An alzheimer's disease prediction method, system, device, and medium

Abstract

This invention relates to the field of medical data processing technology, and discloses an Alzheimer's disease prediction method, system, device, and medium, comprising: obtaining spectral data of a target group in a non-invasive manner; training an Alzheimer's disease prediction model based on the spectral data; the target group includes healthy individuals and individuals with Alzheimer's disease. This invention non-invasively obtains samples from the target group, utilizing the advantages of laser-induced breakdown spectroscopy (LAS) for simultaneous detection and analysis of all elements at high speed. It achieves qualitative or quantitative analysis of the chemical composition of material surfaces by measuring the emission spectrum of micro-plasma, accurately and rapidly obtaining spectral data of the target group. The Alzheimer's disease prediction model trained based on the spectral data can effectively assist doctors in early screening of individuals, enabling early intervention for patients with early-stage Alzheimer's disease, and has broad application prospects.

Description

Technical Field

[0001] This invention relates to the field of medical data processing technology, and in particular to a method, system, device and medium for predicting Alzheimer's disease. Background Technology

[0002] In 2020, 55 million people worldwide were diagnosed with dementia, and the number of patients is doubling every 20 years. Alzheimer's disease (AD), one of the major types of dementia, is a neurodegenerative disease without obvious tissue lesions. To date, there is no effective cure for AD; early diagnosis and timely intervention are the only effective measures to slow disease progression. AD is a continuous process comprising three stages: the asymptomatic stage (preclinical AD), the pre-dementia stage (mild cognitive impairment caused by AD, MCI), and the dementia stage (dementia caused by AD). The pathophysiological changes in AD begin 15 to 20 years before the appearance of clinical symptoms. Early prediction of AD allows for earlier diagnosis, buying time for timely intervention.

[0003] Currently, the primary method for diagnosing Alzheimer's disease is still behavioral and cognitive testing, which is characterized by delays and uncertainties. Detection of biomarkers can lead to earlier and more accurate diagnoses; however, biomarker testing is expensive and requires invasive procedures, which can potentially cause harm to the patient.

[0004] Therefore, the medical field urgently needs to develop non-invasive and rapid auxiliary diagnostic technologies for early screening of diseases such as Alzheimer's. Summary of the Invention

[0005] This invention provides a method, system, device, and medium for predicting Alzheimer's disease, which addresses the shortcomings of existing technologies in detecting Alzheimer's disease, such as lag, uncertainty, and high detection costs, and enables non-invasive and rapid assistance to doctors in early screening for Alzheimer's disease.

[0006] This invention provides a method for constructing an Alzheimer's disease prediction model, comprising:

[0007] Obtain spectral data of the target population in a non-invasive manner;

[0008] Based on spectral data, an Alzheimer's disease prediction model was trained.

[0009] The target groups include healthy individuals and those with Alzheimer's disease.

[0010] According to the present invention, a method for constructing an Alzheimer's disease prediction model, wherein obtaining spectral data of a target population in a non-invasive manner includes:

[0011] Collect excrement samples from the target group;

[0012] Based on excrement samples, spectral data of the target population were obtained using laser-induced breakdown spectroscopy.

[0013] According to the present invention, a method for constructing an Alzheimer's disease prediction model, wherein the Alzheimer's disease prediction model is trained based on spectral data, comprising:

[0014] The spectral data is preprocessed, including outlier processing and spectral area normalization.

[0015] An Alzheimer's disease prediction model was trained based on the preprocessed spectral data.

[0016] According to the method for constructing an Alzheimer's disease prediction model provided by the present invention, the abnormal data processing of spectral data includes:

[0017] The spectral difference was obtained based on the spectra of healthy individuals and individuals with Alzheimer's disease;

[0018] Identify anomalous spectral peaks based on spectral differences;

[0019] Based on abnormal spectral peaks, spectral lines are identified and assigned, and abnormal data is filtered out.

[0020] According to the method for constructing an Alzheimer's disease prediction model provided by the present invention, the spectral area normalization processing of the spectral data specifically includes:

[0021] The spectral area of ​​the spectral data is normalized using a normalization formula.

[0022] The normalization formula is as follows:

[0023]

[0024] In the normalization formula, x represents the spectral data after spectral area normalization. ij This represents the original spectral data in the i-th row and j-th column of a spectral matrix with m rows and n columns.

[0025] According to the present invention, a method for constructing an Alzheimer's disease prediction model, wherein the Alzheimer's disease prediction model is trained based on spectral data, comprising:

[0026] Spectral data are mapped using kernel functions;

[0027] Based on the mapped spectral data, a linear boundary is obtained to train an Alzheimer's disease prediction model.

[0028] According to the method for constructing an Alzheimer's disease prediction model provided by the present invention, the expression of the kernel function is:

[0029]

[0030] In the expression for the kernel function, K represents the kernel function, x i Let x represent the spectral vector in the i-th row. j Represents the spectral vector of the j-th column. Represents a mapping;

[0031] The expression for the linear boundary is:

[0032]

[0033]

[0034]

[0035] In the expression for the linear boundary, x test Let b represent the vector to be measured, and α represent the intercept. i The Lagrange multiplier corresponding to the row vector, α j The Lagrange multiplier corresponding to the column vector, y i Let y represent the label vector of the i-th row. j Let X represent the label vector of the j-th column, and let X represent the overall spectral matrix. i Let x represent the spectral vector in the i-th row. j This represents the spectral vector of the j-th column.

[0036] This invention also provides a system for constructing an Alzheimer's disease prediction model, comprising:

[0037] A data receiving module, configured to receive spectral data of a target population obtained in a non-invasive manner;

[0038] The model training module is configured to train an Alzheimer's disease prediction model based on the spectral data received by the data receiving module.

[0039] The target groups include healthy individuals and individuals with Alzheimer's disease.

[0040] Those skilled in the art will understand that the Alzheimer's disease prediction model construction system of the present invention can be implemented in software, hardware, or a combination of both. When implemented in hardware, the data receiving module is electrically connected to the model training module. When implemented in a combination of software and hardware, the model training module can be implemented by the processor of a hardware device such as a host computer, industrial control computer, or personal computer executing logical instructions stored in memory or calling program instructions. In this case, the data receiving module can be built into the hardware device or be a separate hardware device that transmits data through communication with the hardware device (e.g., wireless or wired communication). When implemented in software, data can be manually input through an input device or obtained from a data acquisition device through an I / O interface, with the processor handling the data reception and model training processes.

[0041] The present invention also provides an early screening system for Alzheimer's disease, comprising:

[0042] A data receiving module, configured to receive spectral data of a subject obtained in a non-invasive manner;

[0043] The prediction module is configured to: obtain the Alzheimer's disease prediction result of the test subject based on the spectral data of the test subject received by the data receiving module and the Alzheimer's disease prediction model obtained by the construction method of the Alzheimer's disease prediction model described above;

[0044] The subjects to be tested include healthy individuals, suspected Alzheimer's patients in the asymptomatic or pre-dementia stage, or patients awaiting exclusion of Alzheimer's disease.

[0045] Those skilled in the art will understand that the Alzheimer's disease prediction system of the present invention can be implemented in software, hardware, or a combination of both. When implemented in hardware, the data receiving module is electrically connected to the prediction module. When implemented in a combination of software and hardware, the prediction module can be implemented by the processor of a hardware device such as a host computer, industrial control computer, or personal computer executing logical instructions stored in memory or calling program instructions. In this case, the data receiving module can be built into the hardware device or be a separate hardware device that transmits data through communication with the hardware device (e.g., wireless or wired communication). When implemented in software, data can be manually input through an input device or obtained from a data acquisition device through an I / O interface, with the processor handling the data reception and prediction process.

[0046] The present invention also provides an electronic device, including a processor and a memory storing a computer program, characterized in that the processor executes the computer program to perform the following steps:

[0047] Receive spectral data of the subject obtained in a non-invasive manner;

[0048] Based on the received spectral data of the test subject, the Alzheimer's disease prediction result of the test subject is obtained by constructing the Alzheimer's disease prediction model through any of the above-mentioned methods.

[0049] The subjects to be tested include healthy individuals, suspected Alzheimer's patients in the asymptomatic or pre-dementia stage, or patients awaiting exclusion of Alzheimer's disease.

[0050] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0051] Receive spectral data of the subject obtained in a non-invasive manner;

[0052] Based on the received spectral data of the test subject, the Alzheimer's disease prediction result of the test subject is obtained by constructing the Alzheimer's disease prediction model through any of the above-mentioned methods.

[0053] The subjects to be tested include healthy individuals, suspected Alzheimer's patients in the asymptomatic or pre-dementia stage, or patients awaiting exclusion of Alzheimer's disease.

[0054] The present invention also provides a computer program product, the computer program product comprising a computer program, the computer program being able to be stored on a non-transitory computer-readable storage medium, and the computer program being executed by a processor, enabling the computer to perform the following steps:

[0055] Receive spectral data of the subject obtained in a non-invasive manner;

[0056] Based on the received spectral data of the test subject, the Alzheimer's disease prediction result of the test subject is obtained by constructing the Alzheimer's disease prediction model through any of the above-mentioned methods.

[0057] The subjects to be tested include healthy individuals, suspected Alzheimer's patients in the asymptomatic or pre-dementia stage, or patients awaiting exclusion of Alzheimer's disease.

[0058] In this invention, Alzheimer's disease can be Alzheimer's disease related to or caused by brain-gut axis dysregulation.

[0059] This invention provides an Alzheimer's disease prediction method, system, device, and medium. It non-invasively obtains samples from a target population (e.g., excrement samples) using laser-induced breakdown spectroscopy (LIBS), leveraging its advantages of simultaneous, high-speed, and comprehensive elemental detection and analysis. By measuring the emission spectrum of micro-plasma, it achieves qualitative or quantitative analysis of the chemical composition of material surfaces, accurately and rapidly obtaining spectral data for the target population. The Alzheimer's disease prediction model trained based on this spectral data can effectively assist doctors in early screening of individuals, enabling early and effective intervention for Alzheimer's patients and slowing the progression of the disease. This invention has broad application prospects in the medical field and can bring significant economic and social value. Attached Figure Description

[0060] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0061] Figure 1 This is a flowchart illustrating a method for constructing an Alzheimer's disease prediction model provided by the present invention.

[0062] Figure 2 The LIBS spectral differences between feces from healthy mice and mice with Alzheimer's disease are shown, where wavelength represents the wavelength.

[0063] Figure 3 The performance evaluation of the Alzheimer's disease prediction model trained by the present invention is shown, where confusionmatrix represents the confusion matrix.

[0064] Figure 4 This is a schematic diagram of the structure of a system for constructing an Alzheimer's disease prediction model provided by the present invention.

[0065] Figure 5 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0066] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, embodiments of this invention, and should not be construed as limiting the invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention. In the description of this invention, it should be understood that the terminology used is for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0067] The following is combined Figures 1-5 This invention describes the Alzheimer's disease prediction method, system, device, and medium provided by the present invention.

[0068] Figure 1 This is a flowchart illustrating the method for constructing the Alzheimer's disease prediction model provided by this invention. (Refer to...) Figure 1 The present invention provides a method for constructing an Alzheimer's disease prediction model, which may include:

[0069] Step S110: Obtain the spectral data of the target population in a non-invasive manner;

[0070] Step S120: Based on spectral data, train an Alzheimer's disease prediction model;

[0071] It should be noted that the target group includes both healthy individuals and individuals with Alzheimer's disease. The target group can be animals, preferably mammals, including laboratory animals such as rodents (e.g., mice, rats, etc.), canines, equines, bovines, felines, primates, and humans. More preferably, rodents, primates, and humans are preferred, with commonly used laboratory rodents (e.g., mice, rats, etc.) and humans being the most preferred. When using the Alzheimer's disease prediction model, the corresponding prediction target should be consistent with the target used to train the Alzheimer's disease prediction model.

[0072] In one embodiment, step S110 may include:

[0073] Collect excrement samples from the target group;

[0074] Based on excrement samples, spectral data of the target population were obtained using laser-induced breakdown spectroscopy.

[0075] It should be noted that excrement samples can be feces, urine, saliva, or sweat, etc., and can be obtained without invasively penetrating the target group's body. Obtaining samples in a non-invasive manner can protect the target group's physical integrity, avoid causing damage to the target group's body, and improve the accuracy of spectral data.

[0076] It should be noted that laser-induced breakdown spectroscopy (LIBS) utilizes a focused pulsed laser to ablate a localized material surface, generating micro-plasma. Qualitative or quantitative analysis of the material's surface chemical composition is achieved by measuring the emission spectrum of this micro-plasma. As an emerging emission spectroscopy technique, LIBS offers unique advantages such as requiring no sample pretreatment, enabling simultaneous detection and analysis of all elements, and boasting rapid speed. It holds broad application prospects in the biomedical field.

[0077] Laser-induced breakdown spectroscopy can reflect the content and variation patterns of various chemical components in excrement samples. The types of blood elements are the same in healthy individuals and those with Alzheimer's disease, but the levels of elements such as Mg, Na, and Ca are higher in the Alzheimer's group, resulting in higher corresponding spectral intensities. Figure 2 The image shows the difference between the average spectrum of fecal samples from mice with Alzheimer's disease and those from healthy mice, revealing significant differences in the spectral intensities of the macroelements Na, Mg, and Ca. Training an Alzheimer's disease prediction model using the target population's attribute labels and the obtained spectral data can improve prediction accuracy. In this embodiment, the spectral data can be direct spectral peaks or other forms of spectral data, such as fitted peak height / area or peak ratios.

[0078] Specifically, when collecting excrement samples from the target group, fecal samples can be taken, and the target group's attribute labels can be recorded. Each fecal sample is compacted to the same thickness (e.g., 1 mm) and allowed to air dry at room temperature (or oven-dried) to maintain a roughly uniform moisture content (e.g., within a preset range). Then, the laser is triggered and adjusted to ensure a good spectral signal and no continuous background (adjusting parameters include laser energy, optimized integration time, and delay time). A focused pulsed laser is used to ablate the surface of the fecal sample, generating plasma, while simultaneously scanning each sample to collect spectral data from different locations. After obtaining the spectral data, the spectral data and corresponding target group attribute labels can be divided into training and testing sets. The training set is used for training the Alzheimer's disease prediction model, and the testing set is used for testing the Alzheimer's disease prediction model. When training the Alzheimer's disease prediction model, sampling can be conducted throughout the entire disease cycle to establish prediction models for different stages of Alzheimer's disease, assisting doctors in diagnosing early, middle, and late-stage Alzheimer's disease.

[0079] In one embodiment, step S120 may include:

[0080] The spectral data is preprocessed, including outlier processing and spectral area normalization.

[0081] The preprocessed spectral data is mapped using kernel functions;

[0082] Based on the mapped spectral data, a linear boundary is obtained to train an Alzheimer's disease prediction model.

[0083] In one embodiment, abnormal data processing of spectral data can be achieved through the following steps:

[0084] The spectral difference was obtained based on the spectra of healthy individuals and individuals with Alzheimer's disease;

[0085] Identify anomalous spectral peaks based on spectral differences;

[0086] Based on abnormal spectral peaks, spectral lines are identified and assigned, and abnormal data is filtered out.

[0087] Abnormal data can occur in both healthy and diseased populations, mainly due to significant data fluctuations, such as the absence of normal data or the collection of noise. Specifically, the spectra of the healthy population and the Alzheimer's disease population are compared as a whole to obtain the spectral difference. If the spectral difference reaches a strong intensity of 10% or more of the original spectral peak (the original spectral peak can be from either the healthy population or the Alzheimer's disease population, with relatively small differences between them), it can be identified as an abnormal spectral peak. Comparing the spectral difference with the original spectral peak helps to identify the location of the abnormal peak and filter out abnormal data.

[0088] In one embodiment, the spectral data is normalized by spectral area, specifically as follows:

[0089] The spectral area of ​​the spectral data is normalized using a normalization formula.

[0090] The normalization formula is as follows:

[0091]

[0092] In the normalization formula, x represents the spectral data after spectral area normalization. ij This represents the raw spectral data (e.g., raw spectral intensity) in the i-th row and j-th column of an m-row, n-column spectral matrix.

[0093] In one embodiment, the expression for the kernel function is:

[0094]

[0095] In the expression for the kernel function, K represents the kernel function, x i Let x represent the spectral vector in the i-th row. j Represents the spectral vector of the j-th column. Represents a mapping;

[0096] When calling the sklearn.svm.SVC() class, the expression for the linear boundary can be:

[0097]

[0098]

[0099] In the expression for the linear boundary, x test Let b represent the vector to be measured, and α represent the intercept. i The Lagrange multiplier corresponding to the row vector, α j The Lagrange multiplier corresponding to the column vector, y i Let y represent the label vector of the i-th row. j Let X represent the label vector of the j-th column, and let X represent the overall spectral matrix. i Let x represent the spectral vector in the i-th row. j This represents the spectral vector in the j-th column, without any other hyperparameter adjustments.

[0100] This embodiment employs a Support Vector Machine (SVM) as the function for training the Alzheimer's disease prediction model. It utilizes a kernel function to map spectral data to a higher-dimensional feature space, facilitating the identification of classification boundaries to distinguish between healthy individuals and those with Alzheimer's disease. This allows for faster and more accurate prediction of whether a candidate has Alzheimer's. Upon completion of the Alzheimer's prediction model training, it is tested using a test set. The prediction results are compared with actual results. If prediction errors are found, the erroneous data is added to the training set, and the Alzheimer's prediction model is iteratively trained again until its predictive performance stabilizes (e.g., the prediction accuracy reaches a certain preset threshold, indicating stable performance).

[0101] Based on the method for constructing the Alzheimer's disease prediction model provided by this invention, a specific embodiment will be provided below.

[0102] Several laboratory mice (including 10 healthy mice (HC) and 10 mice with Alzheimer's disease (AD)) were purchased from Cavensburg (Suzhou) Model Animal Research Co., Ltd. (3×TgAD). The mice were housed in pairs in a 12-hour light-dark cycle (light on from 7:00 AM to 7:00 PM), with a relative humidity of 53% and a temperature of 24±1℃. They had free access to water (RO purified water) and food (maintained formulated feed for both mice and rats). The bedding was cleaned and changed weekly. After a period of rearing, the mice were stimulated to defecate by grasping the scruff of their necks. Ten fecal pellets were collected from each of the healthy and Alzheimer's mice (one pellet per mouse). The fresh fecal pellets were collected directly into 1.5 ml Eppendorf tubes. All work areas and experimental equipment were disinfected with 75% ethanol. The dried fecal particles were then placed on a three-dimensional translation stage, and the sample height was adjusted to the focal point of the LIBS system. The pulsed laser of the LIBS system was focused onto the surface of the fecal particles, ablating the surface to generate plasma. Simultaneously, the spectrometer parameters and laser energy were adjusted until a good spectral signal was obtained without a continuous background. The optimized laser energy was 50 mJ (power density of 45 GW / cm²), the spectrometer integration time was 5 μs, and the delay time was 0.3 μs. Under the optimized acquisition parameters, the fecal particles were translated and their positions were scanned, acquiring 100 LIBS spectra at different locations, resulting in a total of 2000 spectral data. Anomaly processing and area normalization were then performed on the spectral data. The preprocessed spectral data from 16 fecal particles (8 normal and 8 diseased, totaling 1600 LIBS spectra) were used as the training set. An Alzheimer's disease prediction model was established using a support state vector machine algorithm (linear kernel function). The model was then used to predict the preprocessed spectral data (400 LIBS spectra in total) of the remaining 4 fecal pellets to obtain a model performance evaluation, such as... Figure 3 As shown, the Alzheimer's disease prediction model trained in this embodiment has a prediction accuracy of 0.93, a precision of 0.91, a sensitivity / recall of 0.96, and a specificity of 0.9, demonstrating excellent performance across all metrics.

[0103] This invention provides an Alzheimer's disease prediction method, system, device, and medium. It non-invasively obtains samples from a target group (e.g., excrement samples) using a non-invasive method. Leveraging the advantages of laser-induced breakdown spectroscopy (LIBS)—simultaneous detection and analysis of all elements at high speed—it achieves qualitative or quantitative analysis of the chemical composition of material surfaces by measuring the emission spectrum of micro-plasma. This provides accurate and rapid spectral data for the target group. The Alzheimer's disease prediction model trained based on this spectral data can effectively assist doctors in early diagnosis and intervention, enabling timely and effective treatment for Alzheimer's patients. This invention has broad application prospects in the medical field and can bring significant economic and social value.

[0104] The following describes the construction system for the Alzheimer's disease prediction model provided by this invention. The construction system for the Alzheimer's disease prediction model described below can be referred to in correspondence with the construction method for the Alzheimer's disease prediction model described above.

[0105] Reference Figure 4 The present invention provides a system for constructing an Alzheimer's disease prediction model, which may include:

[0106] A data receiving module, configured to receive spectral data of a target population obtained in a non-invasive manner;

[0107] The model training module is configured to train an Alzheimer's disease prediction model based on the spectral data received by the data receiving module.

[0108] The target groups include healthy individuals and individuals with Alzheimer's disease.

[0109] The present invention also provides an Alzheimer's disease prediction system, which may include:

[0110] A data receiving module, configured to receive spectral data of a subject obtained in a non-invasive manner;

[0111] The prediction module is configured to: obtain the Alzheimer's disease prediction result of the test subject based on the spectral data of the test subject received by the data receiving module and the Alzheimer's disease prediction model obtained by the construction method of the Alzheimer's disease prediction model described above;

[0112] The subjects to be tested include healthy individuals, suspected Alzheimer's patients in the asymptomatic or pre-dementia stage, or patients awaiting exclusion of Alzheimer's disease.

[0113] It should be noted that the test subject can be an animal, preferably a mammal, including rodents, canines, equines, bovines, felines, primates, and humans, with primates and humans being more preferred.

[0114] Figure 5 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 5 As shown, the electronic device may include a processor 810, a communications interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communications interface 820, and the memory 830 communicate with each other via the communication bus 840. The processor 810 can call logical instructions in the memory 830 to execute the following steps:

[0115] Receive spectral data of the subject obtained in a non-invasive manner;

[0116] Based on the received spectral data of the test subject, the Alzheimer's disease prediction model obtained through the Alzheimer's disease prediction model construction method described above is used to obtain the Alzheimer's disease prediction result of the test subject;

[0117] The subjects to be tested include healthy individuals, suspected Alzheimer's patients in the asymptomatic or pre-dementia stage, or patients awaiting exclusion of Alzheimer's disease.

[0118] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer 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 steps 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 USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0119] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program, the computer program being able to be stored on a non-transitory computer-readable storage medium, and the computer program being executed by a processor, enabling the computer to perform the following steps:

[0120] Receive spectral data of the subject obtained in a non-invasive manner;

[0121] Based on the received spectral data of the test subject, the Alzheimer's disease prediction model obtained through the Alzheimer's disease prediction model construction method described above is used to obtain the Alzheimer's disease prediction result of the test subject;

[0122] The subjects to be tested include healthy individuals, suspected Alzheimer's patients in the asymptomatic or pre-dementia stage, or patients awaiting exclusion of Alzheimer's disease.

[0123] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0124] Receive spectral data of the subject obtained in a non-invasive manner;

[0125] Based on the received spectral data of the test subject, the Alzheimer's disease prediction model obtained through the Alzheimer's disease prediction model construction method described above is used to obtain the Alzheimer's disease prediction result of the test subject;

[0126] The subjects to be tested include healthy individuals, suspected Alzheimer's patients in the asymptomatic or pre-dementia stage, or patients awaiting exclusion of Alzheimer's disease.

[0127] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0128] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0129] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for constructing an Alzheimer's disease prediction model, characterized in that, include: Obtain spectral data of the target population in a non-invasive manner; Based on spectral data, an Alzheimer's disease prediction model was trained. The target groups include healthy individuals and individuals with Alzheimer's disease; The method of obtaining the spectral data of the target population in a non-invasive manner includes: Collect excrement samples from the target group; Based on excrement samples, spectral data of the target population were obtained using laser-induced breakdown spectroscopy. The Alzheimer's disease prediction model trained based on spectral data includes: The spectral data is preprocessed, including outlier processing and spectral area normalization. Based on the preprocessed spectral data, an Alzheimer's disease prediction model was trained. Specifically, the spectral area normalization process for the spectral data includes: The spectral area of ​​the spectral data is normalized using a normalization formula. The normalization formula is as follows: ， In the normalization formula, This represents the spectral data after spectral area normalization. This represents the original spectral data in the i-th row and j-th column of an m-row, n-column spectral matrix; The Alzheimer's disease prediction model trained based on spectral data includes: Spectral data are mapped using kernel functions; Based on the mapped spectral data, a linear boundary is obtained to train an Alzheimer's disease prediction model; The expression for the kernel function is as follows: ， In the expression of the kernel function, Represents the kernel function. Represents the spectral vector of the i-th row. Represents the spectral vector of the j-th column. Represents a mapping; The expression for the linear boundary is: ， ， ， In the expression for the linear boundary, Let b represent the vector to be measured, and let b represent the intercept. Corresponding to the Lagrange multiplier of the row vector, Corresponding to the Lagrange multiplier of column vectors, Let represent the label vector of the i-th row. Let X represent the label vector of the j-th column, and let X represent the overall spectral matrix. Represents the spectral vector of the i-th row. This represents the spectral vector of the j-th column.

2. The method for constructing an Alzheimer's disease prediction model according to claim 1, characterized in that, The abnormal data processing of the spectral data includes: The spectral difference was obtained based on the spectra of healthy individuals and individuals with Alzheimer's disease; Identify anomalous spectral peaks based on spectral differences; Based on abnormal spectral peaks, spectral lines are identified and assigned, and abnormal data is filtered out.

3. A system for constructing an Alzheimer's disease prediction model, characterized in that, include: A data receiving module, configured to receive spectral data of a target population obtained in a non-invasive manner; The model training module is configured to train an Alzheimer's disease prediction model based on the spectral data received by the data receiving module. The target groups include healthy individuals and individuals suffering from Alzheimer's disease; The method of obtaining the spectral data of the target population in a non-invasive manner includes: Collect excrement samples from the target group; Based on excrement samples, spectral data of the target population were obtained using laser-induced breakdown spectroscopy. The Alzheimer's disease prediction model trained based on spectral data includes: The spectral data is preprocessed, including outlier processing and spectral area normalization. Based on the preprocessed spectral data, an Alzheimer's disease prediction model was trained. Specifically, the spectral area normalization process for the spectral data includes: The spectral area of ​​the spectral data is normalized using a normalization formula. The normalization formula is as follows: ， In the normalization formula, This represents the spectral data after spectral area normalization. This represents the original spectral data in the i-th row and j-th column of an m-row, n-column spectral matrix; The Alzheimer's disease prediction model trained based on spectral data includes: Spectral data are mapped using kernel functions; Based on the mapped spectral data, a linear boundary is obtained to train an Alzheimer's disease prediction model; The expression for the kernel function is as follows: ， In the expression of the kernel function, Represents the kernel function. Represents the spectral vector of the i-th row. Represents the spectral vector of the j-th column. Represents a mapping; The expression for the linear boundary is: ， ， ， In the expression for the linear boundary, Let b represent the vector to be measured, and let b represent the intercept. Corresponding to the Lagrange multiplier of the row vector, Corresponding to the Lagrange multiplier of column vectors, Let represent the label vector of the i-th row. Let X represent the label vector of the j-th column, and let X represent the overall spectral matrix. Represents the spectral vector of the i-th row. This represents the spectral vector of the j-th column.

4. An early screening system for Alzheimer's disease, characterized in that, include: A data receiving module, configured to receive spectral data of a subject obtained in a non-invasive manner; The prediction module is configured to: obtain the Alzheimer's disease prediction result of the test subject based on the spectral data of the test subject received by the data receiving module and the Alzheimer's disease prediction model obtained by the construction method of the Alzheimer's disease prediction model as described in claim 1 or 2; The subjects to be tested include healthy individuals, suspected Alzheimer's patients in the asymptomatic or pre-dementia stage, or patients awaiting exclusion of Alzheimer's disease.

5. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor performs the following steps: Receive spectral data of the subject obtained in a non-invasive manner; Based on the received spectral data of the test subject, the Alzheimer's disease prediction model obtained by the construction method of the Alzheimer's disease prediction model as described in claim 1 or 2 is used to obtain the Alzheimer's disease prediction result of the test subject; The subjects to be tested include healthy individuals, suspected Alzheimer's patients in the asymptomatic or pre-dementia stage, or patients awaiting exclusion of Alzheimer's disease.

Images