Osteoporosis diagnosis method and system based on multi-omics combined machine learning

Abstract

The invention discloses an osteoporosis diagnosis method and system based on multi-omics joint machine learning, relates to the field of artificial intelligence medicine, and is particularly designed for the requirements of pilot groups. The method comprises the following steps: acquiring multiple omics data of bone mass reduction and osteoporosis states, screening out a differential expression biomarker set, and performing gene function enrichment analysis to identify a target biomarker with diagnostic significance. Furthermore, a osteoporosis diagnosis model is constructed based on the biomarkers, and diagnosis information of a target user is generated by using the model. According to the method, the physiological features of the pilot and multi-omics data are combined, personalized and accurate osteoporosis risk prediction and diagnosis are achieved, and the method is suitable for health management of high-intensity occupational groups such as the pilot.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence medicine, and in particular to an osteoporosis diagnosis method and system based on multi-omics combined with machine learning. Background Art

[0002] With the aging of the population, the incidence of osteoporosis has increased rapidly, which has brought severe challenges to the health management of special occupational groups, especially pilots and other high-intensity occupational groups. Osteoporosis can significantly increase the risk of fractures, affecting the physical health and occupational safety of pilots. Therefore, early screening and accurate prediction of osteoporosis are of great significance to protecting the occupational health of pilots. At present, the diagnosis of osteoporosis usually relies on bone density detection methods, such as dual-energy X-ray absorptiometry (DXA) and quantitative CT (QCT), but these methods require professional equipment and personnel, and it is difficult to achieve high-frequency and wide-range screening. Based on this, osteoporosis risk prediction methods for specific populations such as pilots need to be explored urgently to achieve early detection and risk control.

[0003] For example, the Chinese patent with the authorization announcement number CN115775628A discloses a method for predicting osteoporosis risk based on a relationship network. This method collects the subject's physiological indicators and questionnaire data, builds a risk prediction model based on the relationship network, and uses network features combined with algorithms such as Logistic regression, SVM, and random forest to predict osteoporosis risk.

[0004] The above method has the problems raised by the background technology. The method is not effective for some specific groups, and its network construction and analysis rely on a fixed network feature construction method, which is difficult to adapt to the special physiological characteristics and occupational needs of the pilot group. Summary of the invention

[0005] In response to the deficiencies of the prior art, the present application discloses an osteoporosis diagnosis method and system based on multi-omics combined with machine learning.

[0006] In the first aspect, the present application provides a method for diagnosing osteoporosis based on multi-omics combined with machine learning, comprising:

[0007] Obtain multi-omics data of the user to be diagnosed;

[0008] Based on the multi-omics data, a preset osteoporosis diagnostic model is used to diagnose bone status and generate an osteopenia risk score; the osteopenia risk score is used for osteoporosis diagnosis;

[0009] The osteoporosis diagnostic model is constructed based on representative differentially expressed biomarkers, which are obtained by performing gene function enrichment analysis on a set of differentially expressed biomarkers, and the set of differentially expressed biomarkers is obtained by screening from a data set of multi-omics data including bone loss and osteoporosis status.

[0010] As an optional implementation, the multi-omics data includes: multi-omics data of a pilot group; the method further includes:

[0011] Based on the public database, a gene interaction network is constructed; wherein the nodes in the gene interaction network represent genes, and the edges represent the interactions between genes; the interactions include: activation, inhibition, and physical binding;

[0012] The physiological data of the pilot group is obtained, and the physiological data of the pilot group is used as an additional feature of the node and / or edge.

[0013] As an optional embodiment, the enrichment analysis includes:

[0014] The preset multimodal attention enhancement network is used to perform gene function enrichment analysis on the differentially expressed biomarker set to determine representative differentially expressed biomarkers in the construction of osteoporosis diagnostic model.

[0015] As an optional implementation, the multimodal attention enhancement network includes: a graph neural network layer, an attention enhancement layer, and a multimodal fusion layer;

[0016] The graph neural network layer is used to model the graph structure of the gene interaction network and extract the feature representation of each gene node through graph convolution operation;

[0017] The attention enhancement layer is used to dynamically adjust the edge weights between gene nodes based on the pilot's physiological data;

[0018] The multimodal fusion layer is used to fuse the feature representation from the multi-omics data with the pilot's physiological data to generate a comprehensive multimodal feature vector.

[0019] As an optional implementation, the dynamically adjusting edge weights between gene nodes based on the pilot's physiological data includes:

[0020] Using a multi-layer perceptron, the standardized physiological data of the pilot is converted into a low-dimensional embedding vector; wherein the embedding vector has the same dimension as the gene node features in the gene interaction network;

[0021] Based on the preset additive attention mechanism, the attention weights between gene nodes are calculated;

[0022] Based on the attention weights, adjusting edge weights between gene nodes in the gene interaction network;

[0023] Using the adjusted edge weights, we screened out gene node pairs with edge weights increased by more than 20% to identify key genes associated with osteopenia;

[0024] Biological pathway enrichment analysis was performed on the screened key genes related to osteopenia to determine the biological pathways in which they were located.

[0025] As an optional implementation, the calculation of attention weights between gene nodes based on a preset additive attention mechanism includes:

[0026] The pilot’s physiological feature embedding vector is input into a fully connected layer to generate a query vector;

[0027] Input the feature vectors of gene nodes into independent fully connected layers to generate key vectors and value vectors;

[0028] Generate an attention score by calculating the similarity between the query vector and each key vector;

[0029] Apply the softmax function to all attention scores to get the normalized attention weights;

[0030] Multiply the normalized attention weights by the corresponding value vector to generate the weighted gene feature representation.

[0031] As an optional implementation, generating an attention score by calculating the similarity between the query vector and each key vector includes:

[0032] After linearly transforming the query vector and the key vector, the similarity score is calculated through a hyperbolic tangent activation function and a trainable weight matrix.

[0033] As an optional implementation, adjusting the edge weights between gene nodes in the gene interaction network based on the attention weights includes:

[0034] For gene node pairs with attention weights exceeding 0.8, their edge weights are increased based on the attention weights, and the increase in edge weights is greater than 20% of the original weights;

[0035] For gene-node pairs with attention weights lower than 0.2, their edge weights are reduced based on the attention weights, and the reduction in edge weights is less than 10% of the original weights.

[0036] In the second aspect, the present application also provides an osteoporosis diagnosis system based on multi-omics combined with machine learning, including:

[0037] A collection unit, used to obtain multi-omics data of the user to be diagnosed;

[0038] A diagnostic unit, configured to diagnose bone status based on the multi-omics data using a preset osteoporosis diagnostic model, and generate an osteopenia risk score; the osteopenia risk score is used for osteoporosis diagnosis;

[0039] The osteoporosis diagnostic model is constructed based on representative differentially expressed biomarkers, which are obtained by performing gene function enrichment analysis on a set of differentially expressed biomarkers, and the set of differentially expressed biomarkers is obtained by screening from a data set of multi-omics data including bone loss and osteoporosis status.

[0040] Compared with the prior art, the present invention has the following beneficial effects: based on the joint analysis of the pilot's physiological data and multi-omics data, the present invention realizes the effective fusion of individual physiological characteristics and genetic characteristics by constructing a multimodal attention enhancement network. Compared with the traditional diagnostic method based on a single-omics data or conventional classification model, the present invention can dynamically adjust the weight of the gene node, so that the model can more sensitively capture the individual's osteoporosis risk, significantly improving the personalization and accuracy of diagnosis.

[0041] The present invention innovatively constructs a multimodal attention enhancement network structure including a graph neural network, an attention enhancement layer, and a multimodal fusion layer. Compared with traditional neural networks, this structure can establish complex feature interaction relationships based on multi-level data, and effectively focus on the impact of different modal information on diagnostic results, thereby improving the robustness and flexibility of the diagnostic model under complex data conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 A flowchart of an osteoporosis diagnosis method based on multi-omics combined with machine learning provided in an embodiment of the present disclosure;

[0043] Figure 2 A flow chart of a method for calculating attention weights between gene nodes provided in an embodiment of the present disclosure;

[0044] Figure 3 Schematic diagram of an osteoporosis diagnosis system based on multi-omics combined with machine learning provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0045] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0046] See also Figure 1 , Figure 1 This is a flow chart of an osteoporosis diagnosis method based on multi-omics combined with machine learning provided in an embodiment of the present disclosure, the method comprising steps S101 to S102, wherein:

[0047] S101: Acquire multi-omics data of the user to be diagnosed;

[0048] S102: Based on the multi-omics data, using a preset osteoporosis diagnosis model, diagnose the bone status and generate an osteopenia risk score; the osteopenia risk score is used for osteoporosis diagnosis;

[0049] The osteoporosis diagnostic model is constructed based on representative differentially expressed biomarkers, which are obtained by performing gene function enrichment analysis on a set of differentially expressed biomarkers, and the set of differentially expressed biomarkers is obtained by screening from a data set of multi-omics data including bone loss and osteoporosis status.

[0050] It should be emphasized that the diagnosis in the present disclosure only involves data processing and does not involve the modification of living bodies. For example, the osteoporosis diagnostic model is used to diagnose bone status and output a bone loss risk score. The specific processing process is to perform model calculations based on multi-omics data to generate a bone loss risk score, which can be used for subsequent auxiliary osteoporosis diagnosis. It can be seen that the method steps of this application do not include any disease treatment content itself.

[0051] The present invention provides an osteoporosis diagnosis method based on multi-omics combined with machine learning, which aims to diagnose the user's bone status by comprehensively analyzing the multi-omics data of the user to be diagnosed, using a preset osteoporosis diagnosis model, and generate an osteopenia risk score. The risk score is used to assess the user's risk of osteoporosis, thereby providing a scientific basis for early prevention and intervention.

[0052] In a specific implementation, first, the multi-omics data of the user to be diagnosed is obtained. The multi-omics data includes genomic data, transcriptomic data, proteomic data and metabolomic data. These data can be obtained by modern biotechnology means such as high-throughput sequencing technology and mass spectrometry analysis.

[0053] For example, genomic data can obtain the user's DNA sequence information through whole genome sequencing (WGS); transcriptomics data can obtain the user's gene expression level through RNA sequencing (RNA-Seq); proteomics data can use liquid chromatography-mass spectrometry (LC-MS / MS) to detect protein expression in serum or tissue samples; metabolomics data can analyze the concentration and type of metabolites in the user's body through nuclear magnetic resonance (NMR) or mass spectrometry technology.

[0054] After acquiring multi-omics data, the data is preprocessed and normalized. Exemplarily, the raw data can be first quality controlled to remove low-quality reads and possible sequencing errors. Then, the data is normalized to eliminate systematic errors and batch effects between different samples.

[0055] For example, for RNA-Seq data, the Fragments Per Kilobase of transcript per Million mapped reads (FPKM) or Transcripts Per Million (TPM) methods can be used for normalization. For proteomics and metabolomics data, the internal standard method or total ion current normalization method can be used. The purpose of data preprocessing is to ensure the accuracy and reliability of subsequent analysis.

[0056] Next, the pre-processed multi-omics data were used to diagnose bone status using a preset osteoporosis diagnostic model, which was constructed based on representative differentially expressed biomarkers.

[0057] In a specific implementation, differentially expressed biomarkers are first screened from a multi-omics data set (such as a public database or previous research data) including osteopenia and osteoporosis states. These markers are genes, proteins or metabolites that have significant expression differences between the healthy control group and the osteoporosis patient group. The screening process can use statistical methods such as t-test, ANOVA analysis, and multiple test correction (such as Benjamini-Hochberg method) to control the false discovery rate.

[0058] The screened differentially expressed biomarker set is then subjected to gene function enrichment analysis to identify representative biomarkers. Enrichment analysis can use databases such as Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) to reveal the commonalities of these markers in biological processes, cellular components, and molecular functions, as well as the signaling pathways in which they participate. Through enrichment analysis, biomarkers related to key processes such as bone metabolism, bone remodeling, and calcium signaling can be identified, thereby providing a biological basis for model construction.

[0059] Based on the above screening and analysis results, an osteoporosis diagnosis model is constructed. The model can be trained using machine learning algorithms, such as support vector machine (SVM), random forest (Random Forest) or gradient boosting decision tree (GradientBoosting DecisionTree, GBDT). The input features of the model are the expression levels of representative differentially expressed biomarkers, and the output is the user's bone status prediction.

[0060] After the model is built, the multi-omics data of the user to be diagnosed is input into the preset osteoporosis diagnosis model to diagnose the bone status and generate an osteopenia risk score. The risk score is a numerical value obtained through a certain mapping relationship based on the predicted output of the model, ranging from 0 to 1. The higher the score, the greater the risk of osteoporosis for the user. Based on the risk score, users can be divided into low-risk, medium-risk and high-risk groups, and corresponding health advice and intervention measures can be given to each group.

[0061] For example, for users with a risk score lower than 0.3, it is recommended to maintain good living habits and conduct regular bone density tests; for users with a risk score between 0.3 and 0.7, it is recommended to increase calcium and vitamin D intake, strengthen physical exercise, and avoid bad habits that affect bone health; for users with a risk score higher than 0.7, it is recommended to seek medical attention in a timely manner and receive professional osteoporosis diagnosis and treatment.

[0062] In this way, by comprehensively analyzing multi-omics data and using machine learning technology, early and accurate diagnosis of osteoporosis is achieved, which has important clinical application value. Compared with traditional bone density detection methods, this method can detect the risk of bone loss earlier, providing a new technical means to prevent osteoporosis and its serious consequences such as fractures.

[0063] As an optional embodiment, the multi-omics data includes: multi-omics data of a pilot group; the method also includes: constructing a gene interaction network based on a public database; wherein the nodes in the gene interaction network represent genes, and the edges represent interactions between genes; the interactions include: activation, inhibition, and physical binding; obtaining the physiological data of the pilot group, and using the physiological data of the pilot group as additional features of the nodes and / or edges.

[0064] In a specific implementation, the multi-omics data includes multi-omics data of a pilot group. Since pilots are in a special occupational environment for a long time, such as high-altitude flight, weightlessness simulation, high stress state, etc., these factors may have a unique impact on bone metabolism. Therefore, data analysis for the pilot group can improve the accuracy and pertinence of osteoporosis diagnosis.

[0065] Samples were collected from each pilot, including blood and urine samples, for extraction of genomic DNA, total RNA, proteins, and metabolites. At the same time, dual-energy X-ray absorptiometry (DXA) was used to measure the pilots' bone density to obtain bone density data.

[0066] Next, high-throughput sequencing and mass spectrometry techniques were used to generate multi-omics data for the pilot population.

[0067] For example, genomic data is obtained through whole genome sequencing (WGS) to obtain the pilot's DNA sequence information and identify genetic variations associated with osteoporosis, such as single nucleotide polymorphisms (SNPs). Transcriptomics data uses RNA sequencing (RNA-Seq) technology to obtain the pilot's gene expression profile and identify differentially expressed genes. Proteomics data uses liquid chromatography-tandem mass spectrometry (LC-MS / MS) technology to identify and quantify proteins in serum or tissue samples. Metabolomics data uses nuclear magnetic resonance spectroscopy (NMR) or mass spectrometry technology to detect the types and concentrations of metabolites in pilots.

[0068] After obtaining multi-omics data, gene interaction networks were constructed based on public biological databases, including the STRING database, BioGRID database, and KEGG database, which provide known and predicted gene and protein interaction information.

[0069] Each node in the network represents a gene, which is derived from the differentially expressed gene set of the pilot population. The edges between nodes represent the interactions between genes, including activation, inhibition, and physical binding. Activation means that a gene product promotes the expression or activity of another gene; inhibition means that a gene product inhibits the expression or activity of another gene; and physical binding means that there is a direct physical interaction between two gene products.

[0070] In the process of constructing the gene interaction network, the interaction information of the genes of interest is extracted from the above database, and the interaction confidence threshold (eg, 0.7) is set to retain only interactions with high confidence.

[0071] In addition, the network representation can be in the form of an adjacency matrix or an adjacency list to facilitate subsequent computational analysis.

[0072] In specific implementations, physiological data include but are not limited to: Bone density data: bone density values ​​obtained through DXA measurement. Biochemical indicators: detection of biochemical indicators related to bone metabolism in the blood, such as the concentration of calcium, phosphorus, vitamin D, parathyroid hormone (PTH), etc. Lifestyle and occupational factors: collection of information on pilots' flight time, cumulative flight time, frequency of exposure to high acceleration environments, as well as eating habits, exercise frequency, smoking and drinking conditions, etc.

[0073] In the specific implementation, the collected physiological data needs to be standardized, and the Z-score standardization method is used to eliminate the dimensional differences between different indicators to facilitate fusion with genetic data.

[0074] During the data integration process, the pilots' physiological data are used as node attributes of the gene interaction network or for adjusting edge weights.

[0075] For example, bone density data can be associated with gene nodes related to bone formation and bone resorption to reflect the individual's bone health status; the weights of the edges between genes can be adjusted according to physiological data to reflect the impact of physiological state on gene interactions. For example, the interactions between certain genes may be enhanced or weakened under high stress conditions.

[0076] For example, to achieve effective data fusion, a graph embedding algorithm (such as GraphSAGE) can be used to convert gene interaction networks and physiological data into low-dimensional vector representations.

[0077] In the specific implementation, Python programming language can be used in combination with NetworkX library to process network structure, and Pandas and NumPy libraries can be used to process data to achieve data preprocessing and integration. In this way, multi-omics data and physiological data can be uniformly represented, laying the foundation for subsequent model construction and analysis.

[0078] In this way, by integrating the pilots' physiological data into the gene interaction network, the diagnostic model can be personalized, and the accuracy and reliability of the model can be improved according to the special physiological and occupational characteristics of the pilot group. The introduction of physiological data enriches the information of the network, can more comprehensively reflect the interaction between genes and the environment, and help discover key genes and biological pathways related to osteoporosis, providing a scientific basis for further research and intervention.

[0079] As an optional embodiment, the enrichment analysis includes:

[0080] The preset multimodal attention enhancement network is used to perform gene function enrichment analysis on the differentially expressed biomarker set to determine representative differentially expressed biomarkers in the construction of osteoporosis diagnostic model.

[0081] In order to conduct more accurate gene function enrichment analysis on the differentially expressed biomarker set, a preset multimodal attention enhancement network can be used to determine the differentially expressed biomarkers that are representative in the construction of osteoporosis diagnostic models.

[0082] In the specific implementation, after obtaining the multi-omics data of the pilot group and screening the differentially expressed biomarker set, a multimodal attention enhancement network was introduced to conduct in-depth functional analysis of these biomarkers. The network is designed to integrate multiple data modalities and capture the complex associations and functional characteristics between genes, thereby improving the accuracy and biological interpretation of enrichment analysis.

[0083] First, multiple types of information are collected for each differentially expressed biomarker to construct a multimodal input data set. This information includes: Gene expression level: data from transcriptomics, reflecting the expression changes of genes under different states. Protein interaction information: data based on public databases (such as STRING), describing the interaction relationship between gene-encoded proteins. Functional annotation and pathway information: using databases such as Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG), the functional classification of genes and the biological pathways involved are obtained. Other omics data: such as metabolomics data, providing information on metabolite levels.

[0084] An exemplary, multimodal attention enhancement network has the following architecture design:

[0085] Input layer: Receives multi-modal input data, and each data modality is processed by a separate sub-network. For example, text data (functional annotations) are converted into vector representations through the embedding layer, and numerical data (gene expression levels) are processed through the fully connected layer.

[0086] Feature extraction layer: Extract high-level feature representations for each modality of data. For example, use convolutional neural networks (CNNs) to process sequence data, or use graph neural networks (GNNs) to process protein interaction networks.

[0087] Attention mechanism layer: The attention mechanism is introduced to assign dynamic weights to the features of different modalities. By calculating the contribution of each modality feature to the enrichment analysis task, the network can automatically focus on the most important information.

[0088] Fusion layer: The weighted features of each modality are fused to form a comprehensive feature representation.

[0089] Output layer: Outputs the enrichment probability or score of each biomarker in different functional categories and pathways.

[0090] During the model training process, the existing gene function classification and pathway participation information can be used to construct training sets and validation sets. The supervised learning method is used to minimize the loss function between the predicted output and the true label. In order to prevent overfitting and improve the generalization ability of the model, regularization techniques and early stopping strategies can be used.

[0091] After training, the differentially expressed biomarker set is input into the trained multimodal attention-enhanced network. The model comprehensively analyzes the multimodal information of each biomarker and outputs its enrichment score in different functional categories and pathways. Based on these scores, key functions and pathways can be determined, and biological processes and signaling pathways related to osteoporosis can be identified, such as bone remodeling, calcium signaling, Wnt signaling pathway, etc.

[0092] In addition, representative biomarkers can be screened and biomarkers with high enrichment in key functions and pathways can be selected as core features for diagnostic model construction.

[0093] By using a multimodal attention-enhanced network, enrichment analysis not only considers gene expression levels, but also integrates protein interactions, functional annotations, and pathway information to capture the complex biological relationships between genes. The introduction of the attention mechanism enables the model to automatically identify and emphasize the most valuable features for diagnosis, improving the accuracy and reliability of enrichment analysis.

[0094] As an optional implementation, the multimodal attention enhancement network includes: a graph neural network layer, an attention enhancement layer, and a multimodal fusion layer;

[0095] The graph neural network layer is used to model the graph structure of the gene interaction network and extract the feature representation of each gene node through graph convolution operation;

[0096] The attention enhancement layer is used to dynamically adjust the edge weights between gene nodes based on the pilot's physiological data;

[0097] The multimodal fusion layer is used to fuse the feature representation from the multi-omics data with the pilot's physiological data to generate a comprehensive multimodal feature vector.

[0098] Among them, the network aims to deeply explore the set of differentially expressed biomarkers, combined with the pilots' physiological data, to build a more accurate and personalized osteoporosis diagnosis model.

[0099] In the specific implementation, the graph neural network layer is used to model the graph structure of the constructed gene interaction network. The nodes in the gene interaction network represent genes, and the edges represent the interactions between genes (such as activation, inhibition, and physical binding). In this layer, the graph convolutional neural network (GCN) is used to process the network.

[0100] Through graph convolution operations, the model can effectively aggregate the neighborhood information of each gene node and extract rich feature representations. These feature representations not only contain the properties of the gene itself, but also incorporate the influence of its neighboring genes, reflecting the position and functional role of the gene in the network topology.

[0101] In the specific implementation, the attention enhancement layer dynamically adjusts the edge weights between gene nodes based on the pilot's physiological data. The pilot's physiological data includes bone density, blood biochemical indicators (such as calcium, phosphorus, vitamin D levels), stress levels, etc. These data are standardized and used as one of the inputs of the model.

[0102] The attention enhancement layer introduces the attention mechanism, which can dynamically adjust the interaction strength between gene nodes according to the individual physiological characteristics of the pilot. For example, this layer calculates the attention weight between each pair of gene nodes, and the weight reflects the importance of the gene pair in osteoporosis diagnosis under the current physiological state. In this way, the model can pay more attention to those gene interactions that are highly related to the pilot's physiological state, improving the accuracy and personalization of the diagnosis.

[0103] In the specific implementation, the multimodal fusion layer is used to fuse the gene feature representation from multi-omics data with the pilot's physiological data to generate a comprehensive multimodal feature vector. In this layer, a fusion strategy is used to effectively combine data from different modalities.

[0104] For example, a method that can be used is feature concatenation, which directly connects the gene feature representation and the physiological data vector to form a high-dimensional feature vector. Another method that can be used is weighted fusion, which assigns different weights according to the importance of each modal feature and then adds them together to obtain a comprehensive feature. Through multimodal fusion, the model can simultaneously utilize molecular information at the gene level and physiological information at the individual level to more comprehensively assess the risk of osteoporosis.

[0105] In this way, through the design of the above multimodal attention-enhanced network, the model fully utilizes the structural information of the gene interaction network, the characteristics of differentially expressed biomarkers, and the physiological data of pilots to achieve accurate diagnosis of osteoporosis. This method not only improves the prediction performance of the model, but also enhances its adaptability to individual differences.

[0106] As an optional implementation, the dynamically adjusting edge weights between gene nodes based on the pilot's physiological data includes:

[0107] Using a multi-layer perceptron, the standardized physiological data of the pilot is converted into a low-dimensional embedding vector; wherein the embedding vector has the same dimension as the gene node features in the gene interaction network;

[0108] Based on the preset additive attention mechanism, the attention weights between gene nodes are calculated;

[0109] Based on the attention weights, adjusting edge weights between gene nodes in the gene interaction network;

[0110] Using the adjusted edge weights, we screened out gene node pairs with edge weights increased by more than 20% to identify key genes associated with osteopenia;

[0111] Biological pathway enrichment analysis was performed on the screened key genes related to osteopenia to determine the biological pathways in which they were located.

[0112] See also Figure 2 , Figure 2 A flow chart of a method for calculating the attention weights between gene nodes provided in an embodiment of the present disclosure. As an optional implementation, the method for calculating the attention weights between gene nodes based on a preset additive attention mechanism includes steps S201 to S205:

[0113] S201: Input the pilot's physiological feature embedding vector into a fully connected layer to generate a query vector;

[0114] S202: Input the feature vectors of the gene nodes into independent fully connected layers to generate key vectors and value vectors;

[0115] S203: Generate an attention score by calculating the similarity between the query vector and each key vector;

[0116] S204: Apply the softmax function to all attention scores to obtain normalized attention weights;

[0117] S205: Multiply the normalized attention weight by the corresponding value vector to generate a weighted gene feature representation.

[0118] Furthermore, the present invention dynamically adjusts the edge weights between gene nodes based on the pilot's physiological data to further improve the accuracy and personalization of the osteoporosis diagnosis model.

[0119] In the specific implementation, the pilots' physiological data need to be standardized to eliminate the dimensional differences between different indicators. These physiological data include bone density, blood biochemical indicators (such as calcium, phosphorus, vitamin D levels), stress levels, etc. After standardization, the data are converted into a low-dimensional embedding vector using a multi-layer perceptron (MLP). The dimension of the embedding vector is the same as the dimension of the gene node features in the gene interaction network, ensuring effective fusion in subsequent calculations.

[0120] In the specific implementation, the attention weights between gene nodes are calculated based on the preset additive attention mechanism. Specifically, the pilot's physiological feature embedding vector is input into a fully connected layer to generate a query vector (Query); at the same time, the feature vectors of the gene nodes are respectively input into independent fully connected layers to generate corresponding key vectors (Key) and value vectors (Value). Then, the similarity between the query vector and each key vector is calculated to generate an attention score. The similarity is calculated by linearly transforming the query vector and the key vector, applying an appropriate activation function (such as ReLU or tanh), and then multiplying them with a trainable weight matrix to obtain the final similarity score. The softmax function is applied to all attention scores to obtain normalized attention weights.

[0121] In the specific implementation, the obtained attention weights are used to dynamically adjust the edge weights between gene nodes in the gene interaction network. For gene node pairs with attention weights exceeding 0.8, they are considered to have a stronger correlation under the current pilot's physiological state, so their edge weights are increased by more than 20% of the original weights. For gene node pairs with attention weights below 0.2, they are considered to have a weaker correlation, and their edge weights are appropriately reduced by less than 10% of the original weights. This dynamic adjustment method enables the model to pay more attention to gene interactions that are closely related to the pilot's physiological state.

[0122] Through the adjusted edge weights, gene node pairs with an increase of more than 20% in edge weights were further screened. These gene node pairs showed stronger interactions under the current physiological state and played a key role in the occurrence and development of osteopenia. After identifying these key genes related to osteopenia, biological pathway enrichment analysis was performed on them.

[0123] Public databases (such as KEGG and Reactome) were used to determine the biological pathways in which these genes were located, such as the Wnt signaling pathway and RANK / RANKL pathway, which are closely related to bone metabolism.

[0124] This information helps to deeply understand the molecular mechanism of osteoporosis in pilots and provide a scientific basis for personalized prevention and treatment plans. By combining the pilots' physiological data with the gene interaction network and dynamically adjusting the edge weights, the model can more accurately reflect individual differences and improve the accuracy and reliability of osteoporosis diagnosis.

[0125] As an optional implementation, generating an attention score by calculating the similarity between the query vector and each key vector includes:

[0126] After linearly transforming the query vector and the key vector, the similarity score is calculated through a hyperbolic tangent activation function and a trainable weight matrix.

[0127] In a specific implementation, in order to effectively capture the complex nonlinear relationship between the pilot's physiological feature embedding vector (query vector) and the gene node feature vector (key vector), the present invention uses a trainable weight matrix to perform a linear transformation on these vectors.

[0128] Specifically, the pilot's physiological feature embedding vector is normalized and input into a fully connected layer. This fully connected layer contains a trainable weight matrix and bias term, which generates a new query vector by linearly transforming the input vector. The purpose of this step is to map the original physiological features to a new feature space so that the model can better learn and represent these features.

[0129] At the same time, the feature vector of each gene node is also input into an independent fully connected layer for similar linear transformation to generate a new key vector. Similarly, these fully connected layers also contain trainable weight matrices and bias terms. By linearly transforming the key vector, the model can adjust the representation of the gene node features so that it is in the same feature space as the query vector, which is convenient for subsequent similarity calculations.

[0130] After completing the linear transformation, the hyperbolic tangent activation function (tanh) is applied to the new query vector and key vector respectively. The role of this activation function is to introduce nonlinearity so that the model can capture more complex nonlinear relationships between the query vector and the key vector. Since the output range of the tanh function is (-1, 1), this helps stabilize the model's training process and prevent problems such as gradient disappearance or explosion.

[0131] Next, the activated query vector is concatenated with each activated key vector to form a longer vector. This concatenated vector contains comprehensive information about the pilot's physiological characteristics and gene node characteristics, and can more comprehensively reflect the correlation between the two. This concatenated vector is then input into another fully connected layer, which also contains a trainable weight matrix and bias term. Through linear transformation and activation of the concatenated vector, the model outputs a similarity score indicating the degree of similarity between the query vector and the key vector.

[0132] With the attention weights, the value vector of each gene node (usually a linear transformation of the key vector) is weighted and summed to generate a comprehensive weighted gene feature representation. This weighted feature representation incorporates gene information that is highly correlated with the pilot's physiological characteristics, providing key input for subsequent osteoporosis risk assessment.

[0133] During the model training process, a training set containing the pilot's physiological data and genetic data is used to update the trainable weight matrix in the model through the back-propagation algorithm and the optimization algorithm (such as the Adam optimizer). Specifically, the model first calculates the predicted output through forward propagation, compares it with the actual bone loss risk score or diagnosis result, and calculates the loss function (such as mean square error loss). Then, the model calculates the gradient of the loss function for each weight matrix through back-propagation. Based on this gradient information, the optimization algorithm adjusts the parameters of the weight matrix so that the loss function is gradually reduced.

[0134] During the training process, the trainable weight matrix gradually learned how to effectively map the pilot's physiological characteristics and gene node characteristics to the corresponding feature space, and how to calculate the similarity score between the two. Through continuous adjustment and optimization, the model can more accurately identify key gene nodes related to osteoporosis, improving the accuracy and reliability of diagnosis.

[0135] It should be noted that in order to prevent the model from overfitting, regularization techniques (such as L2 regularization) and methods such as Dropout may be used during the training process. In this way, using a trainable weight matrix and tanh activation function, the model has strong learning ability and adaptability, can handle complex biological data and individual differences of pilots, and provides technical support for personalized medicine.

[0136] Based on the same inventive concept, an osteoporosis diagnosis system based on multi-omics combined with machine learning is also provided in the embodiments of the present application. Since the principle of solving the problem by the osteoporosis diagnosis system based on multi-omics combined with machine learning in the embodiments of the present application is similar to the above-mentioned osteoporosis diagnosis method based on multi-omics combined with machine learning in the embodiments of the present disclosure, the implementation of the system can refer to the implementation of the method, and the repeated parts will not be repeated.

[0137] Reference Figure 3 As shown, Figure 3 A schematic diagram of an osteoporosis diagnosis system based on multi-omics combined with machine learning provided in an embodiment of the present disclosure, the system comprising:

[0138] The acquisition unit 10 is used to obtain multi-omics data of the user to be diagnosed;

[0139] A diagnosis unit 20, for diagnosing bone status based on the multi-omics data using a preset osteoporosis diagnosis model, and generating an osteopenia risk score; the osteopenia risk score is used for osteoporosis diagnosis;

[0140] The osteoporosis diagnostic model is constructed based on representative differentially expressed biomarkers, which are obtained by performing gene function enrichment analysis on a set of differentially expressed biomarkers, and the set of differentially expressed biomarkers is obtained by screening from a data set of multi-omics data including bone loss and osteoporosis status.

[0141] For the description of the processing flow of each unit in the system and the interaction flow between each unit, reference may be made to the relevant description in the above method embodiment, which will not be described in detail here.

[0142] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0143] It should be understood that determining B based on A does not mean determining B only based on A. B can also be determined based on A and / or other information.

[0144] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0145] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0146] The preferred embodiments of the present invention disclosed above are only used to help explain the present application. The preferred embodiments do not describe all the details in detail, nor do they limit the present application to specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present application, so that those skilled in the art can understand and use the present application well. The present application is limited only by the claims and their full scope and equivalents.

Claims

1. A method for diagnosing osteoporosis based on multi-omics combined with machine learning, characterized in that: include: Obtain multi-omics data of the user to be diagnosed; Based on the multi-omics data, a preset osteoporosis diagnostic model is used to diagnose bone status and generate an osteopenia risk score; the osteopenia risk score is used for osteoporosis diagnosis; The osteoporosis diagnostic model is constructed based on representative differentially expressed biomarkers, which are obtained by performing gene function enrichment analysis on a set of differentially expressed biomarkers, and the set of differentially expressed biomarkers is obtained by screening from a data set of multi-omics data including bone loss and osteoporosis status.

2. The method according to claim 1, characterized in that The multi-omics data includes: multi-omics data of a pilot group; the method further includes: Based on the public database, a gene interaction network is constructed; wherein the nodes in the gene interaction network represent genes, and the edges represent the interactions between genes; the interactions include: activation, inhibition, and physical binding; The physiological data of the pilot group is obtained, and the physiological data of the pilot group is used as an additional feature of the node and / or edge.

3. The method according to claim 2, characterized in that The enrichment analysis includes: The preset multimodal attention enhancement network is used to perform gene function enrichment analysis on the differentially expressed biomarker set to determine representative differentially expressed biomarkers in the construction of osteoporosis diagnostic model.

4. The method according to claim 3, characterized in that The multimodal attention enhancement network includes: graph neural network layer, attention enhancement layer, and multimodal fusion layer; The graph neural network layer is used to model the graph structure of the gene interaction network and extract the feature representation of each gene node through graph convolution operation; The attention enhancement layer is used to dynamically adjust the edge weights between gene nodes based on the pilot's physiological data; The multimodal fusion layer is used to fuse the feature representation from the multi-omics data with the pilot's physiological data to generate a comprehensive multimodal feature vector.

5. The method according to claim 4, characterized in that The dynamically adjusting edge weights between gene nodes based on the pilot's physiological data includes: Using a multi-layer perceptron, the standardized physiological data of the pilot is converted into a low-dimensional embedding vector; wherein the embedding vector has the same dimension as the gene node features in the gene interaction network; Based on the preset additive attention mechanism, the attention weights between gene nodes are calculated; Based on the attention weights, adjusting edge weights between gene nodes in the gene interaction network; Using the adjusted edge weights, we screened out gene node pairs with edge weights increased by more than 20% to identify key genes associated with osteopenia; Biological pathway enrichment analysis was performed on the screened key genes related to osteopenia to determine the biological pathways in which they were located.

6. The method according to claim 5, characterized in that The calculation of the attention weights between gene nodes based on the preset additive attention mechanism includes: The pilot’s physiological feature embedding vector is input into a fully connected layer to generate a query vector; Input the feature vectors of gene nodes into independent fully connected layers to generate key vectors and value vectors; Generate an attention score by calculating the similarity between the query vector and each key vector; Apply the softmax function to all attention scores to get the normalized attention weights; Multiply the normalized attention weights by the corresponding value vector to generate the weighted gene feature representation.

7. The method according to claim 6, characterized in that The generation of the attention score by calculating the similarity between the query vector and each key vector includes: After linearly transforming the query vector and the key vector, the similarity score is calculated through a hyperbolic tangent activation function and a trainable weight matrix.

8. The method according to claim 7, characterized in that The adjusting the edge weights between gene nodes in the gene interaction network based on the attention weights comprises: For gene node pairs with attention weights exceeding 0.8, their edge weights are increased based on the attention weights, and the increase in edge weights is greater than 20% of the original weights; For gene-node pairs with attention weights lower than 0.2, their edge weights are reduced based on the attention weights, and the reduction in edge weights is less than 10% of the original weights.

9. Osteoporosis diagnosis system based on multi-omics combined with machine learning, characterized by: include: A collection unit, used to obtain multi-omics data of the user to be diagnosed; A diagnostic unit, configured to diagnose bone status based on the multi-omics data using a preset osteoporosis diagnostic model, and generate an osteopenia risk score; the osteopenia risk score is used for osteoporosis diagnosis; The osteoporosis diagnostic model is constructed based on representative differentially expressed biomarkers, which are obtained by performing gene function enrichment analysis on a set of differentially expressed biomarkers, and the set of differentially expressed biomarkers is obtained by screening from a data set of multi-omics data including bone loss and osteoporosis status.

Images