and an AUC of 0.923 (95% CI 0.891-0.955). Key feature contribution analysis showed that the 1365 cm⁻¹ peak intensity ratio (SHAP value 0.32) and eGFR (0.28) were the most predictive. The early warning accuracy for high-risk patients reached 92.7%.
[0045] like Figures 2-3 As shown, in a second aspect, embodiments of this application provide a DKD risk assessment system based on surface-enhanced Raman scattering. The system includes a memory and a processor. The memory includes a program for a DKD risk assessment method based on surface-enhanced Raman scattering. When the program for the DKD risk assessment method based on surface-enhanced Raman scattering is executed by the processor, it performs the following steps: Gold nanorods with a predetermined aspect ratio were prepared by seed growth method, and carboxyl or amino functional groups were introduced on the surface of the gold nanorods to obtain SERS active substrates. Urine samples were obtained, and AGEs molecules in the urine samples were separated and enriched based on centrifugation, solid-phase extraction and molecular sieve purification, and the filtrate was collected. The filtrate was incubated with a SERS active substrate under specific pH conditions to allow AGEs molecules to adsorb onto the surface of the SERS active substrate, thus obtaining an AGEs-SERS complex. Collect the SERS characteristic spectrum of the AGEs-SERS complex and extract the spectral characteristic parameters; A classification model is constructed based on machine learning algorithms, and spectral feature parameters are analyzed to obtain the DKD risk classification assessment results.
[0046] According to embodiments of the present invention, gold nanorods with a predetermined aspect ratio are prepared based on a seed growth method, and carboxyl or amino functional groups are introduced onto the surface of the gold nanorods to obtain a SERS-active substrate, specifically including: Prepare a 0.1 mM chloroauric acid aqueous solution as a precursor and add hexadecyltrimethylammonium bromide as a surfactant; Gold seeds of 2-5 nm are formed by injection of sodium borohydride solution to obtain gold seed solution; The gold seed solution was added to a growth solution containing chloroauric acid, hexadecyltrimethylammonium bromide, silver nitrate and ascorbic acid, and reacted at 25-30℃ for 12-24 hours. The concentration of silver nitrate was adjusted to 0.01-0.1mM to control the aspect ratio of the gold nanorods within the range of 2.5-4.0, thus synthesizing gold nanorods. The synthesized gold nanorods were purified by centrifugation and dispersed in a 1 mM solution of mercaptoundecanoic acid or mercaptoethylamine ethanol. The mixture was stirred at room temperature, and carboxyl or amino functional groups were modified on the surface of the gold nanorods to obtain the SERS active substrate.
[0047] According to an embodiment of the present invention, a urine sample is obtained, and AGEs molecules in the urine sample are separated and enriched based on centrifugation, solid-phase extraction, and molecular sieve purification, and the filtrate is collected. Specifically, the process includes: Obtain a urine sample, centrifuge the urine sample to obtain the supernatant, and then filter the supernatant through a micropore. The supernatant after microfiltration is fused and rinsed, and the supernatant after rinsing is then eluted. The supernatant after elution was concentrated to obtain the extract; The extract was purified by analytical sieve to obtain the filtrate.
[0048] A third aspect of the present invention provides a computer-readable storage medium including a DKD risk assessment method program based on surface-enhanced Raman scattering, wherein when the DKD risk assessment method program based on surface-enhanced Raman scattering is executed by a processor, the program implements the steps of the DKD risk assessment method based on surface-enhanced Raman scattering as described in any of the preceding claims.
[0049] This invention discloses a method, system, and medium for DKD risk assessment based on surface-enhanced Raman scattering (SERS). The method involves preparing gold nanorods with a predetermined aspect ratio using a seed-growth method, introducing carboxyl or amino functional groups onto the surface of the gold nanorods to obtain an SERS-active substrate. Urine samples are obtained, and AGEs molecules in the urine samples are separated and enriched using centrifugation, solid-phase extraction, and molecular sieve purification. The filtrate is collected. Under specific pH conditions, the filtrate is incubated with the SERS-active substrate, allowing AGEs molecules to adsorb onto the surface of the SERS-active substrate, resulting in an AGEs-SERS complex. The SERS characteristic spectrum of the AGEs-SERS complex is acquired, and spectral characteristic parameters are extracted. A classification model is constructed based on a machine learning algorithm, and the spectral characteristic parameters are analyzed to obtain the DKD risk grading assessment result. Through the high sensitivity and spectral characteristics of SERS technology, rapid and accurate detection of AGEs in urine is achieved, enabling clinicians to take intervention measures before significant kidney function damage occurs.
[0050] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.
[0051] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.
[0052] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.
[0053] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0054] Alternatively, if the integrated units of the present invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.
Claims
1. A risk assessment method for DKD based on surface-enhanced Raman scattering, characterized in that, include: Gold nanorods with a predetermined aspect ratio were prepared by seed growth method, and carboxyl or amino functional groups were introduced on the surface of the gold nanorods to obtain SERS active substrates. Urine samples were obtained, and AGEs molecules in the urine samples were separated and enriched based on centrifugation, solid-phase extraction and molecular sieve purification, and the filtrate was collected. The filtrate was incubated with a SERS active substrate under specific pH conditions to allow AGEs molecules to adsorb onto the surface of the SERS active substrate, thus obtaining an AGEs-SERS complex. The SERS characteristic spectrum of the AGEs-SERS complex was acquired, and the spectral characteristic parameters were extracted; A classification model is constructed based on machine learning algorithms, and spectral feature parameters are analyzed to obtain the DKD risk classification assessment results.
2. The DKD risk assessment method based on surface-enhanced Raman scattering according to claim 1, characterized in that, Gold nanorods with a predetermined aspect ratio were prepared using a seed growth method. Carboxyl or amino functional groups were then introduced onto the surface of the gold nanorods to obtain a SERS-active substrate, specifically including: Prepare a 0.1 mM chloroauric acid aqueous solution as a precursor and add hexadecyltrimethylammonium bromide as a surfactant; Gold seeds of 2-5 nm are formed by injection of sodium borohydride solution to obtain gold seed solution; The gold seed solution was added to a growth solution containing chloroauric acid, hexadecyltrimethylammonium bromide, silver nitrate and ascorbic acid, and reacted at 25-30℃ for 12-24 hours. The concentration of silver nitrate was adjusted to 0.01-0.1mM to control the aspect ratio of the gold nanorods within the range of 2.5-4.0, thereby synthesizing gold nanorods. The synthesized gold nanorods were purified by centrifugation and dispersed in a 1 mM solution of mercaptoundecanoic acid or mercaptoethylamine ethanol. The mixture was stirred at room temperature, and carboxyl or amino functional groups were modified on the surface of the gold nanorods to obtain the SERS active substrate.
3. The DKD risk assessment method based on surface-enhanced Raman scattering according to claim 2, characterized in that, Urine samples were obtained, and AGEs molecules in the urine samples were separated and enriched based on centrifugation, solid-phase extraction, and molecular sieve purification. The filtrate was then collected. Specifically, the process included: Obtain a urine sample, centrifuge the urine sample to obtain the supernatant, and then filter the supernatant through a micropore. The supernatant after microfiltration is fused and rinsed, and the supernatant after rinsing is then eluted. The supernatant after elution was concentrated to obtain the extract; The extract was purified by analytical sieve to obtain the filtrate.
4. The DKD risk assessment method based on surface-enhanced Raman scattering according to claim 3, characterized in that, The filtrate was incubated with a SERS-active substrate under specific pH conditions, allowing AGEs molecules to adsorb onto the surface of the SERS-active substrate, resulting in an AGEs-SERS complex, specifically comprising: The filtrate was mixed with gold nanorods at pH 5.5–6.
5. Incubate with shaking at a volume ratio of 1:50 for 30-40 minutes; Adding MgCl2 solution induces aggregation, causing AGEs molecules to adsorb onto the surface of gold nanorods via electrostatic interactions or covalent bonds, thus obtaining AGEs-SERS complexes.
5. The DKD risk assessment method based on surface-enhanced Raman scattering according to claim 4, characterized in that, The SERS characteristic spectra of the AGEs-SERS complex were collected, and the spectral characteristic parameters were extracted, specifically including: The SERS characteristic spectrum of the AGEs-SERS complex was acquired using a 785nm laser with a power of 5mW in a backscattering geometry. The collected SERS characteristic spectra were baseline corrected and smoothed and denoised to obtain the preprocessed SERS characteristic spectra. Based on SERS characteristic spectra, multiple characteristic peaks are located by peak detection, and the integral intensity and peak height of each characteristic peak are calculated. The spectral characteristic parameters are obtained by calculating the relative ratio of the integral intensity to the peak height of the characteristic peak based on the integral intensity and peak height of the characteristic peak.
6. The DKD risk assessment method based on surface-enhanced Raman scattering according to claim 5, characterized in that, A classification model is constructed based on machine learning algorithms, and spectral feature parameters are analyzed to obtain the DKD risk grading assessment results, specifically including: Urine samples were collected from patients with clinically diagnosed low, medium, and high risk levels of DKD. Obtain the spectral characteristic parameters and corresponding clinical parameters of urine samples, and construct a dataset containing sample feature matrix and risk level labels; The dataset was divided into a training set and a test set in a 7:3 ratio. Input the training set feature matrix and corresponding risk level labels into the model, and optimize the model hyperparameters by traversing parameter combinations through grid search. The trained model was evaluated using 5-fold cross-validation, and the evaluation results were obtained. Optimize the model hyperparameters based on the evaluation results until the model performance meets the target. The feature matrix of the test set or the sample to be tested is input into the model with qualified performance. The model analyzes the correlation between spectral feature parameters and DKD risk and outputs the DKD risk classification assessment result corresponding to the urine sample.
7. A DKD risk assessment system based on surface-enhanced Raman scattering, characterized in that, The system includes a memory and a processor. The memory contains a program for a DKD risk assessment method based on surface-enhanced Raman scattering (SERS). When the program for the DKD risk assessment method based on SERS is executed by the processor, it performs the following steps: Gold nanorods with a predetermined aspect ratio were prepared by seed growth method, and carboxyl or amino functional groups were introduced on the surface of the gold nanorods to obtain SERS active substrates. Urine samples were obtained, and AGEs molecules in the urine samples were separated and enriched based on centrifugation, solid-phase extraction and molecular sieve purification, and the filtrate was collected. The filtrate was incubated with a SERS active substrate under specific pH conditions to allow AGEs molecules to adsorb onto the surface of the SERS active substrate, thus obtaining an AGEs-SERS complex. The SERS characteristic spectrum of the AGEs-SERS complex was acquired, and the spectral characteristic parameters were extracted; A classification model is constructed based on machine learning algorithms, and spectral feature parameters are analyzed to obtain the DKD risk classification assessment results.
8. The DKD risk assessment system based on surface-enhanced Raman scattering according to claim 7, characterized in that, Gold nanorods with a predetermined aspect ratio were prepared using a seed growth method. Carboxyl or amino functional groups were then introduced onto the surface of the gold nanorods to obtain a SERS-active substrate, specifically including: Prepare a 0.1 mM chloroauric acid aqueous solution as a precursor and add hexadecyltrimethylammonium bromide as a surfactant; Gold seeds of 2-5 nm are formed by injection of sodium borohydride solution to obtain gold seed solution; The gold seed solution was added to a growth solution containing chloroauric acid, hexadecyltrimethylammonium bromide, silver nitrate and ascorbic acid, and reacted at 25-30℃ for 12-24 hours. The concentration of silver nitrate was adjusted to 0.01-0.1mM to control the aspect ratio of the gold nanorods within the range of 2.5-4.0, thereby synthesizing gold nanorods. The synthesized gold nanorods were purified by centrifugation and dispersed in a 1 mM solution of mercaptoundecanoic acid or mercaptoethylamine ethanol. The mixture was stirred at room temperature, and carboxyl or amino functional groups were modified on the surface of the gold nanorods to obtain the SERS active substrate.
9. The DKD risk assessment system based on surface-enhanced Raman scattering according to claim 8, characterized in that, Urine samples were obtained, and AGEs molecules in the urine samples were separated and enriched based on centrifugation, solid-phase extraction, and molecular sieve purification. The filtrate was then collected. Specifically, the process included: Obtain a urine sample, centrifuge the urine sample to obtain the supernatant, and then filter the supernatant through a micropore. The supernatant after microfiltration is fused and rinsed, and the supernatant after rinsing is then eluted. The supernatant after elution was concentrated to obtain the extract; The extract was purified by analytical sieve to obtain the filtrate.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a DKD risk assessment method program based on surface-enhanced Raman scattering, which, when executed by a processor, implements the steps of the DKD risk assessment method based on surface-enhanced Raman scattering as described in any one of claims 1 to 6.