An artificial intelligence-based auxiliary method for infertility analysis, diagnosis and treatment

Abstract

The present invention relates to the technical field of patient diagnosis and treatment based on artificial intelligence. Specifically, it relates to a method for analyzing and assisting in the diagnosis and treatment of infertility based on artificial intelligence, including: obtaining multi-source data of patients; performing semantic analysis on the multi-source data of patients, and combining machine learning algorithms to identify the symptom characteristic data of patients' infertility; constructing a disease risk prediction model based on deep learning algorithms, using the symptom characteristic data of patients' infertility as input to predict the risk probability of patients' infertility diseases; based on the predicted risk probability of patients' infertility diseases, extracting the potential causes of patients' infertility through association rule mining algorithms, and generating personalized diagnosis and treatment plans for patients in combination with a preset medical knowledge base. The present invention realizes the accurate identification of the causes of infertility and the generation of personalized treatment plans through multi-modal time series data analysis, dynamic cause mining, and medical knowledge constraint verification, significantly improving the accuracy of diagnosis and treatment and the efficiency of clinical decision-making.

Description

Technical Field

[0001] The present invention relates to the technical field of patient diagnosis and treatment based on artificial intelligence. Specifically, it relates to a method for analyzing, diagnosing and treating infertility based on artificial intelligence. Background Art

[0002] Infertility, as a global reproductive health problem, its causes involve multi-dimensional complex mechanisms such as endocrine disorders, reproductive organ abnormalities, genetic factors, and environmental exposures. During the clinical diagnosis and treatment process, it is necessary to integrate multi-modal data such as the patient's hormone level monitoring data over several months, ultrasonic imaging features, medical history records, and lifestyle, which not only has significant time-dynamic characteristics but also has a non-linear coupling relationship between symptom manifestations and causal associations. Traditional medical decision-making has significant limitations in multi-modal data processing. On the one hand, a single prediction model often only focuses on static feature analysis and ignores the temporal correlation of symptom evolution. On the other hand, traditional association rule mining algorithms lack the constraint of medical prior knowledge and are prone to generating a large number of pseudo-associations that do not conform to clinical pathological mechanisms, which not only reduces the credibility of cause identification but may also mislead the design of treatment plans. More critically, in the existing technology chain, the prediction model and the treatment plan generation module are mutually separated, lacking a closed-loop verification mechanism from cause identification to clinical intervention, resulting in limited personalization and executability of the diagnosis and treatment plan. Summary of the Invention

[0003] The purpose of the present invention is to provide a method for analyzing, diagnosing and treating infertility based on artificial intelligence, which realizes accurate identification of infertility causes and generation of personalized treatment plans through multi-modal time-series data analysis, dynamic cause mining, and medical knowledge constraint verification, significantly improving the accuracy of diagnosis and treatment and the efficiency of clinical decision-making.

[0004] The present invention is realized through the following technical solutions:

[0005] A method for analyzing, diagnosing and treating infertility based on artificial intelligence, the steps of the method include:

[0006] Obtain multi-source data of the patient;

[0007] Perform semantic analysis on the multi-source data of the patient, and combine machine learning algorithms to identify the symptom feature data of the patient's infertility;

[0008] Build a disease risk prediction model based on deep learning algorithms, use the symptom feature data of the patient's infertility as input, and predict the risk probability of the patient's infertility disease;

[0009] Based on the predicted risk probability of the patient's infertility disease, extract the potential causes of the patient's infertility through association rule mining algorithms, and generate a personalized diagnosis and treatment plan for the patient in combination with a preset medical knowledge base.

[0010] Optionally, the semantic analysis of the multi-source data of the patient is specifically as follows:

[0011] Perform natural language preprocessing on the multi-source data of the patient;

[0012] Construct a bidirectional long short-term memory network model, add inter-layer residual connections in the bidirectional long short-term memory network architecture, and integrate the attention mechanism to perform semantic analysis on the preprocessed multi-source data of the patient to generate a semantic vector of the patient's infertility;

[0013] Construct a vector comparison space based on a preset medical knowledge base, calculate the similarity between the semantic vector of the patient's infertility and the vector of the medical knowledge base through the weighted cosine similarity algorithm, and introduce a semantic association correction factor to optimize the similarity calculation result, and generate a set of candidate symptoms of the patient's infertility with confidence scores;

[0014] Screen the set of candidate symptoms of the patient's infertility through a dynamic threshold filtering module, and output the symptom feature data of the patient's infertility.

[0015] Optionally, the natural language preprocessing of the multi-source data of the patient is specifically as follows:

[0016] Standardize the format of the multi-source data of the patient to generate a text data stream with a unified encoding;

[0017] Perform word segmentation on the multi-source data of the patient, and synchronously remove invalid stop words;

[0018] Determine the medical semantic attributes of the multi-source data of the patient through part-of-speech tagging;

[0019] Output the multi-source data of the patient after completing natural language preprocessing.

[0020] Optionally, the construction of the disease risk prediction model based on the deep learning algorithm is specifically as follows:

[0021] Input the symptom feature data of the patient's infertility;

[0022] Run the first convolutional neural network model and the second convolutional neural network model in parallel, and respectively extract features from the symptom feature data of the patient's infertility;

[0023] Input the feature extraction result of the first convolutional neural network model into the forward LSTM model to calculate the forward hidden state, and input the feature extraction result of the second convolutional neural network model into the backward LSTM model to calculate the backward hidden state;

[0024] Jointly splice the forward hidden state and the backward hidden state, and apply the multi-head attention mechanism to obtain the context semantic vector;

[0025] Aggregate the context semantic vectors, and after activation by the Sigmoid function, output the predicted risk probability of the patient's infertility disease.

[0026] Optionally, the parallel running of the first convolutional neural network model and the second convolutional neural network model is specifically as follows:

[0027] Define the first convolutional neural network model, which uses a 3×1 convolutional kernel to extract local features of the patient's infertility symptom feature data;

[0028] Define the second convolutional neural network model, which uses a 7×1 convolutional kernel to extract global features of the patient's infertility symptom feature data;

[0029] And both the first convolutional neural network model and the second convolutional neural network model are one-dimensional convolutional neural network models.

[0030] Optionally, the joint splicing of the forward hidden state and the backward hidden state and the application of the multi-head attention mechanism to obtain the context semantic vector are specifically as follows:

[0031] Jointly splice the forward hidden state and the backward hidden state to form a joint hidden state matrix containing bidirectional temporal information;

[0032] Split the joint hidden state matrix into multiple attention heads for parallel computing:

[0033] Each attention head independently calculates the correlation weights of the query vector, key vector, and value vector;

[0034] Calculate the similarity matrix of the query vector and the key vector, and successively perform scaling processing and softmax normalization, and perform weighted aggregation on the value vector to generate the context semantic vector.

[0035] Optionally, the feature aggregation of the context semantic vector is specifically as follows:

[0036] Perform mean pooling on the context semantic vector in the time dimension to extract global semantic features;

[0037] Perform max pooling on the local features and global features of the patient's infertility symptom feature data in the time dimension respectively;

[0038] Splice the global semantic features, local features, and global features of the patient's infertility symptom feature data to form a comprehensive feature vector, and after activation by the Sigmoid function, output the predicted risk probability of the patient's infertility disease.

[0039] Optionally, the extraction of potential causes of the patient's infertility by the association rule mining algorithm is specifically as follows:

[0040] Extract the multi-source data of the patient through the association rule mining algorithm to obtain the set of candidate symptom-cause pairs for the patient's infertility;

[0041] According to the time window of the incubation period of the cause in the preset medical knowledge base, verify the time matching of the set of candidate symptom-cause pairs for the patient's infertility, and output the potential causes of the patient's infertility, including the cause name and the temporal confidence level.

[0042] Optionally, the generation of the personalized diagnosis and treatment plan for the patient by combining the preset medical knowledge base is specifically as follows:

[0043] Input the potential causes of the patient's infertility, the predicted risk probability of the patient's infertility disease, and the baseline incidence data of each cause in the preset medical knowledge base;

[0044] Calculate the comprehensive score for each cause in the potential causes of the patient's infertility. The scoring basis includes: the predicted risk probability of the patient's infertility disease, the temporal confidence level of the symptom and the cause, and the baseline incidence of the selected cause;

[0045] Generate a priority list of the causes of the patient's infertility and sort it from high to low according to the score;

[0046] Execute the preset hierarchical decision according to the priority list of the causes of the patient's infertility, and output the personalized diagnosis and treatment plan for the patient.

[0047] Optionally, the preset hierarchical decision includes:

[0048] The first priority decision: Define that this decision is executed when the score is greater than or equal to the first threshold, and match the emergency diagnosis and treatment plan in the preset medical knowledge base;

[0049] The second priority decision: Define that this decision is executed when the score is greater than or equal to the second threshold and lower than the first threshold, and select the treatment plan that balances the efficacy and cost factors to represent the personalized diagnosis and treatment plan for the patient;

[0050] The third priority decision: Define that this decision is executed when the score is lower than the second threshold, mark the potential causes of the patient's infertility to be confirmed, generate a list of causes for manual review, and trim the list of causes for manual review in combination with the medical record data in the multi-source data of the patient, and output the final diagnosis and treatment plan for the patient.

[0051] The technical solution of the present invention has at least the following advantages and beneficial effects:

[0052] Through the deep integration of temporal feature extraction and correlation analysis under medical constraints, the present invention constructs a complete technical closed-loop from risk prediction to etiology mining and then to treatment decision-making. On the one hand, by using bidirectional LSTM and multi-head attention mechanisms, it accurately captures the local fluctuations and global trends of symptom evolution, significantly enhancing the temporal sensitivity of risk prediction. On the other hand, it innovatively combines the dynamic support degree adjustment mechanism with medical knowledge base verification to ensure that etiology mining not only meets statistical significance but also satisfies clinical pathological logic. Through the multi-level feature pooling strategy, it effectively integrates semantic features across time steps and key event response values, enabling the model to have both global trend perception and local anomaly detection capabilities. In addition, the treatment plan generation module not only strictly verifies the proportion of indication and contraindication indicators but also enforces the physiological cycle time window constraint to ensure the safety and operability of clinical intervention. The present invention forms a self-optimizing intelligent diagnosis and treatment ecosystem by automatically triggering multidisciplinary consultations and knowledge base iterative updates, and finally outputs personalized plan documents to achieve seamless connection between precision medicine and clinical practice. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a schematic flowchart of a method for analyzing, diagnosing and treating infertility based on artificial intelligence provided by the present invention;

[0054] Figure 2 It is a schematic logical diagram of semantic analysis provided by the present invention;

[0055] Figure 3 It is a schematic logical diagram of a disease risk prediction model provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0056] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Usually, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0057] As Figure 1 shown, the present invention provides one of the embodiments: a method for analyzing, diagnosing and treating infertility based on artificial intelligence, and the steps of the method include:

[0058] Obtain multi-source data of the patient;

[0059] Perform semantic analysis on the multi-source data of the patient, and combine machine learning algorithms to identify the symptom feature data of the patient's infertility;

[0060] Construct a disease risk prediction model based on deep learning algorithms, using the symptom characteristic data of patients with infertility as input to predict the risk probability of patients' infertility diseases;

[0061] Based on the predicted risk probability of patients' infertility diseases, extract the potential causes of patients' infertility through association rule mining algorithms, and generate personalized diagnosis and treatment plans for patients in combination with a preset medical knowledge base.

[0062] In this embodiment, this embodiment mainly includes Step 1: Collect multi-source data of patients, including symptom descriptions, age, gender, medical history, examination results, etc.; Step 2, use natural language processing technology to perform semantic analysis on the symptom descriptions of patients, and combine machine learning algorithms to identify key symptoms related to infertility; Step 3, construct a disease risk prediction model based on deep learning algorithms, use multi-dimensional data such as the age, gender, medical history, and examination results of patients as input, and train the model to predict the risk probability of patients suffering from infertility diseases such as polycystic ovary syndrome, varicocele, and endometriosis; Step 4, use association rule mining algorithms to analyze the association relationships between multi-source data of patients, and find out the potential causes of infertility; Step 5, according to the analysis results of Step 2, Step 3, and Step 4, combine clinical practice guidelines and expert experience knowledge bases to generate personalized diagnosis and treatment suggestions for doctors, including diagnosis suggestions, treatment suggestions, and follow-up suggestions.

[0063] As Figure 2 shown, in a further implementation of the above embodiment, the semantic analysis of the multi-source data of patients is specifically as follows:

[0064] Perform natural language preprocessing on the multi-source data of patients;

[0065] Construct a bidirectional long short-term memory network model, add inter-layer residual connections in the bidirectional long short-term memory network architecture, and integrate an attention mechanism to perform semantic analysis on the preprocessed multi-source data of patients to generate a semantic vector of patients' infertility;

[0066] Construct a vector comparison space based on a preset medical knowledge base, calculate the similarity between the semantic vector of patients' infertility and the vectors in the medical knowledge base through the weighted cosine similarity algorithm, and introduce a semantic association correction factor to optimize the similarity calculation result, and generate a set of candidate symptoms of patients' infertility with confidence scores;

[0067] Screen the set of candidate symptoms of patients' infertility through a dynamic threshold filtering module, and output the symptom characteristic data of patients' infertility.

[0068] In implementation, the algorithm description of the optimized cosine similarity algorithm in this embodiment is as follows: Weight adjustment: The original cosine similarity algorithm treats each dimension of the vector equally. However, in the analysis of infertility symptoms, the importance of different symptom dimensions varies. For example, the importance of symptoms related to "abnormal ovulation" may be higher than that of "occasional abdominal pain" in the diagnosis of infertility. Through expert experience and statistical analysis of a large number of case data, weights are assigned to different dimensions of the symptom description semantic vector and the infertility symptom semantic vector in the medical knowledge base. Let the symptom description semantic vector be , where is the symptom description semantic vector, with dimension n, , are respectively the nth and ith elements of the vector, and the medical knowledge base symptom semantic vector is , where is the medical knowledge base symptom semantic vector, with dimension n, , are respectively the nth and ith elements of the vector, and the weight vector is , where , are respectively the nth and ith elements of the weight vector. The optimized dot product calculation becomes , where k is the weighted medical knowledge base vector, and n is the dimension of the vector. When calculating the cosine similarity, the adjusted dot product is used to make the algorithm more focused on important symptom dimensions and improve the accuracy of key symptom recognition. Semantic relevance correction: In addition to the angle between vector spaces, this embodiment introduces a semantic relevance correction factor and uses the medical ontology knowledge base to clarify the semantic relationship between symptoms. For example, "irregular menstruation" and "menstrual disorder" have similar semantics, and the correlation degree should be appropriately increased when calculating the similarity. The shortest path or semantic association strength between two symptoms is found through a semantic network or knowledge graph. Let the correction factor k be K>1 (indicating strong semantic association, k=1 indicating ordinary association, 0<k<1 indicating weak semantic association). The optimized cosine similarity formula is , where is the optimized weighted cosine similarity, making the similarity calculation more in line with medical semantic logic.

[0069] In this embodiment, the improved bidirectional long short-term memory network (Bi-LSTM) structure is applied: In the Bi-LSTM model based on the attention mechanism, an inter-layer residual connection is added. A direct connection is established between different hidden layers to allow information to be transmitted more smoothly, alleviate the problem of gradient disappearance, and enable the model to better learn the complex dependencies in the long sequence symptom description. For example, in a multi-layer Bi-LSTM, the In addition to passing the output of the layer to the layer for normal processing, it is also directly added to the input of the layer, that is , where represents the index of the current layer, is the hidden layer output of the layer, h is the hidden layer output, W is the weight matrix, is the weight matrix of the layer, b is the bias, is the bias term of the layer, is the activation function, which enhances the model's learning ability for symptom semantics. Using a language model pre-trained on a large-scale medical text dataset, transfer the medical semantic features learned by the pre-trained model to the current infertility symptom analysis model. Freeze the parameters of some layers of the pre-trained model and fine-tune the model on infertility symptom data. In this way, the model can utilize the general medical semantic knowledge captured by the pre-trained model to learn the relevant feature representations of infertility symptoms faster and more accurately, improving the recognition effect of key symptoms, especially with obvious advantages when the data volume is limited.

[0070] Furthermore, the natural language preprocessing of the patient's multi-source data is specifically as follows:

[0071] Standardize the format of the patient's multi-source data to generate a text data stream with unified encoding;

[0072] Segment the words of the patient's multi-source data and synchronously remove invalid stop words;

[0073] Determine the medical semantic attributes of the patient's multi-source data through part-of-speech tagging;

[0074] Output the patient's multi-source data after completing natural language preprocessing.

[0075] In specific implementation, this embodiment adopts a word segmentation algorithm combining rules and statistical learning to segment the text of the patient's symptom description into individual words. For example, "menstrual disorder accompanied by lower abdominal distension" is segmented into "menstrual disorder", "and", "accompanied by", "lower abdominal distension". Then remove stop words such as "de" and "le" that have no practical meaning to reduce data redundancy. Then perform part-of-speech tagging to determine the part of speech of each word, such as nouns, verbs, adjectives, etc., so as to understand the text meaning more accurately and prepare for subsequent analysis.

[0076] As Figure 3 shown, in the specific application of this embodiment, the disease risk prediction model constructed based on the deep learning algorithm is specifically as follows:

[0077] Input the infertility symptom feature data of the patient.

[0078] Run the first convolutional neural network model and the second convolutional neural network model in parallel to extract feature data of the infertility symptoms of patients respectively.

[0079] The parallel running of the first convolutional neural network model and the second convolutional neural network model is specifically as follows:

[0080] Define the first convolutional neural network model, which uses a 3×1 convolutional kernel to extract local features of the infertility symptom feature data of patients;

[0081] Define the second convolutional neural network model, which uses a 7×1 convolutional kernel to extract global features of the infertility symptom feature data of patients;

[0082] Both the first convolutional neural network model and the second convolutional neural network model are one-dimensional convolutional neural network models.

[0083] Input the feature extraction result of the first convolutional neural network model into the forward LSTM model to calculate the forward hidden state, and input the feature extraction result of the second convolutional neural network model into the backward LSTM model to calculate the backward hidden state.

[0084] Jointly splice the forward hidden state and the backward hidden state, and apply the multi-head attention mechanism to obtain the context semantic vector.

[0085] The joint splicing of the forward hidden state and the backward hidden state and the application of the multi-head attention mechanism to obtain the context semantic vector is specifically as follows:

[0086] Jointly splice the forward hidden state and the backward hidden state to form a joint hidden state matrix containing bidirectional temporal information;

[0087] Split the joint hidden state matrix into multiple attention heads for parallel computing:

[0088] Each attention head independently calculates the correlation weights of the query vector, key vector, and value vector;

[0089] Calculate the similarity matrix between the query vector and the key vector, and successively perform scaling processing and softmax normalization, and perform weighted aggregation on the value vector to generate the context semantic vector.

[0090] Aggregate the features of the context semantic vector, and output the predicted risk probability of the infertility disease of the patient after activation by the Sigmoid function.

[0091] The aggregation of the features of the context semantic vector is specifically as follows:

[0092] Perform mean pooling on the context semantic vector in the time dimension to extract global semantic features;

[0093] Perform max pooling in the time dimension on the local and global features of the patient's infertility symptom characteristic data respectively;

[0094] Concatenate the global semantic features, the local and global features of the patient's infertility symptom characteristic data to form a comprehensive feature vector, and output the predicted risk probability of the patient's infertility disease after activation by the Sigmoid function.

[0095] During implementation, obtain the infertility symptom characteristic data of the patient as input, including hormone level sequences, ultrasound imaging indicators, and other clinical manifestation information. To fully capture the local patterns and global trends of the data, a first convolutional neural network model and a second convolutional neural network model are respectively defined to perform feature extraction on the patient data in parallel. The first convolutional neural network model uses a 3×1 convolutional kernel to focus on local associations between short-term or adjacent time steps; the second convolutional neural network model uses a 7×1 convolutional kernel, mainly for detecting global changes within a larger time range. Both belong to one-dimensional convolutional neural network models, aiming to perform efficient and targeted convolutional scans on the time dimension of the infertility symptom characteristic data. After completing the parallel convolutional extraction, input the feature map obtained by the first convolutional neural network model into the forward LSTM model to obtain the forward hidden state; input the feature map obtained by the second convolutional neural network model into the backward LSTM model to obtain the backward hidden state. By jointly concatenating the forward and backward hidden states, a joint hidden state matrix containing bidirectional temporal information is formed. In this matrix, each time step aggregates the feature representations from front to back and from back to front, better retaining the complete context of the patient's symptoms evolving over time. Subsequently, to further highlight the importance of different time steps or feature dimensions, apply a multi-head attention mechanism to the joint hidden state matrix. Specifically, split the joint hidden state matrix into multiple attention heads for parallel computing. Each head respectively generates query vectors, key vectors, and value vectors, calculates the similarity scores, and after scaling and softmax normalization, weights and aggregates the value vectors, finally outputting the context semantic vector. This process can adaptively emphasize or weaken the contributions of different temporal positions to the model's judgment, making the prediction more targeted and interpretable. After obtaining the context semantic vector, perform deep information fusion based on the feature aggregation strategy. Its core steps include: performing mean pooling on the context semantic vector in the time dimension to extract the overall semantic information across time steps; performing max pooling on the local features and global features output from the first convolutional neural network and the second convolutional neural network respectively to retain their most significant feature values; concatenating the global semantic vector, local features, and global features again to form a unified comprehensive feature vector. Finally, activate the comprehensive feature vector through the Sigmoid function to output the predicted risk probability of the patient's infertility disease. This probability value can be further applied to clinical auxiliary diagnosis, etiology mining, and treatment plan recommendation, providing a more efficient and accurate risk assessment tool for the field of reproductive medicine.

[0096] In a further implementation of this embodiment, the potential causes of the patient's infertility are extracted by the association rule mining algorithm, specifically as follows:

[0097] Extract the multi-source data of the patient through the association rule mining algorithm to obtain a set of candidate symptom-cause pairs for the patient's infertility;

[0098] According to the time window of the incubation period of the cause of disease in the preset medical knowledge base, verify the time matching of the set of candidate symptoms-causes of infertility and sterility of the patient, and output the potential causes of infertility and sterility of the patient, including the name of the cause and the chronological confidence level.

[0099] The generation of a personalized diagnosis and treatment plan for the patient by combining the preset medical knowledge base is specifically as follows:

[0100] Input the potential causes of infertility and sterility of the patient, the predicted risk probability of the infertility and sterility disease of the patient, and the data of the baseline incidence rate of each cause in the preset medical knowledge base;

[0101] Calculate the comprehensive score for each cause in the potential causes of infertility and sterility of the patient. The scoring basis includes: the predicted risk probability of the infertility and sterility disease of the patient, the chronological confidence level of the symptom and the cause, and the baseline incidence rate of the selected cause;

[0102] Generate a priority list of the causes of infertility and sterility of the patient, and sort them from high to low according to the score;

[0103] Execute the preset hierarchical decision according to the priority list of the causes of infertility and sterility of the patient, and output the personalized diagnosis and treatment plan for the patient.

[0104] The preset hierarchical decision includes:

[0105] The first priority decision: Define that this decision is executed when the score is greater than or equal to the first threshold, and match the emergency diagnosis and treatment plan in the preset medical knowledge base;

[0106] The second priority decision: Define that this decision is executed when the score is greater than or equal to the second threshold and lower than the first threshold. Select a treatment plan that balances the efficacy and cost factors to represent the personalized diagnosis and treatment plan for the patient;

[0107] The third priority decision: Define that this decision is executed when the score is lower than the second threshold. Mark the potential causes of infertility and sterility of the patient to be confirmed, generate a list of causes for manual review, and trim the list of causes for manual review in combination with the medical record data in the multi-source data of the patient, and output the final diagnosis and treatment plan for the patient.

[0108] Specifically, in this embodiment, the association rule mining algorithm is first used to extract the multi-source data of patients to obtain a set of candidate symptom-cause pairs. The multi-source data includes multi-dimensional information such as hormone level records, ultrasonic detection indexes, medical history documents, and living habits. Through a rule mining model trained and verified under a large-scale clinical sample, a preliminary association identification is performed on the infertility candidate symptoms and the possible corresponding causes to form symptom-cause pairs with basic statistical significance. According to the time window of the cause incubation period defined in the preset medical knowledge base, a time matching verification is performed on the above set of candidate symptom-cause pairs to ensure that only the causes that can produce corresponding clinical manifestations within the incubation period are retained. The potential causes of the patient's infertility output by this step will be recorded according to the cause name and the chronological confidence for use by the subsequent diagnosis and treatment plan generation module. In this embodiment, the potential causes of infertility, the predicted risk probability of the patient's infertility disease, and the benchmark incidence data of each cause in the preset medical knowledge base are jointly used as inputs to comprehensively score each cause. The main scoring basis includes: the predicted risk probability of the patient's infertility disease, the chronological confidence of the symptoms and causes, and the benchmark incidence of this cause in the overall population. After scoring, all causes are sorted from high to low according to the scores to form a priority list of infertility causes. According to this priority list, a hierarchical decision-making process is executed to output the patient's personalized diagnosis and treatment plan. The hierarchical decision-making process is specifically divided into three priorities: the first priority decision corresponds to the cause with a score greater than or equal to the first threshold, and directly matches the diagnosis and treatment plan marked as emergency treatment or critical intervention in the preset medical knowledge base; the second priority decision corresponds to the cause with a score greater than or equal to the second threshold but lower than the first threshold, and a modified treatment plan that takes into account both efficacy and cost is selected to reasonably allocate medical resources while meeting the patient's medical needs; the third priority decision corresponds to the cause with a score lower than the second threshold, which is temporarily marked as to be confirmed, a list of manually reviewed causes is generated, and the medical record documents and examination results in the patient's multi-source data are retrieved for in-depth comparison. If it is found during the expert consultation or further examination and verification process that there is indeed clinical evidence for this cause, it will be included in the final patient diagnosis and treatment plan; if the verification is invalid or there is not enough evidence to support it, it will be excluded to ensure the safety and accuracy of the diagnosis and treatment plan. It can be understood that the treatment plans are all existing treatment plans in the medical knowledge base. Finally, through the complete processes of the above-mentioned association rule mining, incubation period verification, comprehensive scoring, and hierarchical decision-making, a personalized diagnosis and treatment plan including diagnostic basis, recommended treatment plan, and auxiliary monitoring indexes is output. This plan can be docked with the patient's medical record system or electronic health record (EHR) to realize the closed-loop management of clinical diagnosis and treatment, which not only improves the accuracy of identifying and intervening in the causes of infertility, but also consolidates the standardized utilization and continuous update of medical data, providing more scientific and efficient decision-making support for clinicians.

[0109] The above are only the preferences of the present invention and are not used to limit the present invention. For those skilled in the art, various modifications and changes can be made to the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An infertility analysis and diagnosis assistance method based on artificial intelligence, characterized in that The steps of the method include: Obtain multi-source data of the patient; Perform semantic analysis on the multi-source data of the patient, and combine machine learning algorithms to identify the characteristic data of the patient's infertility symptoms; Build a disease risk prediction model based on a deep learning algorithm, use the characteristic data of the patient's infertility symptoms as input, and predict the risk probability of the patient's infertility disease; Based on the predicted risk probability of the patient's infertility disease, extract the potential causes of the patient's infertility through an association rule mining algorithm, and generate a personalized diagnosis and treatment plan for the patient in combination with a preset medical knowledge base; The semantic analysis of the multi-source data of the patient is specifically as follows: Perform natural language preprocessing on the multi-source data of the patient; Build a bidirectional long short-term memory network model, add an inter-layer residual connection in the bidirectional long short-term memory network architecture, and integrate an attention mechanism to perform semantic analysis on the preprocessed multi-source data of the patient to generate a semantic vector of the patient's infertility; Build a vector comparison space based on a preset medical knowledge base, calculate the similarity between the semantic vector of the patient's infertility and the vector in the medical knowledge base through a weighted cosine similarity algorithm, and introduce a semantic association correction factor to optimize the similarity calculation result, and generate a set of candidate symptoms of the patient's infertility with confidence scores; Screen the set of candidate symptoms of the patient's infertility through a dynamic threshold filtering module, and output the characteristic data of the patient's infertility symptoms; The building of the disease risk prediction model based on a deep learning algorithm is specifically as follows: Input the characteristic data of the patient's infertility symptoms; Run the first convolutional neural network model and the second convolutional neural network model in parallel, and respectively extract features from the characteristic data of the patient's infertility symptoms; Input the feature extraction result of the first convolutional neural network model into a forward LSTM model to calculate the forward hidden state, and input the feature extraction result of the second convolutional neural network model into a backward LSTM model to calculate the backward hidden state; Jointly splice the forward hidden state and the backward hidden state, and apply a multi-head attention mechanism to obtain a context semantic vector; Aggregate the features of the context semantic vector, and output the predicted risk probability of the patient's infertility disease after activation by the Sigmoid function; The parallel running of the first convolutional neural network model and the second convolutional neural network model is specifically as follows: Define the first convolutional neural network model, which uses a 3×1 convolutional kernel to extract local features of the characteristic data of the patient's infertility symptoms; Define the second convolutional neural network model, which uses a 7×1 convolutional kernel to extract global features of the characteristic data of the patient's infertility symptoms; And both the first convolutional neural network model and the second convolutional neural network model are one-dimensional convolutional neural network models.

2. The infertility analysis and diagnosis and treatment assistance method based on artificial intelligence according to claim 1, wherein The natural language preprocessing of the multi-source data of the patient is specifically as follows: Standardize the format of the multi-source data of the patient to generate a text data stream with a unified encoding; Perform word segmentation on the multi-source data of the patient and synchronously remove invalid stop words; Determine the medical semantic attributes of the multi-source data of the patient through part-of-speech tagging; Output the multi-source data of the patient after completing natural language preprocessing.

3. The infertility analysis and diagnosis and treatment assistance method based on artificial intelligence according to claim 1, characterized in that, Jointly concatenating the forward hidden state and the backward hidden state, and applying a multi-head attention mechanism to obtain a context semantic vector, which is specifically as follows: Jointly concatenating the forward hidden state and the backward hidden state to form a joint hidden state matrix containing bidirectional temporal information; Splitting the joint hidden state matrix into multiple attention heads for parallel computing: Each attention head independently calculates the correlation weights of the query vector, the key vector, and the value vector; Calculating the similarity matrix between the query vector and the key vector, and successively performing scaling processing and softmax normalization, and performing weighted aggregation on the value vector to generate a context semantic vector.

4. The method for infertility analysis and diagnosis assistance based on artificial intelligence according to claim 3, wherein, The feature aggregation of the context semantic vector is specifically as follows: Performing mean pooling on the context semantic vector in the time dimension to extract global semantic features; Performing max pooling on the local features and global features of the patient's infertility symptom feature data in the time dimension respectively; Concatenating the global semantic features, the local features and global features of the patient's infertility symptom feature data to form a comprehensive feature vector, and outputting the predicted risk probability of the patient's infertility disease after activation by the Sigmoid function.

5. The artificial-intelligence-based infertility analysis and diagnosis and treatment assistance method according to claim 4, wherein Extracting the potential causes of the patient's infertility through the association rule mining algorithm, which is specifically as follows: Extracting the patient's multi-source data through the association rule mining algorithm to obtain a set of candidate symptom-cause pairs for the patient's infertility; Verifying the time matching of the set of candidate symptom-cause pairs for the patient's infertility according to the time window of the disease incubation period in the preset medical knowledge base, and outputting the potential causes of the patient's infertility, including the cause name and the temporal confidence level.

6. The artificial intelligence-based infertility analysis, diagnosis and treatment assistance method according to claim 5, wherein Generating a personalized diagnosis and treatment plan for the patient in combination with the preset medical knowledge base, which is specifically as follows: Inputting the potential causes of the patient's infertility, the predicted risk probability of the patient's infertility disease, and the benchmark incidence data of each cause in the preset medical knowledge base; Calculating a comprehensive score for each cause in the potential causes of the patient's infertility, and the scoring basis includes: the predicted risk probability of the patient's infertility disease, the temporal confidence level of the symptom and the cause, and the benchmark incidence of the selected cause; Generating a priority list of the causes of the patient's infertility and sorting them from high to low according to the scores; Performing a preset hierarchical decision according to the priority list of the causes of the patient's infertility, and outputting a personalized diagnosis and treatment plan for the patient.

7. The artificial intelligence-based infertility analysis and diagnosis assistance method according to claim 6, wherein, The preset hierarchical decision includes: The first priority decision: defining that this decision is executed when the score is greater than or equal to the first threshold, and matching the emergency diagnosis and treatment plan in the preset medical knowledge base; The second priority decision: defining that this decision is executed when the score is greater than or equal to the second threshold and lower than the first threshold, and selecting a treatment plan that balances the efficacy and cost factors to represent the personalized diagnosis and treatment plan for the patient; The third priority decision: defining that this decision is executed when the score is lower than the second threshold, marking the potential causes of the patient's infertility to be confirmed, generating a list of manually reviewed causes, and trimming the list of manually reviewed causes in combination with the medical record data in the patient's multi-source data, and outputting the final diagnosis and treatment plan for the patient.

Images