Processing method and device for hypertrophic cardiomyopathy risk prediction model

Abstract

The embodiment of the invention relates to a processing method and device for a hypertrophic cardiomyopathy risk prediction model. The method comprises the following steps: constructing the hypertrophic cardiomyopathy risk prediction model; constructing a generalized crowd model data set as a first data set, and constructing a specified crowd model data set as a second data set based on the specified crowd characteristics; performing pre-training on the model based on the first data set, and then performing fine tuning on the model based on the second data set; after fine tuning is finished, inputting a tested person myocardial strain parameter set meeting the characteristics of the specified crowd into the model for prediction, and feeding back a prediction result to a user. According to the invention, prediction accuracy and prediction stability can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and device for processing a hypertrophic cardiomyopathy risk prediction model. Background Art

[0002] Hypertrophic cardiomyopathy (HCM) is a non-ischemic cardiomyopathy characterized by myocardial hypertrophy. It carries the risk of sudden death and is a cause of sudden death during exercise. Some HCM patients experience no early clinical symptoms, and some even die suddenly at the first symptom. Therefore, early prognosis of HCM is crucial.

[0003] Myocardial strain technology, based on cardiac magnetic resonance (CMR) imaging, is a sensitive and noninvasive method for measuring myocardial strain parameters. It can quantitatively assess a range of myocardial strain parameters in the left heart (left atrium and left ventricle). Myocardial strain assessment can identify early signs of myocardial dysfunction before a patient's left ventricular ejection fraction (LVEF) decreases. This means that myocardial strain parameters can be used to predict the risk of hypertrophic cardiomyopathy at an early stage.

[0004] An analysis of traditional hypertrophic cardiomyopathy risk prediction schemes revealed several issues: 1) Most schemes only consider left ventricular parameters and ignore left atrial parameters. Left ventricular parameters are primarily based on three categories: global longitudinal strain (GLS), global circumferential strain (GCS), and global radial strain (GRS), without considering other detailed branches. Under these circumstances, the accuracy of hypertrophic cardiomyopathy risk prediction is limited by the parameter dimensions and is difficult to further improve. 2) Although myocardial strain technology can provide accurate myocardial strain parameters, the interpretation of the results requires physicians to possess extensive professional knowledge and experience. This means that the stability of hypertrophic cardiomyopathy risk prediction is limited by individual professional experience, making it difficult to achieve stable prediction quality.

[0005] To address these issues, the present invention proposes a deep learning model based on artificial intelligence technology that can predict the risk of hypertrophic cardiomyopathy based on multiple types of myocardial strain parameters of the left atrium and left ventricle. This model can improve prediction accuracy based on multi-dimensional parameters and also improve prediction stability by breaking away from the limitations of personal experience. In addition to the three traditional left ventricular parameters (GLS, GCS, and GRS), the multiple types of myocardial strain parameters mentioned here also include six other parameters: circumferential transmural strain difference (cTSD) between the left ventricular epicardium and endocardium, left ventricular peak systolic strain rate (PSSR), left ventricular peak diastolic strain rate (PDSR), left atrial passive strain (Ee), left atrial early diastolic peak negative strain rate (sRe), and left atrial late diastolic peak negative strain rate (sRa). Summary of the Invention

[0006] The purpose of the present invention is to provide a processing method, device, electronic device and computer-readable storage medium for a hypertrophic cardiomyopathy risk prediction model in response to the defects of the existing technology. The present invention constructs a hypertrophic cardiomyopathy risk prediction model that can predict risks based on six types of left ventricular parameters (GLS parameters, GCS parameters, GRS parameters, cTSD parameters, PSSR parameters, PDSR parameters) and three types of left atrial parameters (Ee parameters, sRe parameters, sRa parameters); and constructs a generalized population model data set (first data set) and a specified population model data set (second data set) through big data collection; and first pre-trains the hypertrophic cardiomyopathy risk prediction model based on the first data set and then fine-tunes the hypertrophic cardiomyopathy risk prediction model based on the second data set; after the model fine-tuning is completed, the hypertrophic cardiomyopathy risk prediction model is used to make predictions based on the myocardial strain parameters of the subject. The present invention can improve the accuracy and stability of predictions.

[0007] To achieve the above objectives, a first aspect of an embodiment of the present invention provides a method for processing a hypertrophic cardiomyopathy risk prediction model, the method comprising:

[0008] A deep learning model for predicting the risk probability of hypertrophic cardiomyopathy is constructed, denoted as a corresponding hypertrophic cardiomyopathy risk prediction model; the hypertrophic cardiomyopathy risk prediction model is used to predict the risk probability of hypertrophic cardiomyopathy based on a myocardial strain parameter vector X input by the model and output a corresponding predicted probability p; the myocardial strain parameter vector X is composed of six categories of left ventricular parameters and three categories of left atrial parameters, the six categories of left ventricular parameters including GLS parameters, GCS parameters, GRS parameters, cTSD parameters, PSSR parameters, and PDSR parameters; the three categories of left atrial parameters including Ee parameters, sRe parameters, and sRa parameters; the predicted probability p is a probability value between 0 and 1;

[0009] Construct a generalized population model dataset as the corresponding first dataset; and construct a specified population model dataset as the corresponding second dataset based on preset specified population characteristics; the first dataset includes multiple first data records; the first data record includes a first training parameter vector and a first label probability; the second dataset includes multiple second data records; the second data record includes a second training parameter vector and a second label probability; the first and second training parameter vectors are both composed of the six categories of left ventricular parameters and the three categories of left atrial parameters; the first and second label probabilities are both label probabilities with a value of 0 or 1; the specified population characteristics include regional configuration, age group configuration and gender configuration; when the regional configuration is not empty, it is composed of one or more regions; when the age group configuration is not empty, it is composed of one or more age groups among the preset adolescent age group, middle-aged age group and elderly age group; the gender configuration includes male, female and both male and female;

[0010] Pre-training the hypertrophic cardiomyopathy risk prediction model based on the first data set;

[0011] After the pre-training is completed, fine-tuning the hypertrophic cardiomyopathy risk prediction model based on the second data set;

[0012] After fine-tuning is completed, the myocardial strain parameter set of the subject input by the user that meets the characteristics of the specified population is used as the corresponding myocardial strain parameter vector X and is input into the hypertrophic cardiomyopathy risk prediction model for prediction, and the predicted probability p output by this prediction is fed back to the current user; the myocardial strain parameter set of the subject is a myocardial strain parameter set of the subject that meets the characteristics of the specified population, and is composed of the six categories of left ventricular parameters and the three categories of left atrial parameters.

[0013] Preferably, the model input end of the hypertrophic cardiomyopathy risk prediction model is used to receive the myocardial strain parameter vector X, and the model output end is used to output the corresponding predicted probability p;

[0014] The hypertrophic cardiomyopathy risk prediction model includes a normalization module, a ventricular feature extraction module, an atrial feature extraction module, a feature splicing module, a multi-head self-attention module, a residual connection module, a feature mapping module and a Sigmoid function module;

[0015] The input end of the standardization module is connected to the input end of the model, and the first and second output ends are respectively connected to the input ends of the ventricle and atrial feature extraction modules; the output ends of the ventricle and atrial feature extraction modules are respectively connected to the first and second input ends of the feature splicing module; the output end of the feature splicing module is respectively connected to the input end of the multi-head self-attention module and the first input end of the residual connection module; the output end of the multi-head self-attention module is connected to the second input end of the residual connection module; the output end of the residual connection module is connected to the input end of the feature mapping module; the output end of the feature mapping module is connected to the input end of the Sigmoid function module; the output end of the Sigmoid function module is connected to the output end of the model;

[0016] The standardization module is used to extract the corresponding six categories of left ventricular parameters and the three categories of left atrial parameters from the myocardial strain parameter vector X; and based on the preset standardization rules, the nine categories of myocardial strain parameters are respectively standardized to obtain the corresponding six categories of left ventricular standardized parameters and three categories of left atrial standardized parameters; and the six categories of left ventricular standardized parameters are used to form the corresponding left ventricular parameter vector X V Send to the ventricular feature extraction module, the corresponding left atrial parameter vector X is composed of the three types of left atrial standardized parameters A Send to the atrial feature extraction module; wherein the standardization rule includes at least the z-score rule; the six categories of left ventricular standardization parameters include GLS standardization parameters, GCS standardization parameters, GRS standardization parameters, cTSD standardization parameters, PSSR standardization parameters and PDSR parameter standardization parameters; the three categories of left atrial standardization parameters include Ee standardization parameters, sRe standardization parameters, and sRa standardization parameters; the left ventricular parameter vector X V It consists of 6 parameters Composition, 1≤index i≤6, 6 parameters One-to-one correspondence with the six types of left ventricular standardized parameters; the left atrial parameter vector X A By 3 parameters Composition, 1≤indexj≤3, 3 parameters One-to-one correspondence with the three types of left atrial standardized parameters;

[0017] The ventricular feature extraction module is composed of two layers of fully connected layers connected in sequence, and a LeakyReLU activation function is connected after each fully connected layer; the ventricular feature extraction module is used to extract the left ventricular parameter vector X V Perform feature extraction to obtain the corresponding feature vector H V Sending to the feature splicing module;

[0018] The eigenvector H V The reasoning process is:

[0019] H V =LeakyReLU(W V,2 LeakyReLU(W V,1 X V +b V,1 )+b V,2 ),

[0020] W V,1 、b V,1 are the weight matrix parameters and offset vector parameters of the first fully connected layer of the ventricular feature extraction module, W V,2 、b V,2 The weight matrix parameters and offset vector parameters of the second fully connected layer of the ventricular feature extraction module; the left ventricular parameter vector X V The shape is 6×1, and the weight matrix parameter W V,1 The shape is C V,1 ×6, the offset vector parameter b V,1 The shape is C V,1 ×1, LeakyReLU(W V,1 X V +b V,1 ) has an output vector shape of C V,1 ×1; the weight matrix parameter W V,2 The shape is C V,2 ×C V,1 , the offset vector parameter b V,2 The shape is C V,2 ×1, the eigenvector H V The shape is C V,2 ×1;C V,1 、C V,2 are two preset feature vector dimensions, and 6<C V,1 <C V,2 ;

[0021] The atrial feature extraction module is composed of two layers of fully connected layers connected in sequence, and a LeakyReLU activation function is connected after each fully connected layer; the atrial feature extraction module is used to extract the left atrial parameter vector X APerform feature extraction to obtain the corresponding feature vector H A Sending to the feature splicing module;

[0022] The eigenvector H A The reasoning process is:

[0023] H A =LeakyReLU(W A,2 LeakyReLU(W A,1 X A +b A,1 )+b A,2 ),

[0024] W A,1 、b A,1 are the weight matrix parameters and offset vector parameters of the first fully connected layer of the atrial feature extraction module, W A,2 、b A,2 The weight matrix parameters and offset vector parameters of the second fully connected layer of the atrial feature extraction module; the left atrial parameter vector X A The shape is 3×1, and the weight matrix parameter W A,1 The shape is C A,1 ×3, the offset vector parameter b A,1 The shape is C A,1 ×1, LeakyReLU(W A,1 X A +b A,1 ) has an output vector shape of C A,1 ×1; the weight matrix parameter W A,2 The shape is C A,2 ×C A,1 , the offset vector parameter b A,2 The shape is C A,2 ×1, the eigenvector H A The shape is C A,2 ×1;C A,1 、C A,2 are two preset feature vector dimensions, and 3<C A,1 <C A,2 , C A,1 <C V,1 , C A,2 <C V,2 ;

[0025] The feature splicing module is used to V and the eigenvector H A Perform vector splicing to obtain the corresponding splicing vector H C Send to the multi-head self-attention module and the residual connection module; the splicing vector HC The shape is C C ×1, C C is a preset feature vector dimension, C C =C V,2 +C A,2 ;

[0026] The multi-head self-attention module consists of N head It consists of a parallel single-head attention unit and a multi-head fusion unit; N head is the preset total number of single heads, the total number of single heads N head is a number greater than 1 and can be represented by the dimension C of the feature vector C A positive integer that is divisible by ; each of the single-head attention units corresponds to an independent set of query / key / value matrix parameters 1≤indexn≤N head , each set of query / key / value matrix parameters The shape is C head ×C C , C head is a preset feature vector dimension, C head =C C / N head ; The multi-head fusion unit corresponds to a fusion matrix parameter W R , the fusion matrix parameter W R The shape is C C ×C C ;

[0027] The multi-head self-attention module is used to C Perform multi-head self-attention operation to obtain the corresponding feature vector H R Send to the residual connection module, specifically: each of the single-head attention units according to the corresponding query / key / value matrix parameters For the splicing vector H C Perform a self-attention operation to obtain the corresponding single-head feature vector H n Send to the multi-head fusion unit; the multi-head fusion unit sends head The single-headed eigenvector H n Perform feature splicing to obtain the corresponding splicing vector H all-head , and based on the fusion matrix parameter W R For the splicing vector H all-head Perform feature mapping to obtain the corresponding feature vector H R Sending to the residual connection module;

[0028] Each of the single-head feature vectors H n The reasoning process is:

[0029]

[0030] Q n , K n 、V n are the query vector, key vector, and value vector of the nth single-head attention unit, is the key vector K n The transposed vector of , softmax() is the Softmax function; the query, key, value vector Q n , K n 、V n The shape is C head ×1, the transposed vector The shape is 1×C head , the single-head eigenvector H n The shape is C head ×1;

[0031] The splicing vector H all-head The shape is (C head ×N head )×1=C C ×1;

[0032] The eigenvector H R The reasoning process is:

[0033] H R =W R H all-head ,

[0034] The eigenvector H R The shape is C C ×1;

[0035] The residual connection module is used to connect the splicing vector H in a residual connection manner. C and the eigenvector H R Fusion is performed to obtain the corresponding feature vector H R+C Sending to the feature mapping module;

[0036] The eigenvector H R+C The reasoning process is:

[0037] H R+c =H c +H R ,

[0038] The eigenvector H R+C The shape is C C ×1;

[0039] The feature mapping module is composed of two layers of fully connected layers connected in sequence, and a Swish activation function is connected after the first fully connected layer; the feature mapping module is used to calculate the feature vector H R+C Calculate the confidence score and send the corresponding confidence score z to the Sigmoid function module;

[0040] The reasoning process of the confidence score z is:

[0041] z=W out,2 Swish(W out,1 H R+C +b out,1 )+b out,2 ,

[0042] W out,1 、b out,1 are the weight matrix parameters and offset vector parameters of the first fully connected layer of the feature mapping module, W out,2 、b out,2 The weight matrix parameters and offset vector parameters of the second fully connected layer of the feature mapping module are as follows: out,1 The shape is C out,1 ×C C , the offset vector parameter b V,1 The shape is C out,1 ×1, Swish(W out,1 H R+C +b out,1 ) has an output vector shape of C out,1 ×1; the weight matrix parameter W out,2 The shape is C out,2 ×C out,1 , the offset vector parameter b out,2 The shape is C out,2 ×1, C out,1 、C out,2 are two preset feature vector dimensions, C out,1 >C C , C out,2 =1; the confidence score z is a scalar;

[0043] The Sigmoid function module is used to bring the confidence score z into the Sigmoid function for calculation and output the calculation result as the corresponding predicted probability p.

[0044] Preferably, the constructing of a generalized population model dataset as the corresponding first dataset specifically includes:

[0045] Step 31, using a preset multi-category data collection channel, collect big data of cardiac magnetic resonance images of multiple groups of people in multiple regions and age groups to obtain corresponding original image sets;

[0046] Among them, the multiple types of data collection channels include at least public medical imaging data sets, public or authorized medical institution databases for data use, and public or authorized medical imaging sharing platforms for data use; the multiple regions include at least two or more regions; the multiple age groups include at least adolescent age groups, middle-aged age groups, and elderly age groups; the multiple types of people include at least healthy people who do not suffer from any heart disease, other patients who do not suffer from hypertrophic cardiomyopathy but suffer from other heart diseases, and target patient groups with hypertrophic cardiomyopathy; the genders of the multiple types of people should cover both males and females; the original image set includes multiple original image records; the original image record includes a first image, a first region, a first gender, a first age group, and a first population type; the first image is a cardiac magnetic resonance image; the first region is one of the multiple regions; the first gender includes males and females; the first age group includes adolescent age groups, middle-aged age groups, and elderly age groups; the first population type includes healthy people who do not suffer from any heart disease, other patients who do not suffer from hypertrophic cardiomyopathy but suffer from other heart diseases, and target patient groups with hypertrophic cardiomyopathy;

[0047] Step 32: Use the first image of each original image record in the original image set as the corresponding current image; quantitatively evaluate the six categories of left ventricular parameters and the three categories of left atrial parameters corresponding to the current image based on preset myocardial strain parameter evaluation software, and form a corresponding first training parameter vector with the evaluated six categories of left ventricular parameters and the three categories of left atrial parameters; identify whether the first population type corresponding to the current image is a target patient population with hypertrophic cardiomyopathy, and if so, set a corresponding first label probability to 1; otherwise, set a corresponding first label probability to 0; and form a corresponding first data record with the first training parameter vector corresponding to the current image and the first label probability; and form a corresponding first data set with all the obtained first data records;

[0048] The myocardial strain parameter assessment software includes at least CVI42 software, CAAS MR Strain software, and Segment software.

[0049] Preferably, the step of constructing a designated population model dataset based on preset designated population characteristics as the corresponding second dataset specifically includes:

[0050] Extracting the corresponding region configuration, age group configuration, and gender configuration from the specified population characteristics;

[0051] and identifying whether the region configuration is empty; if so, using the original image set as the corresponding first image set; if not, forming the corresponding first image set from the original image records of the first region in the original image set that satisfy the region configuration;

[0052] and identifying whether the age group configuration is empty; if so, using the first image set as the corresponding second image set; if not, forming the corresponding second image set from the original image records in the first image set whose first age group meets the age group configuration;

[0053] and identifying the gender configuration; if the gender configuration is male, the corresponding third image set is composed of the original image records in the second image set in which the first gender is male; if the gender configuration is female, the corresponding third image set is composed of the original image records in the second image set in which the first gender is female; if the gender configuration is unisex, the second image set is used as the corresponding third image set;

[0054] The first data records corresponding to each of the original image records of the third image set in the first data set are extracted as a corresponding second data record; and all the obtained second data records form a corresponding second data set.

[0055] Preferably, the pre-training of the hypertrophic cardiomyopathy risk prediction model based on the first data set specifically includes:

[0056] Step 51: randomly segmenting the first data set into sub-datasets based on a preset first segmentation ratio to obtain corresponding first training sets and first evaluation sets; and using the full model parameter set of the hypertrophic cardiomyopathy risk prediction model as the corresponding current model parameter set; and segmenting the first training set into multiple first batch record sets according to a preset maximum number of records in a single batch;

[0057] wherein both the first training set and the first evaluation set consist of a plurality of the first data records; the ratio of the total number of records in the first training set to the total number of records in the first evaluation set satisfies the first segmentation ratio; and the total number of the first data records in the first batch of records is less than or equal to the maximum number of records in a single batch;

[0058] Step 52: taking the first of the first record sets as the corresponding current record set;

[0059] Step 53: Perform a round of traversal on all the first data records in the current record set; and during this round of traversal, use the first data record currently traversed as the corresponding current training record; and input the first training parameter vector of the current training record as the current myocardial strain parameter vector X into the hypertrophic cardiomyopathy risk prediction model for prediction, and use the predicted probability p output by this prediction as the corresponding first predicted probability; and form a corresponding first prediction-label pair from the current first predicted probability and the first label probability of the current training record; and at the end of this round of traversal, bring all the first prediction-label pairs obtained in this round of traversal into a preset first model loss function to calculate and obtain a corresponding first loss value;

[0060] Wherein, the first model loss function is implemented based on L1 loss function, L2 loss function or binary cross entropy loss function;

[0061] Step 54: Identify whether the first loss value satisfies a preset first loss value range. If the first loss value satisfies the first loss value range, identify whether the current record set is the last first record set of the first training set. If so, proceed to step 55. If not, use the next first record set as the new current record set and return to step 53. If the first loss value does not satisfy the first loss value range, perform a round of modulation on the current model parameter set based on a preset first model optimizer in a direction that minimizes the first model loss function, and return to step 53 after this round of modulation.

[0062] Wherein, the first model optimizer includes at least AdamW optimizer, Adam optimizer, and SGD optimizer;

[0063] Step 55: Perform a round of traversal on all the first data records in the first evaluation set; and during this round of traversal, use the currently traversed first data record as the corresponding current evaluation record; and input the first training parameter vector of the current evaluation record as the current myocardial strain parameter vector X into the hypertrophic cardiomyopathy risk prediction model for prediction, and use the predicted probability p output by this prediction as the corresponding second predicted probability; and form a corresponding second prediction-label pair from the current second predicted probability and the first label probability of the current evaluation record; and at the end of this round of traversal, bring all the obtained second prediction-label pairs into the preset first model evaluation function to calculate and obtain the corresponding first evaluation value;

[0064] Wherein, the first model evaluation function is implemented based on MAE function, MSE function or RMSE function;

[0065] Step 56, identifying whether the first evaluation value meets the preset first evaluation value range; if not, returning to step 52; if so, confirming that the model pre-training is completed.

[0066] Preferably, fine-tuning the hypertrophic cardiomyopathy risk prediction model based on the second data set specifically includes:

[0067] Step 61, based on a preset second segmentation ratio, the second data set is randomly segmented into sub-data sets to obtain corresponding second training sets and second evaluation sets; and the fusion matrix parameter W of the multi-head self-attention module of the hypertrophic cardiomyopathy risk prediction model is used. R And the weight matrix parameter W of the feature mapping module out,1 、W out,2 Forming a corresponding current fine-tuning parameter set; and dividing the second training set into multiple second batch record sets according to a preset maximum number of records in a single batch;

[0068] wherein both the second training set and the second evaluation set are composed of a plurality of second data records; the ratio of the total number of records in the second training set to the total number of records in the second evaluation set satisfies the second split ratio; and the total number of the second data records in the second batch of records is less than or equal to the maximum number of records in a single batch;

[0069] Step 62, taking the first of the second batch of record sets as the corresponding current record set;

[0070] Step 63: Perform a round of traversal on all the second data records in the current record set; and during this round of traversal, use the currently traversed second data record as the corresponding current training record; and input the second training parameter vector of the current training record as the current myocardial strain parameter vector X into the hypertrophic cardiomyopathy risk prediction model for prediction, and use the predicted probability p output by this prediction as the corresponding third predicted probability; and form a corresponding third prediction-label pair by combining the current third predicted probability and the second label probability of the current training record; and at the end of this round of traversal, bring all the third prediction-label pairs obtained in this round of traversal into the preset second model loss function to calculate the corresponding second loss value;

[0071] Wherein, the second model loss function is implemented based on the binary cross entropy loss function;

[0072] Step 64: Identify whether the second loss value satisfies a preset second loss value range. If the second loss value satisfies the second loss value range, identify whether the current record set is the last second batch of record sets in the second training set. If so, proceed to step 65. If not, use the next second batch of record sets as the new current record set and return to step 63. If the second loss value does not satisfy the second loss value range, perform a round of fine-tuning on the current model parameter set based on a preset second model optimizer in a direction to minimize the second model loss function, and return to step 63 after this round of fine-tuning.

[0073] Wherein, the second model optimizer includes at least AdamW optimizer, Adam optimizer, and SGD optimizer;

[0074] Step 65: Perform a round of traversal on all the second data records of the second evaluation set; and during this round of traversal, use the currently traversed second data record as the corresponding current evaluation record; and input the second training parameter vector of the current evaluation record as the current myocardial strain parameter vector X into the hypertrophic cardiomyopathy risk prediction model for prediction, and use the predicted probability p output by this prediction as the corresponding fourth predicted probability; and form a corresponding fourth prediction-label pair by the current fourth predicted probability and the second label probability of the current evaluation record; and at the end of this round of traversal, calculate the accuracy, precision, recall and F1 score based on all the obtained fourth prediction-label pairs to obtain the corresponding first accuracy, first precision, first recall and first F1 score;

[0075] Step 66: Identify whether the first accuracy, the first precision, the first recall, and the first F1 score each satisfy the corresponding first accuracy range, first precision range, first recall range, and first F1 score range; if not, return to step 62; if so, confirm that the model fine-tuning is completed.

[0076] A second aspect of an embodiment of the present invention provides a device for implementing the processing method of the hypertrophic cardiomyopathy risk prediction model described in the first aspect, the device comprising: a model construction module, a data set construction module, a model pre-training module, a model fine-tuning module, and a model application module;

[0077] The model construction module is used to construct a deep learning model for predicting the risk probability of hypertrophic cardiomyopathy, which is recorded as the corresponding hypertrophic cardiomyopathy risk prediction model; the hypertrophic cardiomyopathy risk prediction model is used to predict the risk probability of hypertrophic cardiomyopathy based on the myocardial strain parameter vector X input by the model and output the corresponding predicted probability p; the myocardial strain parameter vector X is composed of six categories of left ventricular parameters and three categories of left atrial parameters, the six categories of left ventricular parameters including GLS parameters, GCS parameters, GRS parameters, cTSD parameters, PSSR parameters, and PDSR parameters; the three categories of left atrial parameters including Ee parameters, sRe parameters, and sRa parameters; the predicted probability p is a probability value between 0 and 1;

[0078] The data set construction module is used to construct a generalized population model data set as the corresponding first data set; and to construct a specified population model data set as the corresponding second data set based on the preset specified population characteristics; the first data set includes multiple first data records; the first data record includes a first training parameter vector and a first label probability; the second data set includes multiple second data records; the second data record includes a second training parameter vector and a second label probability; the first and second training parameter vectors are both composed of the six categories of left ventricular parameters and the three categories of left atrial parameters; the first and second label probabilities are both label probabilities with a value of 0 or 1; the specified population characteristics include regional configuration, age group configuration and gender configuration; when the regional configuration is not empty, it is composed of one or more regions; when the age group configuration is not empty, it is composed of one or more age groups among the preset adolescent age group, middle-aged age group and elderly age group; the gender configuration includes male, female and both male and female;

[0079] The model pre-training module is used to pre-train the hypertrophic cardiomyopathy risk prediction model based on the first data set;

[0080] The model fine-tuning module is used to fine-tune the hypertrophic cardiomyopathy risk prediction model based on the second data set after the pre-training is completed;

[0081] The model application module is configured to, after fine-tuning is completed, input the myocardial strain parameter set of the subject that meets the characteristics of the specified population, as the corresponding myocardial strain parameter vector X, into the hypertrophic cardiomyopathy risk prediction model for prediction, and feed back the predicted probability p output by this prediction to the current user; the myocardial strain parameter set of the subject is a myocardial strain parameter set of the subject that meets the characteristics of the specified population, and is composed of the six categories of left ventricular parameters and the three categories of left atrial parameters.

[0082] A third aspect of an embodiment of the present invention provides an electronic device, including: a memory, a processor, and a transceiver;

[0083] The processor is configured to be coupled to the memory, read and execute instructions in the memory, so as to implement the method steps described in the first aspect above;

[0084] The transceiver is coupled to the processor, and the processor controls the transceiver to send and receive messages.

[0085] A fourth aspect of an embodiment of the present invention provides a computer-readable storage medium, which stores computer instructions. When the computer instructions are executed by a computer, the computer executes the instructions of the method described in the first aspect above.

[0086] The embodiment of the present invention provides a processing method, device, electronic device and computer-readable storage medium for a hypertrophic cardiomyopathy risk prediction model. As can be seen from the above content, the embodiment of the present invention constructs a hypertrophic cardiomyopathy risk prediction model that can predict risks based on six types of left ventricular parameters (GLS parameters, GCS parameters, GRS parameters, cTSD parameters, PSSR parameters, PDSR parameters) and three types of left atrial parameters (Ee parameters, sRe parameters, sRa parameters); and constructs a generalized population model data set (first data set) and a specified population model data set (second data set) through big data collection; and first pre-trains the hypertrophic cardiomyopathy risk prediction model based on the first data set and then fine-tunes the hypertrophic cardiomyopathy risk prediction model based on the second data set; after the model fine-tuning is completed, the hypertrophic cardiomyopathy risk prediction model is used to make predictions based on the myocardial strain parameters of the subject. The embodiment of the present invention not only improves the prediction accuracy, but also improves the prediction stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 A schematic diagram of a processing method for a hypertrophic cardiomyopathy risk prediction model provided in Example 1 of the present invention;

[0088] Figure 2 This is a module diagram of the hypertrophic cardiomyopathy risk prediction model provided in Example 1 of the present invention;

[0089] Figure 3 This is a module structure diagram of a processing device for a hypertrophic cardiomyopathy risk prediction model provided in Example 2 of the present invention;

[0090] Figure 4 This is a structural diagram of an electronic device provided in Example 3 of the present invention. DETAILED DESCRIPTION

[0091] To make the objectives, technical solutions, and advantages of the present invention more apparent, the present invention will be further described in detail below with reference to the accompanying drawings. It should be understood that the embodiments described herein are merely some, rather than all, of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are intended to fall within the scope of protection of the present invention.

[0092] The first embodiment of the present invention provides a method for processing a hypertrophic cardiomyopathy risk prediction model, such as Figure 1 As shown in the schematic diagram of a processing method for a hypertrophic cardiomyopathy risk prediction model provided in Example 1 of the present invention, the method mainly includes the following steps:

[0093] Step 1: Build a deep learning model for predicting the risk probability of hypertrophic cardiomyopathy, which is recorded as the corresponding hypertrophic cardiomyopathy risk prediction model.

[0094] Here, the hypertrophic cardiomyopathy risk prediction model of the embodiment of the present invention is used to predict the risk probability of hypertrophic cardiomyopathy based on the myocardial strain parameter vector X input by the model and output the corresponding predicted probability p; Figure 2 The following is a schematic diagram of the module of the hypertrophic cardiomyopathy risk prediction model provided in Example 1 of the present invention. The myocardial strain parameter vector X is composed of six categories of left ventricular parameters and three categories of left atrial parameters; the six categories of left ventricular parameters include GLS parameters, GCS parameters, GRS parameters, cTSD parameters, PSSR parameters, and PDSR parameters; and the three categories of left atrial parameters include Ee parameters, sRe parameters, and sRa parameters. The predicted probability p is a probability value between 0 and 1.

[0095] like Figure 2 As shown, the model input end of the hypertrophic cardiomyopathy risk prediction model is used to receive the myocardial strain parameter vector X, and the model output end is used to output the corresponding prediction probability p.

[0096] The model components of the hypertrophic cardiomyopathy risk prediction model include: normalization module, ventricular feature extraction module, atrial feature extraction module, feature splicing module, multi-head self-attention module, residual connection module, confidence assessment module and Sigmoid function module.

[0097] The connection relationship of the model components of the hypertrophic cardiomyopathy risk prediction model is as follows: the input end of the normalization module is connected to the model input end, and the first and second output ends are connected to the input ends of the ventricular and atrial feature extraction modules respectively; the output ends of the ventricular and atrial feature extraction modules are connected to the first and second input ends of the feature splicing module respectively; the output end of the feature splicing module is connected to the input end of the multi-head self-attention module and the first input end of the residual connection module respectively; the output end of the multi-head self-attention module is connected to the second input end of the residual connection module; the output end of the residual connection module is connected to the input end of the confidence assessment module; the output end of the confidence assessment module is connected to the input end of the Sigmoid function module; and the output end of the Sigmoid function module is connected to the output end of the model.

[0098] The functions of the model components of the hypertrophic cardiomyopathy risk prediction model are shown below.

[0099] 1) Standardized modules:

[0100] The standardization module of the embodiment of the present invention is used to extract the corresponding six categories of left ventricular parameters and three categories of left atrial parameters from the myocardial strain parameter vector X; and based on the preset standardization rules, the nine categories of myocardial strain parameters are respectively standardized to obtain the corresponding six categories of left ventricular standardized parameters and three categories of left atrial standardized parameters; and the six categories of left ventricular standardized parameters are used to form the corresponding left ventricular parameter vector X V Send to the ventricular feature extraction module, the corresponding left atrial parameter vector X is composed of three types of left atrial standardized parameters A Send to the atrial feature extraction module.

[0101] Among them, the standardization rules at least include the z-score rule; the six categories of left ventricular standardization parameters include GLS standardization parameters, GCS standardization parameters, GRS standardization parameters, cTSD standardization parameters, PSSR standardization parameters and PDSR parameter standardization parameters; the three categories of left atrial standardization parameters include Ee standardization parameters, sRe standardization parameters, and sRa standardization parameters; the left ventricular parameter vector X V It consists of 6 parameters Composition, 1≤index i≤6, 6 parameters One-to-one correspondence with the six types of left ventricular standardized parameters; left atrial parameter vector X A By 3 parameters Composition, 1≤indexj≤3, 3 parameters One-to-one correspondence with the three types of left atrial standardized parameters.

[0102] 2) Ventricular feature extraction module:

[0103] The ventricular feature extraction module of the embodiment of the present invention is composed of two layers of fully connected layers connected in sequence, and a LeakyReLU activation function is connected after each fully connected layer. The ventricular feature extraction module is used to extract the left ventricular parameter vector X V Perform feature extraction to obtain the corresponding feature vector H V Send to the feature stitching module.

[0104] Here, the eigenvector H V The reasoning process is:

[0105] H V =LeakyReLU(W V,2 LeakyReLU(W V,1 X V +b V,1 )+b V,2 );

[0106] Among them, W V,1 、b V,1 are the weight matrix parameters and offset vector parameters of the first fully connected layer of the ventricular feature extraction module, W V,2 、b V,2 The weight matrix parameters and offset vector parameters of the second fully connected layer of the ventricular feature extraction module; the left ventricular parameter vector X V The shape is 6×1, and the weight matrix parameter W V,1 The shape is C V,1 ×6, offset vector parameter b V,1 The shape is C V,1 ×1, LeakyReLU(W V,1 X V +b V,1 ) has an output vector shape of C V,1 ×1; weight matrix parameter W V,2 The shape is C V,2 ×C V,1 , offset vector parameter b V,2 The shape is C V,2 ×1, eigenvector H V The shape is C V,2 ×1;C V,1 、C V,2 are two preset feature vector dimensions, and 6<C V,1 <C V,2 , for example C V,1 =12, C V,2 =16.

[0107] 3) Atrial feature extraction module:

[0108] The atrial feature extraction module of the embodiment of the present invention is composed of two layers of fully connected layers connected in sequence, and a LeakyReLU activation function is connected after each fully connected layer. The atrial feature extraction module is used to extract the left atrial parameter vector X A Perform feature extraction to obtain the corresponding feature vector H A Send to the feature stitching module.

[0109] Here, the eigenvector H A The reasoning process is:

[0110] H A =LeakyReLU(W A,2 LeakyReLU(W A,1 X A +b A,1 )+b A,2 );

[0111] Among them, W A,1 、b A,1 are the weight matrix parameters and offset vector parameters of the first fully connected layer of the atrial feature extraction module, W A,2 、b A,2 The weight matrix parameters and offset vector parameters of the second fully connected layer of the atrial feature extraction module; the left atrial parameter vector X A The shape is 3×1, and the weight matrix parameter W A,1 The shape is C A,1 ×3, offset vector parameter b A,1 The shape is C A,1 ×1, LeakyReLU(W A,1 X A +b A,1 ) has an output vector shape of C A,1 ×1; weight matrix parameter W A,2 The shape is C A,2 ×C A,1 , offset vector parameter b A,2 The shape is C A,2 ×1, eigenvector H A The shape is C A,2 ×1;C A,1 、C A,2 are two preset feature vector dimensions, and 3<C A,1 <C A,2 , C A,1 <C V,1 , C A,2 <C V,2 , for example C A,1 =6, C A,2 =8.

[0112] 4) Feature splicing module:

[0113] The feature concatenation module of the embodiment of the present invention is used to concatenate the feature vector H V and the eigenvector H A Perform vector splicing to obtain the corresponding splicing vector H C Send to the multi-head self-attention module and residual connection module.

[0114] Here, the splicing vector H C The shape is C C ×1, C C is a preset feature vector dimension, C C =C V,2 +C A,2 .

[0115] 5) Multi-head self-attention module:

[0116] The multi-head self-attention module of the embodiment of the present invention is composed of N head It consists of a parallel single-head attention unit and a multi-head fusion unit; where N head The total number of preset single heads, N head is a number greater than 1 and can be C A divisible positive integer, such as N head =4; each single-head attention unit corresponds to an independent set of query / key / value matrix parameters 1≤indexn≤N head , each set of query / key / value matrix parameters The shape is C head ×C C , C head is a preset feature vector dimension, C head =C C / N head ; Multi-head fusion unit corresponds to a fusion matrix parameter W R , fusion matrix parameter W R The shape is C C ×C C .

[0117] The multi-head self-attention module is used to concatenate the vector H C Perform multi-head self-attention operation to obtain the corresponding feature vector H R Send to the residual connection module, specifically: each single-head attention unit according to the corresponding query / key / value matrix parameters The concatenation vector H C Perform a self-attention operation to obtain the corresponding single-head feature vector H n Send to the multi-head fusion unit; the multi-head fusion unit sends head A single-headed eigenvector H nPerform feature splicing to obtain the corresponding splicing vector H all-head , and based on the fusion matrix parameter W R The concatenation vector H all-head Perform feature mapping to obtain the corresponding feature vector H R Send to the residual connection module.

[0118] Here, each single-headed eigenvector H n The reasoning process is:

[0119]

[0120] Among them, Q n , K n 、V n are the query vector, key vector, and value vector of the n-th single-head attention unit, is the key vector K n The transposed vector of , softmax() is the Softmax function; query, key, value vector Q n , K n 、V n The shape is C head ×1, The shape is 1×C head , single-headed eigenvector H n The shape is C head ×1.

[0121] In addition, the splicing vector H all-head The shape is (C head ×N head )×1=C C ×1.

[0122] In addition, the eigenvector H R The reasoning process is:

[0123] H R =W R H all-head ;

[0124] Among them, the eigenvector H R The shape is C C ×1.

[0125] 6) Residual Connection Module:

[0126] The residual connection module of the embodiment of the present invention is used to connect the splicing vector H in a residual connection manner. C and the eigenvector H R Fusion is performed to obtain the corresponding feature vector H R+C Send to the confidence assessment module.

[0127] Here, the eigenvector H R+C The reasoning process is:

[0128] H R+c =H c +H R ;

[0129] Among them, the eigenvector H R+C The shape is C C ×1.

[0130] 7) Confidence Assessment Module:

[0131] The confidence evaluation module of the embodiment of the present invention is composed of two layers of fully connected layers connected in sequence, and a Swish activation function is connected after the first fully connected layer. The confidence evaluation module is used to calculate the confidence of the feature vector H according to the feature vector H. R+C The confidence score is calculated and the corresponding confidence score z is sent to the Sigmoid function module.

[0132] Here, the reasoning process of the confidence score z is:

[0133] z=W out,2 Swish(W out,1 H R+c +b out,1 )+b out,2 ;

[0134] Among them, W out,1 、b out,1 are the weight matrix parameters and offset vector parameters of the first fully connected layer of the confidence assessment module, W out,2 、b out,2 The weight matrix parameters and offset vector parameters of the second fully connected layer of the confidence assessment module; the weight matrix parameters W out,1 The shape is C out,1 ×C C , offset vector parameter b V,1 The shape is C out,1 ×1, Swish(W out,1 H R+C +b out,1 ) has an output vector shape of C out,1 ×1; weight matrix parameter W out,2 The shape is C out,2 ×C out,1 , offset vector parameter b out,2 The shape is C out,2 ×1, C out,1 、C out,2 are two preset feature vector dimensions, C out,1 >C C For example, C out,1=32; In addition, because the confidence score z is a scalar, the feature vector dimension C out,2 =1.

[0135] 8) Sigmoid function module:

[0136] The Sigmoid function module of the embodiment of the present invention is used to bring the confidence score z into the Sigmoid function for calculation and output the calculation result as the corresponding prediction probability p.

[0137] It should be noted that the confidence score z can be considered an assessment score used to assess the confidence that the current sample (i.e., the myocardial strain parameter vector X input to the current model) is a positive sample (i.e., the current myocardial strain parameter vector X is the myocardial strain parameter vector of a patient with hypertrophic cardiomyopathy). A higher confidence score z indicates a higher probability that the current sample is a positive sample. Conversely, a lower confidence score z indicates a lower probability that the current sample is a positive sample and a higher probability that it is a negative sample. The predicted probability p calculated from the confidence score z can be considered the probability that the current sample is a positive sample. A higher predicted probability p indicates a greater probability that the current sample is a positive sample. Conversely, a lower predicted probability p indicates a lower probability that the current sample is a positive sample and a higher probability that it is a negative sample.

[0138] Step 2: construct a generalized population model dataset as the corresponding first dataset; and construct a specified population model dataset based on preset specified population characteristics as the corresponding second dataset;

[0139] The first data set includes a plurality of first data records; the first data record includes a first training parameter vector and a first label probability; the second data set includes a plurality of second data records; the second data record includes a second training parameter vector and a second label probability; the first and second training parameter vectors are both composed of six categories of left ventricular parameters and three categories of left atrial parameters; and the first and second label probabilities are both label probabilities with a value of 0 or 1;

[0140] The specified population characteristics include region configuration, age group configuration, and gender configuration. If the region configuration is not empty, it consists of one or more regions. If the age group configuration is not empty, it consists of one or more preset age groups among the youth age group, middle-aged age group, and elderly age group. Gender configuration includes male, female, and both male and female.

[0141] The current step 2 specifically includes:

[0142] Step 21, constructing a generalized population model dataset as the corresponding first dataset;

[0143] Specifically, step 211 includes collecting big data of cardiac magnetic resonance images of multiple groups of people in multiple regions and age groups through pre-set multiple data collection channels to obtain corresponding original image sets;

[0144] Here, the multiple types of data collection channels in the embodiment of the present invention include at least public medical imaging data sets, medical institution databases that are public or have obtained data use authorization, and medical imaging sharing platforms that are public or have obtained data use authorization;

[0145] The multiple regions in the embodiment of the present invention include at least two or more regions;

[0146] The multiple age groups in the embodiment of the present invention include at least a youth age group, a middle-aged age group, and an elderly age group;

[0147] The multiple groups of people in the embodiment of the present invention include at least healthy people who do not suffer from any heart disease, other patients who do not suffer from hypertrophic cardiomyopathy but suffer from other heart diseases, and a target patient population suffering from hypertrophic cardiomyopathy;

[0148] The gender of the multiple groups of people in the embodiment of the present invention should cover both male and female;

[0149] The original image set of the embodiment of the present invention includes multiple original image records; the original image records include a first image, a first region, a first gender, a first age group, and a first population type; the first image is a cardiac magnetic resonance image; the first region is one of multiple regions; the first gender includes male and female; the first age group includes adolescent age group, middle-aged age group, and elderly age group; the first population type includes healthy people without any heart disease, other patients without hypertrophic cardiomyopathy but with other heart diseases, and a target patient population with hypertrophic cardiomyopathy;

[0150] Step 212: The first image of each original image record in the original image set is used as the corresponding current image; the six categories of left ventricular parameters and three categories of left atrial parameters corresponding to the current image are quantitatively evaluated based on the preset myocardial strain parameter evaluation software, and the evaluated six categories of left ventricular parameters and three categories of left atrial parameters are used to form a corresponding first training parameter vector; the first population type corresponding to the current image is identified as a target patient population with hypertrophic cardiomyopathy, and if so, a corresponding first label probability is set to 1; otherwise, a corresponding first label probability is set to 0; the first training parameter vector corresponding to the current image and the first label probability are used to form a corresponding first data record; and all the obtained first data records are used to form a corresponding first data set;

[0151] Here, the myocardial strain parameter assessment software of the embodiment of the present invention includes at least CVI42 software, CAAS MRStrain software, and Segment software;

[0152] Step 22: constructing a designated population model dataset as a corresponding second dataset based on preset designated population characteristics;

[0153] Specifically, it includes: step 221, extracting corresponding region configuration, age group configuration and gender configuration from the specified population characteristics;

[0154] Step 222: Identify whether the region configuration is empty; if so, use the original image set as the corresponding first image set; if not, form the corresponding first image set from the original image records of the first region in the original image set that meets the region configuration;

[0155] Step 223, it is determined whether the age group configuration is empty; if so, the first image set is used as the corresponding second image set; if not, the corresponding second image set is composed of the original image records of the first age group in the first image set that meet the age group configuration;

[0156] Step 224 , the gender configuration is identified; if the gender configuration is male, the corresponding third image set is composed of the original image records of the second image set whose first gender is male; if the gender configuration is female, the corresponding third image set is composed of the original image records of the second image set whose first gender is female; if the gender configuration is unisex, the second image set is used as the corresponding third image set;

[0157] In step 225 , first data records corresponding to the respective original image records of the third image set in the first data set are extracted as corresponding second data records; and all the obtained second data records are combined into a corresponding second data set.

[0158] Step 3, pre-training a hypertrophic cardiomyopathy risk prediction model based on the first data set;

[0159] Specifically, the method includes: step 31, randomly dividing the first data set into sub-datasets based on a preset first division ratio to obtain a corresponding first training set and a first evaluation set; and using the full model parameter set of the hypertrophic cardiomyopathy risk prediction model as the corresponding current model parameter set; and dividing the first training set into multiple first batch record sets according to a preset maximum number of records in a single batch;

[0160] The first split ratio is a preset ratio parameter, such as 8:2; the first training set and the first evaluation set are both composed of multiple first data records; the ratio of the total number of records in the first training set to the total number of records in the first evaluation set satisfies the first split ratio; the maximum number of records in a single batch is a preset positive integer, such as 10; the total number of first data records in the first record set is less than or equal to the maximum number of records in a single batch;

[0161] Step 32, taking the first batch of record sets as the corresponding current record set;

[0162] Step 33: Perform a round of traversal on all first data records in the current record set; during this round of traversal, use the currently traversed first data record as the corresponding current training record; input the first training parameter vector of the current training record as the current myocardial strain parameter vector X into the hypertrophic cardiomyopathy risk prediction model for prediction, and use the predicted probability p output by this prediction as the corresponding first predicted probability; and form a corresponding first prediction-label pair from the current first predicted probability and the first label probability of the current training record; and at the end of this round of traversal, bring all first prediction-label pairs obtained in this round of traversal into the preset first model loss function to calculate the corresponding first loss value;

[0163] The first model loss function is implemented based on the L1 loss function, the L2 loss function or the binary cross entropy loss function;

[0164] Step 34: Identify whether the first loss value satisfies a preset first loss value range. If so, identify whether the current record set is the last first record set of the first training set. If so, proceed to step 35. If not, use the next first record set as the new current record set and return to step 33. If the first loss value does not satisfy the first loss value range, perform a round of modulation on the current model parameter set based on a preset first model optimizer in a direction that minimizes the first model loss function. After this round of modulation, return to step 33.

[0165] The first loss value range is a preset numerical range; the first model optimizer includes at least an AdamW optimizer, an Adam optimizer, and an SGD optimizer;

[0166] Step 35: Perform a round of traversal on all first data records of the first evaluation set; during this round of traversal, use the currently traversed first data record as the corresponding current evaluation record; input the first training parameter vector of the current evaluation record as the current myocardial strain parameter vector X into the hypertrophic cardiomyopathy risk prediction model for prediction, and use the predicted probability p output by this prediction as the corresponding second predicted probability; and form a corresponding second prediction-label pair from the current second predicted probability and the first label probability of the current evaluation record; and at the end of this round of traversal, bring all the obtained second prediction-label pairs into the preset first model evaluation function to calculate and obtain the corresponding first evaluation value;

[0167] Wherein, the first model evaluation function is implemented based on the MAE function, the MSE function or the RMSE function;

[0168] Step 36, identifying whether the first evaluation value meets the preset first evaluation value range; if not, returning to step 32; if so, confirming that the model pre-training is completed.

[0169] Here, the first evaluation value range is a preset numerical range.

[0170] Step 4: After the pre-training is completed, fine-tune the hypertrophic cardiomyopathy risk prediction model based on the second data set;

[0171] Specifically, step 41 includes randomly dividing the second data set into sub-data sets based on a preset second segmentation ratio to obtain corresponding second training sets and second evaluation sets; and using the fusion matrix parameter W of the multi-head self-attention module of the hypertrophic cardiomyopathy risk prediction model R And the weight matrix parameter W of the confidence assessment module out,1 、W out,2 Form the corresponding current fine-tuning parameter set; and divide the second training set into multiple second batch record sets according to the preset maximum number of records in a single batch;

[0172] The second split ratio is a preset ratio parameter, such as 8:2; the second training set and the second evaluation set both consist of multiple second data records; the ratio of the total number of records in the second training set to the total number of records in the second evaluation set satisfies the second split ratio; the total number of second data records in the second batch of records is less than or equal to the maximum number of records in a single batch;

[0173] Step 42, taking the first second batch of record sets as the corresponding current record sets;

[0174] Step 43: Perform a round of traversal on all second data records in the current record set; during this round of traversal, use the currently traversed second data record as the corresponding current training record; input the second training parameter vector of the current training record as the current myocardial strain parameter vector X into the hypertrophic cardiomyopathy risk prediction model for prediction, and use the predicted probability p output by this prediction as the corresponding third predicted probability; and form a corresponding third prediction-label pair using the current third predicted probability and the second label probability of the current training record; and at the end of this round of traversal, bring all third prediction-label pairs obtained in this round of traversal into the preset second model loss function to calculate the corresponding second loss value;

[0175] Among them, the second model loss function is implemented based on the binary cross entropy loss function;

[0176] Step 44: Identify whether the second loss value satisfies a preset second loss value range. If so, identify whether the current record set is the last second batch of record sets in the second training set. If so, proceed to step 45. If not, use the next second batch of record sets as the new current record set and return to step 43. If the second loss value does not satisfy the second loss value range, perform a round of fine-tuning on the current model parameter set based on a preset second model optimizer in a direction to minimize the second model loss function, and return to step 43 after this round of fine-tuning.

[0177] The second loss value range is a preset numerical range; the second model optimizer includes at least an AdamW optimizer, an Adam optimizer, and an SGD optimizer;

[0178] Step 45: Perform a round of traversal on all second data records of the second evaluation set; during this round of traversal, use the currently traversed second data record as the corresponding current evaluation record; input the second training parameter vector of the current evaluation record as the current myocardial strain parameter vector X into the hypertrophic cardiomyopathy risk prediction model for prediction, and use the predicted probability p output by this prediction as the corresponding fourth predicted probability; and form a corresponding fourth prediction-label pair by combining the current fourth predicted probability and the second label probability of the current evaluation record; and at the end of this round of traversal, calculate the accuracy, precision, recall, and F1 score based on all the obtained fourth prediction-label pairs to obtain the corresponding first accuracy, first precision, first recall, and first F1 score;

[0179] In step 46, it is determined whether the first accuracy, the first precision, the first recall, and the first F1 score each satisfy the corresponding first accuracy range, the first precision range, the first recall range, and the first F1 score range; if not, the process returns to step 42; if so, the process confirms that the model fine-tuning is complete.

[0180] Here, the first accuracy range, the first precision range, the first recall range, and the first F1 score range are four preset value ranges.

[0181] Step 5: After fine-tuning, the myocardial strain parameter set of the subject input by the user that meets the characteristics of the specified population is used as the corresponding myocardial strain parameter vector X and input into the hypertrophic cardiomyopathy risk prediction model for prediction, and the prediction probability p output by this prediction is fed back to the current user.

[0182] Here, the myocardial strain parameter set of the subject implemented in the present invention is a myocardial strain parameter set of the subject that meets the characteristics of a specified population, and is composed of six categories of left ventricular parameters and three categories of left atrial parameters.

[0183] Figure 3 This is a module structure diagram of a processing device for a hypertrophic cardiomyopathy risk prediction model provided in the second embodiment of the present invention. The device is a terminal device or server that implements the aforementioned method embodiment, or can be a device that enables the aforementioned terminal device or server to implement the aforementioned method embodiment. For example, the device can be a device or chip system of the aforementioned terminal device or server. Figure 3 As shown, the device includes: a model construction module 201, a data set construction module 202, a model pre-training module 203, a model fine-tuning module 204, and a model application module 205.

[0184] The model construction module 201 is used to construct a deep learning model for predicting the risk probability of hypertrophic cardiomyopathy, which is recorded as the corresponding hypertrophic cardiomyopathy risk prediction model; the hypertrophic cardiomyopathy risk prediction model is used to predict the risk probability of hypertrophic cardiomyopathy based on the myocardial strain parameter vector X input by the model and output the corresponding predicted probability p; the myocardial strain parameter vector X is composed of six categories of left ventricular parameters and three categories of left atrial parameters, the six categories of left ventricular parameters including GLS parameters, GCS parameters, GRS parameters, cTSD parameters, PSSR parameters, and PDSR parameters; the three categories of left atrial parameters including Ee parameters, sRe parameters, and sRa parameters; the predicted probability p is a probability value between 0 and 1.

[0185] The data set construction module 202 is used to construct a generalized population model data set as the corresponding first data set; and to construct a specified population model data set as the corresponding second data set based on the preset specified population characteristics; the first data set includes multiple first data records; the first data record includes a first training parameter vector and a first label probability; the second data set includes multiple second data records; the second data record includes a second training parameter vector and a second label probability; the first and second training parameter vectors are both composed of six categories of left ventricular parameters and three categories of left atrial parameters; the first and second label probabilities are both label probabilities with a value of 0 or 1; the specified population characteristics include regional configuration, age group configuration and gender configuration; when the regional configuration is not empty, it is composed of one or more regions; when the age group configuration is not empty, it is composed of one or more age groups among the preset youth age group, middle-aged age group and elderly age group; the gender configuration includes male, female and both male and female.

[0186] The model pre-training module 203 is used to pre-train the hypertrophic cardiomyopathy risk prediction model based on the first data set.

[0187] The model fine-tuning module 204 is configured to fine-tune the hypertrophic cardiomyopathy risk prediction model based on the second data set after the pre-training is completed.

[0188] After fine-tuning is completed, the model application module 205 is configured to input the user's myocardial strain parameter set that meets the characteristics of the specified population as the corresponding myocardial strain parameter vector X into the hypertrophic cardiomyopathy risk prediction model for prediction, and to feed back the predicted probability p output by this prediction to the current user. The myocardial strain parameter set of the subject is a myocardial strain parameter set that meets the characteristics of the specified population and consists of six categories of left ventricular parameters and three categories of left atrial parameters.

[0189] An embodiment of the present invention provides a processing device for a hypertrophic cardiomyopathy risk prediction model, which can execute the method steps in the above method embodiment. Its implementation principles and technical effects are similar and will not be repeated here.

[0190] It should be noted that the division of the modules of the above devices is merely a division of logical functions. In actual implementation, they can be fully or partially integrated into a physical entity or physically separated. Furthermore, these modules can all be implemented in the form of software called by a processing element; or all be implemented in the form of hardware; or some modules can be implemented in the form of software called by a processing element, and some modules can be implemented in the form of hardware. For example, the model building module can be a separate processing element, or it can be integrated into a chip of the above device. In addition, it can be stored in the form of program code in the memory of the above device, and called by a processing element of the above device to perform the functions of the above-mentioned module. The implementation of other modules is similar. In addition, these modules can all or partly be integrated together, or they can be implemented independently. The processing element described here can be an integrated circuit with signal processing capabilities. During implementation, each step of the above method or each of the above modules can be completed by hardware integrated logic circuits in the processor element or instructions in the form of software.

[0191] For example, the above modules may be one or more integrated circuits configured to implement the above methods, such as one or more application-specific integrated circuits (ASICs), one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs). For another example, when a module is implemented by scheduling program code through a processing element, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call program code. For another example, these modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0192] In the above embodiments, all or part of the embodiments may be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments may be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the above method embodiments are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The above-mentioned computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the above-mentioned computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, Bluetooth, microwave, etc.) means. The above-mentioned computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media. The above-mentioned available medium can be a magnetic medium (such as a floppy disk, hard disk, tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state disk (SSD)), etc.

[0193] Figure 4 This is a schematic diagram of the structure of an electronic device provided in the third embodiment of the present invention. The electronic device can be a terminal device or server that implements the method of the aforementioned embodiment, or it can be a terminal device or server that implements the method of the aforementioned embodiment connected to the aforementioned terminal device or server. Figure 4 As shown, the electronic device may include: a processor 301 (such as a CPU), a memory 302, and a transceiver 303; the transceiver 303 is coupled to the processor 301, and the processor 301 controls the transceiver 303's transceiver actions. Various instructions may be stored in the memory 302 for completing various processing functions and implementing the processing steps described in the aforementioned embodiment method. Preferably, the electronic device involved in the embodiment of the present invention further includes: a power supply 304, a system bus 305, and a communication port 306. The system bus 305 is used to realize communication connections between components. The above-mentioned communication port 306 is used for connecting and communicating between the electronic device and other peripherals.

[0194] exist Figure 4The system bus 305 mentioned in the figure can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The system bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus. The communication interface is used to realize communication between the database access device and other devices (such as clients, read-write libraries, and read-only libraries). The memory may include random access memory (RAM) and may also include non-volatile memory (Non-Volatile Memory), such as at least one disk storage.

[0195] The above-mentioned processors can be general-purpose processors, including central processing units (CPUs), network processors (NPs), graphics processing units (GPUs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0196] It should be noted that an embodiment of the present invention further provides a computer-readable storage medium, which stores instructions. When the computer-readable storage medium is run on a computer, it enables the computer to execute the methods and processing procedures provided in the above embodiments.

[0197] The embodiment of the present invention provides a processing method, device, electronic device and computer-readable storage medium for a hypertrophic cardiomyopathy risk prediction model. As can be seen from the above content, the embodiment of the present invention constructs a hypertrophic cardiomyopathy risk prediction model that can predict risks based on six types of left ventricular parameters (GLS parameters, GCS parameters, GRS parameters, cTSD parameters, PSSR parameters, PDSR parameters) and three types of left atrial parameters (Ee parameters, sRe parameters, sRa parameters); and constructs a generalized population model data set (first data set) and a specified population model data set (second data set) through big data collection; and first pre-trains the hypertrophic cardiomyopathy risk prediction model based on the first data set and then fine-tunes the hypertrophic cardiomyopathy risk prediction model based on the second data set; after the model fine-tuning is completed, the hypertrophic cardiomyopathy risk prediction model is used to make predictions based on the myocardial strain parameters of the subject. The embodiment of the present invention not only improves the prediction accuracy, but also improves the prediction stability.

[0198] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein may be implemented using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0199] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for processing a hypertrophic cardiomyopathy risk prediction model, characterized in that: The method comprises: A deep learning model for predicting the risk probability of hypertrophic cardiomyopathy is constructed, denoted as a corresponding hypertrophic cardiomyopathy risk prediction model; the hypertrophic cardiomyopathy risk prediction model is used to predict the risk probability of hypertrophic cardiomyopathy based on a myocardial strain parameter vector X input by the model and output a corresponding predicted probability p; the myocardial strain parameter vector X is composed of six categories of left ventricular parameters and three categories of left atrial parameters, the six categories of left ventricular parameters including GLS parameters, GCS parameters, GRS parameters, cTSD parameters, PSSR parameters, and PDSR parameters; the three categories of left atrial parameters including Ee parameters, sRe parameters, and sRa parameters; the predicted probability p is a probability value between 0 and 1; Construct a generalized population model dataset as the corresponding first dataset; and construct a specified population model dataset as the corresponding second dataset based on preset specified population characteristics; the first dataset includes multiple first data records; the first data record includes a first training parameter vector and a first label probability; the second dataset includes multiple second data records; the second data record includes a second training parameter vector and a second label probability; the first and second training parameter vectors are both composed of the six categories of left ventricular parameters and the three categories of left atrial parameters; the first and second label probabilities are both label probabilities with a value of 0 or 1; the specified population characteristics include regional configuration, age group configuration and gender configuration; when the regional configuration is not empty, it is composed of one or more regions; when the age group configuration is not empty, it is composed of one or more age groups among the preset adolescent age group, middle-aged age group and elderly age group; the gender configuration includes male, female and both male and female; Pre-training the hypertrophic cardiomyopathy risk prediction model based on the first data set; After the pre-training is completed, fine-tuning the hypertrophic cardiomyopathy risk prediction model based on the second data set; After fine-tuning is completed, the myocardial strain parameter set of the subject input by the user that meets the characteristics of the specified population is used as the corresponding myocardial strain parameter vector X and is input into the hypertrophic cardiomyopathy risk prediction model for prediction, and the predicted probability p output by this prediction is fed back to the current user; the myocardial strain parameter set of the subject is a myocardial strain parameter set of the subject that meets the characteristics of the specified population, and is composed of the six categories of left ventricular parameters and the three categories of left atrial parameters.

2. The method for processing the hypertrophic cardiomyopathy risk prediction model according to claim 1, characterized in that: The model input end of the hypertrophic cardiomyopathy risk prediction model is used to receive the myocardial strain parameter vector X, and the model output end is used to output the corresponding predicted probability p; The hypertrophic cardiomyopathy risk prediction model includes a normalization module, a ventricular feature extraction module, an atrial feature extraction module, a feature splicing module, a multi-head self-attention module, a residual connection module, a feature mapping module and a Sigmoid function module; The input end of the standardization module is connected to the input end of the model, and the first and second output ends are respectively connected to the input ends of the ventricle and atrial feature extraction modules; the output ends of the ventricle and atrial feature extraction modules are respectively connected to the first and second input ends of the feature splicing module; the output end of the feature splicing module is respectively connected to the input end of the multi-head self-attention module and the first input end of the residual connection module; the output end of the multi-head self-attention module is connected to the second input end of the residual connection module; the output end of the residual connection module is connected to the input end of the feature mapping module; the output end of the feature mapping module is connected to the input end of the Sigmoid function module; the output end of the Sigmoid function module is connected to the output end of the model; The standardization module is used to extract the corresponding six categories of left ventricular parameters and the three categories of left atrial parameters from the myocardial strain parameter vector X; and based on the preset standardization rules, the nine categories of myocardial strain parameters are respectively standardized to obtain the corresponding six categories of left ventricular standardized parameters and three categories of left atrial standardized parameters; and the six categories of left ventricular standardized parameters are used to form the corresponding left ventricular parameter vector X V Send to the ventricular feature extraction module, the corresponding left atrial parameter vector X is composed of the three types of left atrial standardized parameters A Send to the atrial feature extraction module; wherein the standardization rule includes at least the z-score rule; the six categories of left ventricular standardization parameters include GLS standardization parameters, GCS standardization parameters, GRS standardization parameters, cTSD standardization parameters, PSSR standardization parameters and PDSR parameter standardization parameters; the three categories of left atrial standardization parameters include Ee standardization parameters, sRe standardization parameters, and sRa standardization parameters; the left ventricular parameter vector X V It consists of 6 parameters Composition, 1≤index i≤6, 6 parameters One-to-one correspondence with the six types of left ventricular standardized parameters; the left atrial parameter vector X A By 3 parameters Composition, 1≤indexj≤3, 3 parameters One-to-one correspondence with the three types of left atrial standardized parameters; The ventricular feature extraction module is composed of two layers of fully connected layers connected in sequence, and a LeakyReLU activation function is connected after each fully connected layer; the ventricular feature extraction module is used to extract the left ventricular parameter vector X V Perform feature extraction to obtain the corresponding feature vector H V Sending to the feature splicing module; The eigenvector H V The reasoning process is: H V =LeakyReLU(W V,2 LeakyReLU(W V,1 X V +b V,1 )+b V,2 ), W V,1 、b V,1 are the weight matrix parameters and offset vector parameters of the first fully connected layer of the ventricular feature extraction module, W V,2 、b V,2 The weight matrix parameters and offset vector parameters of the second fully connected layer of the ventricular feature extraction module; the left ventricular parameter vector X V The shape is 6×1, and the weight matrix parameter W V,1 The shape is C V,1 ×6, the offset vector parameter b V,1 The shape is C V,1 ×1, LeakyReLU(W V,1 X V +b V,1 ) has an output vector shape of C V,1 ×1; the weight matrix parameter W V,2 The shape is C V,2 ×C V,1 , the offset vector parameter b V,2 The shape is C V,2 ×1, the eigenvector H V The shape is C V,2 ×1;C V,1 、C V,2 are two preset feature vector dimensions, and 6<C V,1 <C V,2 ; The atrial feature extraction module is composed of two layers of fully connected layers connected in sequence, and a LeakyReLU activation function is connected after each fully connected layer; the atrial feature extraction module is used to extract the left atrial parameter vector X A Perform feature extraction to obtain the corresponding feature vector H A Sending to the feature splicing module; The eigenvector H A The reasoning process is: H A =LeakyReLU(W A,2 LeakyReLU(W A,1 X A +b A,1 )+b A,2 ), W A,1 、b A,1 are the weight matrix parameters and offset vector parameters of the first fully connected layer of the atrial feature extraction module, W A,2 、b A,2 The weight matrix parameters and offset vector parameters of the second fully connected layer of the atrial feature extraction module; the left atrial parameter vector X A The shape is 3×1, and the weight matrix parameter W A,1 The shape is C A,1 ×3, the offset vector parameter b A,1 The shape is C A,1 ×1, LeakyReLU(W A,1 X A +b A,1 ) has an output vector shape of C A,1 ×1; the weight matrix parameter W A,2 The shape is C A,2 ×C A,1 , the offset vector parameter b A,2 The shape is C A,2 ×1, the eigenvector H A The shape is C A,2 ×1;C A,1 、C A,2 are two preset feature vector dimensions, and 3<C A,1 <C A,2 , C A,1 <C V,1 , C A,2 <C V,2 ; The feature splicing module is used to V and the eigenvector H A Perform vector splicing to obtain the corresponding splicing vector H C Send to the multi-head self-attention module and the residual connection module; the splicing vector H C The shape is C C ×1, C C is a preset feature vector dimension, C C =C V,2 +C A,2 ; The multi-head self-attention module consists of N head It consists of a parallel single-head attention unit and a multi-head fusion unit; N head is the preset total number of single heads, the total number of single heads N head is a number greater than 1 and can be represented by the dimension C of the feature vector C A positive integer that is divisible by ; each of the single-head attention units corresponds to an independent set of query / key / value matrix parameters 1≤indexn≤N head , each set of query / key / value matrix parameters The shape is C head ×C C , C head is a preset feature vector dimension, C head =C C / N head ; The multi-head fusion unit corresponds to a fusion matrix parameter W R , the fusion matrix parameter W R The shape is C C ×C C ; The multi-head self-attention module is used to C Perform multi-head self-attention operation to obtain the corresponding feature vector H R Send to the residual connection module, specifically: each of the single-head attention units according to the corresponding query / key / value matrix parameters For the splicing vector H C Perform a self-attention operation to obtain the corresponding single-head feature vector H n Send to the multi-head fusion unit; the multi-head fusion unit sends head The single-headed eigenvector H n Perform feature splicing to obtain the corresponding splicing vector H all-head , and based on the fusion matrix parameter W R For the splicing vector H all-head Perform feature mapping to obtain the corresponding feature vector H R Sending to the residual connection module; Each of the single-head feature vectors H n The reasoning process is: Q n , K n 、V n are the query vector, key vector, and value vector of the nth single-head attention unit, is the key vector K n The transposed vector of , softmax() is the Softmax function; the query, key, value vector Q n , K n 、V n The shape is C head ×1, the transposed vector The shape is 1×C head , the single-head eigenvector H n The shape is C head ×1; The splicing vector H all-head The shape is (C head ×N head )×1=C C ×1; The eigenvector H R The reasoning process is: H R =W R H all-head , The eigenvector H R The shape is C C ×1; The residual connection module is used to connect the splicing vector H in a residual connection manner. C and the eigenvector H R Fusion is performed to obtain the corresponding feature vector H R+C Sending to the feature mapping module; The eigenvector H R+C The reasoning process is: H R+c =H c +H R , The eigenvector H R+C The shape is C C ×1; The feature mapping module is composed of two layers of fully connected layers connected in sequence, and a Swish activation function is connected after the first fully connected layer; the feature mapping module is used to calculate the feature vector H R+C Calculate the confidence score and send the corresponding confidence score z to the Sigmoid function module; The reasoning process of the confidence score z is: z=W out,2 Swish(W out,1 H R+C +b out,1 )+b out,2 , W out,1 、b out,1 are the weight matrix parameters and offset vector parameters of the first fully connected layer of the feature mapping module, W out,2 、b out,2 The weight matrix parameters and offset vector parameters of the second fully connected layer of the feature mapping module are as follows: out,1 The shape is C out,1 ×C C , the offset vector parameter b V,1 The shape is C out,1 ×1, Swish(W out,1 H R+C +b out,1 ) has an output vector shape of C out,1 ×1; the weight matrix parameter W out,2 The shape is C out,2 ×C out,1 , the offset vector parameter b out,2 The shape is C out,2 ×1, C out,1 、C out,2 are two preset feature vector dimensions, C out,1 >C C , C out,2 =1; the confidence score z is a scalar; The Sigmoid function module is used to bring the confidence score z into the Sigmoid function for calculation and output the calculation result as the corresponding predicted probability p.

3. The method for processing the hypertrophic cardiomyopathy risk prediction model according to claim 1, characterized in that: The step of constructing a generalized population model dataset as the corresponding first dataset specifically includes: Step 31, using a preset multi-category data collection channel, collect big data of cardiac magnetic resonance images of multiple groups of people in multiple regions and age groups to obtain corresponding original image sets; Among them, the multiple types of data collection channels include at least public medical imaging data sets, public or authorized medical institution databases for data use, and public or authorized medical imaging sharing platforms for data use; the multiple regions include at least two or more regions; the multiple age groups include at least adolescent age groups, middle-aged age groups, and elderly age groups; the multiple types of people include at least healthy people who do not suffer from any heart disease, other patients who do not suffer from hypertrophic cardiomyopathy but suffer from other heart diseases, and target patient groups with hypertrophic cardiomyopathy; the genders of the multiple types of people should cover both males and females; the original image set includes multiple original image records; the original image record includes a first image, a first region, a first gender, a first age group, and a first population type; the first image is a cardiac magnetic resonance image; the first region is one of the multiple regions; the first gender includes males and females; the first age group includes adolescent age groups, middle-aged age groups, and elderly age groups; the first population type includes healthy people who do not suffer from any heart disease, other patients who do not suffer from hypertrophic cardiomyopathy but suffer from other heart diseases, and target patient groups with hypertrophic cardiomyopathy; Step 32: Use the first image of each original image record in the original image set as the corresponding current image; quantitatively evaluate the six categories of left ventricular parameters and the three categories of left atrial parameters corresponding to the current image based on preset myocardial strain parameter evaluation software, and form a corresponding first training parameter vector with the evaluated six categories of left ventricular parameters and the three categories of left atrial parameters; identify whether the first population type corresponding to the current image is a target patient population with hypertrophic cardiomyopathy, and if so, set a corresponding first label probability to 1; otherwise, set a corresponding first label probability to 0; and form a corresponding first data record with the first training parameter vector corresponding to the current image and the first label probability; and form a corresponding first data set with all the obtained first data records; The myocardial strain parameter assessment software includes at least CVI42 software, CAAS MR Strain software, and Segment software.

4. The method for processing the hypertrophic cardiomyopathy risk prediction model according to claim 3, characterized in that: The step of constructing a designated population model dataset as a corresponding second dataset based on preset designated population characteristics specifically includes: Extracting the corresponding region configuration, age group configuration, and gender configuration from the specified population characteristics; and identifying whether the region configuration is empty; if so, using the original image set as the corresponding first image set; if not, forming the corresponding first image set from the original image records of the first region in the original image set that satisfy the region configuration; and identifying whether the age group configuration is empty; if so, using the first image set as the corresponding second image set; if not, forming the corresponding second image set from the original image records in the first image set whose first age group meets the age group configuration; and identifying the gender configuration; if the gender configuration is male, the corresponding third image set is composed of the original image records in the second image set in which the first gender is male; if the gender configuration is female, the corresponding third image set is composed of the original image records in the second image set in which the first gender is female; if the gender configuration is unisex, the second image set is used as the corresponding third image set; The first data records corresponding to each of the original image records of the third image set in the first data set are extracted as a corresponding second data record; and all the obtained second data records form a corresponding second data set.

5. The method for processing the hypertrophic cardiomyopathy risk prediction model according to claim 1, characterized in that: Pre-training the hypertrophic cardiomyopathy risk prediction model based on the first data set specifically includes: Step 51: randomly segmenting the first data set into sub-datasets based on a preset first segmentation ratio to obtain corresponding first training sets and first evaluation sets; and using the full model parameter set of the hypertrophic cardiomyopathy risk prediction model as the corresponding current model parameter set; and segmenting the first training set into multiple first batch record sets according to a preset maximum number of records in a single batch; wherein both the first training set and the first evaluation set consist of a plurality of the first data records; the ratio of the total number of records in the first training set to the total number of records in the first evaluation set satisfies the first segmentation ratio; and the total number of the first data records in the first batch of records is less than or equal to the maximum number of records in a single batch; Step 52: taking the first of the first record sets as the corresponding current record set; Step 53: Perform a round of traversal on all the first data records in the current record set; and during this round of traversal, use the first data record currently traversed as the corresponding current training record; and input the first training parameter vector of the current training record as the current myocardial strain parameter vector X into the hypertrophic cardiomyopathy risk prediction model for prediction, and use the predicted probability p output by this prediction as the corresponding first predicted probability; and form a corresponding first prediction-label pair from the current first predicted probability and the first label probability of the current training record; and at the end of this round of traversal, bring all the first prediction-label pairs obtained in this round of traversal into a preset first model loss function to calculate and obtain a corresponding first loss value; Wherein, the first model loss function is implemented based on L1 loss function, L2 loss function or binary cross entropy loss function; Step 54: Identify whether the first loss value satisfies a preset first loss value range. If the first loss value satisfies the first loss value range, identify whether the current record set is the last first record set of the first training set. If so, proceed to step 55. If not, use the next first record set as the new current record set and return to step 53. If the first loss value does not satisfy the first loss value range, perform a round of modulation on the current model parameter set based on a preset first model optimizer in a direction that minimizes the first model loss function, and return to step 53 after this round of modulation. Wherein, the first model optimizer includes at least AdamW optimizer, Adam optimizer, and SGD optimizer; Step 55: Perform a round of traversal on all the first data records in the first evaluation set; and during this round of traversal, use the currently traversed first data record as the corresponding current evaluation record; and input the first training parameter vector of the current evaluation record as the current myocardial strain parameter vector X into the hypertrophic cardiomyopathy risk prediction model for prediction, and use the predicted probability p output by this prediction as the corresponding second predicted probability; and form a corresponding second prediction-label pair from the current second predicted probability and the first label probability of the current evaluation record; and at the end of this round of traversal, bring all the obtained second prediction-label pairs into the preset first model evaluation function to calculate and obtain the corresponding first evaluation value; Wherein, the first model evaluation function is implemented based on MAE function, MSE function or RMSE function; Step 56, identifying whether the first evaluation value meets the preset first evaluation value range; if not, returning to step 52; if so, confirming that the model pre-training is completed.

6. The method for processing the hypertrophic cardiomyopathy risk prediction model according to claim 1, characterized in that: The fine-tuning of the hypertrophic cardiomyopathy risk prediction model based on the second data set specifically includes: Step 61, based on a preset second segmentation ratio, the second data set is randomly segmented into sub-data sets to obtain corresponding second training sets and second evaluation sets; and the fusion matrix parameter W of the multi-head self-attention module of the hypertrophic cardiomyopathy risk prediction model is used. R And the weight matrix parameter W of the feature mapping module out,1 、W out,2 Forming a corresponding current fine-tuning parameter set; and dividing the second training set into multiple second batch record sets according to a preset maximum number of records in a single batch; wherein both the second training set and the second evaluation set are composed of a plurality of second data records; the ratio of the total number of records in the second training set to the total number of records in the second evaluation set satisfies the second split ratio; and the total number of the second data records in the second batch of records is less than or equal to the maximum number of records in a single batch; Step 62, taking the first of the second batch of record sets as the corresponding current record set; Step 63: Perform a round of traversal on all the second data records in the current record set; and during this round of traversal, use the currently traversed second data record as the corresponding current training record; and input the second training parameter vector of the current training record as the current myocardial strain parameter vector X into the hypertrophic cardiomyopathy risk prediction model for prediction, and use the predicted probability p output by this prediction as the corresponding third predicted probability; and form a corresponding third prediction-label pair by combining the current third predicted probability and the second label probability of the current training record; and at the end of this round of traversal, bring all the third prediction-label pairs obtained in this round of traversal into the preset second model loss function to calculate the corresponding second loss value; Wherein, the second model loss function is implemented based on the binary cross entropy loss function; Step 64: Identify whether the second loss value satisfies a preset second loss value range. If the second loss value satisfies the second loss value range, identify whether the current record set is the last second batch of record sets in the second training set. If so, proceed to step 65. If not, use the next second batch of record sets as the new current record set and return to step 63. If the second loss value does not satisfy the second loss value range, perform a round of fine-tuning on the current model parameter set based on a preset second model optimizer in a direction to minimize the second model loss function, and return to step 63 after this round of fine-tuning. Wherein, the second model optimizer includes at least AdamW optimizer, Adam optimizer, and SGD optimizer; Step 65: Perform a round of traversal on all the second data records of the second evaluation set; and during this round of traversal, use the currently traversed second data record as the corresponding current evaluation record; and input the second training parameter vector of the current evaluation record as the current myocardial strain parameter vector X into the hypertrophic cardiomyopathy risk prediction model for prediction, and use the predicted probability p output by this prediction as the corresponding fourth predicted probability; and form a corresponding fourth prediction-label pair by the current fourth predicted probability and the second label probability of the current evaluation record; and at the end of this round of traversal, calculate the accuracy, precision, recall and F1 score based on all the obtained fourth prediction-label pairs to obtain the corresponding first accuracy, first precision, first recall and first F1 score; Step 66: Identify whether the first accuracy, the first precision, the first recall, and the first F1 score each satisfy the corresponding first accuracy range, first precision range, first recall range, and first F1 score range; if not, return to step 62; if so, confirm that the model fine-tuning is completed.

7. A device for executing the method for processing the hypertrophic cardiomyopathy risk prediction model according to any one of claims 1 to 6, characterized in that: The device includes: a model construction module, a data set construction module, a model pre-training module, a model fine-tuning module, and a model application module; The model construction module is used to construct a deep learning model for predicting the risk probability of hypertrophic cardiomyopathy, which is recorded as the corresponding hypertrophic cardiomyopathy risk prediction model; the hypertrophic cardiomyopathy risk prediction model is used to predict the risk probability of hypertrophic cardiomyopathy based on the myocardial strain parameter vector X input by the model and output the corresponding predicted probability p; the myocardial strain parameter vector X is composed of six categories of left ventricular parameters and three categories of left atrial parameters, the six categories of left ventricular parameters including GLS parameters, GCS parameters, GRS parameters, cTSD parameters, PSSR parameters, and PDSR parameters; the three categories of left atrial parameters including Ee parameters, sRe parameters, and sRa parameters; the predicted probability p is a probability value between 0 and 1; The data set construction module is used to construct a generalized population model data set as the corresponding first data set; and to construct a specified population model data set as the corresponding second data set based on the preset specified population characteristics; the first data set includes multiple first data records; the first data record includes a first training parameter vector and a first label probability; the second data set includes multiple second data records; the second data record includes a second training parameter vector and a second label probability; the first and second training parameter vectors are both composed of the six categories of left ventricular parameters and the three categories of left atrial parameters; the first and second label probabilities are both label probabilities with a value of 0 or 1; the specified population characteristics include regional configuration, age group configuration and gender configuration; when the regional configuration is not empty, it is composed of one or more regions; when the age group configuration is not empty, it is composed of one or more age groups among the preset adolescent age group, middle-aged age group and elderly age group; the gender configuration includes male, female and both male and female; The model pre-training module is used to pre-train the hypertrophic cardiomyopathy risk prediction model based on the first data set; The model fine-tuning module is used to fine-tune the hypertrophic cardiomyopathy risk prediction model based on the second data set after the pre-training is completed; The model application module is configured to, after fine-tuning is completed, input the myocardial strain parameter set of the subject that meets the characteristics of the specified population, as the corresponding myocardial strain parameter vector X, into the hypertrophic cardiomyopathy risk prediction model for prediction, and feed back the predicted probability p output by this prediction to the current user; the myocardial strain parameter set of the subject is a myocardial strain parameter set of the subject that meets the characteristics of the specified population, and is composed of the six categories of left ventricular parameters and the three categories of left atrial parameters.

8. An electronic device, characterized in that: include: memory, processors, and transceivers; The processor is configured to be coupled to the memory, read and execute instructions in the memory, so as to implement the method according to any one of claims 1 to 6; The transceiver is coupled to the processor, and the processor controls the transceiver to send and receive messages.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a computer, the computer is caused to execute the method according to any one of claims 1 to 6.

Images