Patient classification method and device based on magnetic resonance parameters, equipment and medium

Abstract

The invention discloses a patient classification method and device based on magnetic resonance parameters, equipment and a medium. The method comprises the steps of obtaining target magnetic resonance parameters of a target patient; the target magnetic resonance parameter is screened from a plurality of preset magnetic resonance parameters based on the heart resynchronization tag of the sample patient and the real result of the occurrence of the preset endpoint event; the target magnetic resonance parameters comprise a myocardial strain parameter, a gadolinium contrast agent delay strengthening result and a left ventricular torsion parameter; inputting the target magnetic resonance parameters of the target patient into a target classification tree, and classifying the target patient by adopting the target classification tree to obtain a target classification result corresponding to the target patient; the target classification tree is obtained by extending a preset classification tree on the basis of cut-off values of the target magnetic resonance parameters and meeting the influence of the cut-off values on heart resynchronization; and outputting a target classification result. According to the method, the patients can be classified more accurately.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of intelligent medical treatment, and particularly relates to a patient classification method and device based on magnetic resonance parameters, an equipment, a computer readable storage medium and a computer program product. BACKGROUND

[0002] Cardiac resynchronization therapy (CRT) is an important instrument treatment method for treating dilated cardiomyopathy (DCM) with intraventricular conduction block and heart failure.

[0003] At present, whether a patient is recommended to perform CRT is mainly determined by referring to the guidelines issued by the American Heart Association and the European Heart Association, and by the QRS complex time limit and morphological characteristics in the 12-lead electrocardiogram. However, the QRS complex time limit and morphological characteristics are too single, and cannot comprehensively evaluate whether the patient is suitable for CRT. Meanwhile, in clinical practice, the patients recommended by a part of the guidelines to perform CRT also have the condition of poor CRT efficacy.

[0004] In summary, the current CRT recommendation has the problem of inaccuracy. SUMMARY

[0005] The application provides a patient classification method and device based on magnetic resonance parameters, an equipment, a computer readable storage medium and a computer program product, which can classify patients more accurately.

[0006] In one aspect, the application provides a patient classification method based on magnetic resonance parameters, which comprises: obtaining a target magnetic resonance parameter of a target patient; the target magnetic resonance parameter is selected from a plurality of preset magnetic resonance parameters of sample patients based on a cardiac resynchronization label of the sample patients and a true result of occurrence of a preset endpoint event; the target magnetic resonance parameter comprises at least one of the following: a myocardial strain parameter, a gadolinium contrast agent delayed enhancement result, and a left ventricular torsion parameter; inputting the target magnetic resonance parameter of the target patient into a target classification tree, and classifying the target patient by using the target classification tree to obtain a target classification result corresponding to the target patient; the target classification tree is obtained by extending a preset classification tree based on a cutoff value of each target magnetic resonance parameter and an influence of meeting each cutoff value on cardiac resynchronization; outputting the target classification result.

[0007] In some embodiments, before obtaining the target magnetic resonance parameter of the target patient, the method further comprises: determining a control patient according to the cardiac resynchronization label of the sample patients; the cardiac resynchronization label of the control patient is not implanted; determine a cutoff value of each of the preset magnetic resonance parameters according to the true results of the preset endpoint events of the control patients; screen the target magnetic resonance parameter from the plurality of preset magnetic resonance parameters of the sample patients according to the cutoff values, and the cardiac resynchronization labels and the true results of the preset endpoint events of the sample patients.

[0008] In some embodiments, the determining the cutoff value of each of the preset magnetic resonance parameters according to the true results of the preset endpoint events of the control patients comprises: plot a receiver operating characteristic curve between each of the preset magnetic resonance parameters and the true results of the preset endpoint events of the control patients according to the true results of the preset endpoint events of the control patients; determine the cutoff value of each of the preset magnetic resonance parameters according to each of the receiver operating characteristic curves.

[0009] In some embodiments, the screening the target magnetic resonance parameter from the plurality of preset magnetic resonance parameters of the sample patients according to the cutoff values, and the cardiac resynchronization labels and the true results of the preset endpoint events of the sample patients comprises: determine a first type of patient and a second type of patient corresponding to each of the preset magnetic resonance parameters from the sample patients according to the cardiac resynchronization labels of the sample patients, and whether each of the preset magnetic resonance parameters meets the corresponding cutoff value; the first type of patient meets the cutoff value of the preset magnetic resonance parameter and the cardiac resynchronization label is implanted; the second type of patient meets the cutoff value of the preset magnetic resonance parameter and the cardiac resynchronization label is not implanted; determine whether there is a significant difference in the true results of the preset endpoint events of the first type of patient and the second type of patient corresponding to each of the preset magnetic resonance parameters; determine the preset magnetic resonance parameter corresponding to the first type of patient and the second type of patient whose true results of the preset endpoint events have a significant difference as the target magnetic resonance parameter.

[0010] In some embodiments, the determining whether there is a significant difference in the true results of the preset endpoint events of the first type of patient and the second type of patient corresponding to each of the preset magnetic resonance parameters comprises: determine a significance value of a Kaplan-Meier survival curve of the first type of patient and the second type of patient corresponding to each of the preset magnetic resonance parameters according to the true results of the preset endpoint events of the first type of patient and the second type of patient corresponding to each of the preset magnetic resonance parameters; determine the first type of patient and the second type of patient whose significance value of the Kaplan-Meier survival curve is less than a preset threshold as the true results of the preset endpoint events having a significant difference In some embodiments, before inputting the target magnetic resonance parameter of the target patient into the target classification tree, the method further comprises: obtaining a preset classification tree; the preset classification tree is used to determine an initial classification result corresponding to a target patient according to a clinical parameter of the target patient; the clinical parameter comprises at least one of the following: left ventricular ejection fraction (LVEF), QRS complex time, QRS complex morphological feature, cardiac function classification; the initial classification result comprises a first initial classification and a second initial classification; determining the initial classification result as an initial intermediate node of a target classification tree; branches of the initial intermediate node comprise the first initial classification and the second initial classification; determining a cutoff value of a gadolinium contrast agent delayed enhancement result as a first intermediate node connected by a branch of the first initial classification, and determining whether the cutoff value of the gadolinium contrast agent delayed enhancement result is met as a branch of the first intermediate node; determining a cutoff value of a ventricular strain parameter and a cutoff value of a left ventricular twist parameter as a second intermediate node connected by a branch of the second initial classification, and determining whether a first preset condition is met as a branch of the second intermediate node; the first preset condition is that the cutoff value of the myocardial strain parameter or the cutoff value of the left ventricular twist parameter is met; connecting the branches of the first intermediate node and the branches of the second intermediate node to the first classification result or the second classification result according to an influence of cardiac resynchronization therapy on meeting each cutoff value pair, to obtain the target classification tree.

[0011] In some embodiments, connecting the branches of the first intermediate node and the branches of the second intermediate node to the first classification result or the second classification result according to an influence of cardiac resynchronization therapy on meeting each cutoff value pair comprises: connecting a branch having a positive influence on cardiac resynchronization therapy to one of the first classification result and the second classification result; connecting a branch having no positive influence on cardiac resynchronization therapy to the other of the first classification result or the second classification result.

[0012] In some embodiments, the cardiac function classification comprises first to fourth classifications in ascending order, and the grade is positively correlated with the degree of patient activity limitation; the QRS complex morphological feature is left bundle branch block (LBBB) or non-LBBB; the first initial classification meets LVEF less than or equal to a first threshold value, QRS complex time greater than or equal to a second threshold value, LBBB, and cardiac function classification being the second to fourth classifications; the second initial classification meets LVEF less than or equal to the first threshold value, QRS complex time greater than or equal to the second threshold value, non-LBBB, and cardiac function classification being the second to fourth classifications, or meets LVEF less than or equal to the first threshold value, QRS complex time less than the second threshold value and greater than a third threshold value, LBBB, and cardiac function classification being the second to fourth classifications; classifying a target patient using the target classification tree to obtain a target classification result corresponding to the target patient, comprising: if the initial classification result of the target patient is the first initial classification, determining whether the gadolinium contrast agent delayed enhancement result meets a cutoff value; the gadolinium contrast agent delayed enhancement result meeting the cutoff value includes: the gadolinium contrast agent delayed enhancement result being negative, the gadolinium contrast agent delayed enhancement result being positive and the enhancement percentage being in [0, 8%]; if it is determined that the gadolinium contrast agent delayed enhancement result meets the cutoff value, determining that the target classification result is one of the first classification result and the second classification result; if it is determined that the gadolinium contrast agent delayed enhancement result does not meet the cutoff value, determining that the target classification result is the other of the first classification result and the second classification result; if the initial classification result of the target patient is the second initial classification, determining whether a first preset condition is met; if it is determined that the first preset condition is met, determining that the target classification result is one of the first classification result and the second classification result; if it is determined that the first preset condition is not met, determining that the target classification result is the other of the first classification result and the second classification result.

[0013] In some embodiments, outputting the target classification result includes at least one of the following: visually displaying the target classification result; generating a cardiac resynchronization indication recommendation report according to the target classification result, and outputting the cardiac resynchronization indication recommendation report; generating a cardiac resynchronization preoperative evaluation report using a preset artificial intelligence prediction model, and outputting the cardiac resynchronization preoperative evaluation report.

[0014] In another aspect, the embodiments of the present application provide a patient classification device based on magnetic resonance parameters, the device comprising: an acquisition module configured to acquire target magnetic resonance parameters of a target patient; the target magnetic resonance parameters are selected from a plurality of preset magnetic resonance parameters of sample patients based on cardiac resynchronization labels of the sample patients and true results of occurrence of a preset endpoint event; the target magnetic resonance parameters include at least one of the following: a myocardial strain parameter, a gadolinium contrast agent delayed enhancement percentage, and a left ventricular torsion parameter; a classification module configured to input the target magnetic resonance parameters of the target patient into a target classification tree, and classify the target patient using the target classification tree to obtain a target classification result corresponding to the target patient; the target classification tree is obtained by extending a preset classification tree based on cutoff values of each target magnetic resonance parameter and an influence of meeting each cutoff value on cardiac resynchronization; an output module configured to output the target classification result.

[0015] In still another aspect, the embodiment of the present application provides a patient classification device based on magnetic resonance parameters, the patient classification device based on magnetic resonance parameters comprising a processor and a memory storing computer program instructions; The processor implements the patient classification method based on magnetic resonance parameters of any one of the above when executing the computer program instructions.

[0016] In still another aspect, the embodiment of the present application provides a computer readable storage medium, the computer readable storage medium storing computer program instructions, the computer program instructions being executed by a processor to implement the patient classification method based on magnetic resonance parameters of any one of the above.

[0017] In still another aspect, the embodiment of the present application provides a computer program product, the instructions in the computer program product being executed by a processor of an electronic device to cause the electronic device to execute the patient classification method based on magnetic resonance parameters of any one of the above.

[0018] The patient classification method based on magnetic resonance parameters, the device, the equipment and the computer readable storage medium of the embodiment of the present application, by acquiring the target magnetic resonance parameter of the target patient; Since the target magnetic resonance parameter is based on the cardiac resynchronization label of the sample patient and the true result of the occurrence of the preset endpoint event, it is screened from a plurality of preset magnetic resonance parameters of the sample patient; The target magnetic resonance parameter comprises at least one of the following: myocardial strain parameter, gadolinium contrast agent delayed enhancement result, left ventricular torsion parameter; Therefore, the target magnetic resonance parameter can comprehensively evaluate whether the physiological state of the patient's heart is suitable for cardiac resynchronization; By inputting the target magnetic resonance parameter of the target patient into the target classification tree, and using the target classification tree to classify the target patient, the target classification result corresponding to the target patient is obtained; Since the target classification tree is obtained by extending the preset classification tree based on the cutoff value of each target magnetic resonance parameter and the influence of meeting each cutoff value on cardiac resynchronization; Therefore, the target classification tree can accurately classify the patient according to the effect of the target patient on cardiac resynchronization under different target magnetic resonance parameters; And output the target classification result, to realize accurate classification of the target patient. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments of the present application. For those skilled in the art, without paying creative labor, other drawings can also be obtained according to these drawings.

[0020] Figure 1 is a flowchart of the patient classification method based on magnetic resonance parameters provided by an embodiment of the present application; Figure 2is a flowchart of a method for patient classification based on magnetic resonance parameters provided by another embodiment of the present application; Figure 3 and Figure 4 is a Kaplan-Meier survival curve of the COR class I patients in yet another embodiment of the present application; Figure 5 is a Kaplan-Meier survival curve of the COR class IIa patients and the COR class IIb patients in yet another embodiment of the present application; Figure 6 is a Kaplan-Meier survival curve of the COR class IIa patients in yet another embodiment of the present application; Figure 7 is a schematic diagram of the probability of occurrence of a predetermined endpoint event for patients of different types in yet another embodiment of the present application; Figure 8 is a logic diagram of the classification of a target patient by a target classification tree in yet another embodiment of the present application; Figure 9 is a structural diagram of a device for patient classification based on magnetic resonance parameters provided by yet another embodiment of the present application; Figure 10 is a hardware structural diagram of a device for patient classification based on magnetic resonance parameters provided by yet another embodiment of the present application. DETAILED DESCRIPTION

[0021] The features and exemplary embodiments of the various aspects of the present application will be described in detail below with reference to the drawings. The following detailed description is merely intended to explain the present application, and is not intended to limit the present application. The present application can be implemented without some of the specific details, which are well known to those skilled in the art. The following description of the embodiments is merely intended to provide a better understanding of the present application by showing examples of the present application.

[0022] It is to be noted that the terms such as first and second, and the like, merely identify one entity or action from another, without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the presence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0023] The data involved in the present application are obtained through legal channels, have obtained the explicit informed consent of the relevant users / patients, and have been anonymized.

[0024] First, the professional terms involved in the present application are introduced.

[0025] Cardiac Resynchronization, in English, is an intervention strategy that corrects the asynchrony of ventricular contraction through electrophysiological regulation technology.

[0026] QRS complex, in English, is a waveform in an electrocardiogram that represents ventricular depolarization, i.e., the electrical activity during contraction, composed of Q wave, R wave, and S wave.

[0027] QRS complex time is the duration of the QRS complex on an electrocardiogram, usually measured in milliseconds, used to assess ventricular conduction abnormalities.

[0028] Left Bundle Branch Block (LBBB) is an electrophysiological abnormality of the heart, characterized by the inability of the left bundle branch to effectively conduct electrical signals, resulting in asynchronous electrical activity between the left and right ventricles.

[0029] Left Ventricular Ejection Fraction (LVEF) is the percentage of blood volume ejected from the left ventricle during each contraction relative to end-diastolic volume, used to measure cardiac pumping function.

[0030] Righ Ventricular Ejection Fraction (RVEF) is the percentage of blood volume ejected from the right ventricle during each contraction relative to end-diastolic volume, used to measure cardiac pumping function.

[0031] Ventricular strain parameters, also known as myocardial strain parameters, are indicators of myocardial deformation measured through imaging techniques, reflecting the function of the ventricle during contraction.

[0032] Global Longitudinal Strain (GLS) is a myocardial strain parameter that represents the overall deformation of the myocardium in the longitudinal direction, i.e., along the long axis of the heart.

[0033] Global Radial Strain (GRS) is a myocardial strain parameter that represents the overall deformation of the myocardium in the radial direction, i.e., perpendicular to the long axis of the heart.

[0034] Global Circumferential Strain (GCS) is a myocardial strain parameter that represents the overall deformation of the myocardium in the circumferential direction, i.e., around the ventricular cavity.

[0035] Late Gadolinium Enhancement (LGE) is a technique used in cardiac magnetic resonance imaging that involves delayed scanning after the administration of gadolinium contrast agent, used to detect areas of myocardial fibrosis or scarring.

[0036] Late Gadolinium Enhancement Percentage is the percentage of LGE area relative to the total mass of left ventricular myocardium, used to quantify the extent of myocardial injury.

[0037] Cardiac Torsion is a ventricular mechanics parameter that represents the difference in the relative rotation angle between the apex and the base of the heart during systole, reflecting the mechanical synergy of myocardial fiber spiral structure.

[0038] Left Ventricular Torsion Parameter is a ventricular mechanics parameter used to assess the mechanical indicators of left ventricular myocardial fiber spiral structure, which is detailed in this application as the torsion parameter.

[0039] N-terminal B-type natriuretic peptide precursor (NT-proBNP) is a key biomarker for evaluating heart function, secreted by ventricular myocardial cells when the heart pressure load increases or myocardial damage occurs. NT-proBNP levels can reflect the status of heart function, commonly expressed as lgNT-proBNP.

[0040] NYHA cardiac function classification, the full name in English, is a standard established by the New York Heart Association, which classifies the heart function of patients with heart failure into I to IV levels according to the difficulty of inducing symptoms in daily activities, from no symptoms to symptoms at rest.

[0041] Receiver Operating Characteristic curve, abbreviated as ROC curve, is the full name in English.

[0042] Kaplan-Meier survival curve, abbreviated as KM curve, is the full name in English.

[0043] Univariate Cox proportional hazards regression model, the full name in English, is a regression model used to analyze the risk of a single predictor variable on the time of event occurrence.

[0044] Multivariate Cox proportional hazards regression model, the full name in English, is a regression model used to analyze the risk of multiple predictor variables on the time of event occurrence while controlling for confounding factors.

[0045] Cardiac Magnetic Resonance (CMR) imaging is a non-invasive medical examination that uses magnetic fields, radio waves, and computer technology to generate detailed images of the heart and its blood vessels.

[0046] Next, the prior art involved in the present application will be described in detail.

[0047] Cardiac resynchronization therapy is an important device treatment for patients with dilated cardiomyopathy (DCM) and heart failure with intraventricular conduction block, such as left bundle branch block. Effective CRT treatment can improve ventricular synchrony, improve cardiac pumping function, and significantly improve patient symptoms, quality of life, and prognosis.

[0048] Currently, the recommendation of whether a patient is suitable for CRT treatment in clinical practice mainly depends on the clinical guidelines issued by authoritative organizations such as the American College of Cardiology (ACC), the American Heart Association (AHA) or the European Society of Cardiology (ESC). The core recommendation criteria of these guidelines are highly dependent on the analysis of a conventional 12-lead electrocardiogram, especially the two indicators of QRS complex duration and QRS complex morphological characteristics, usually taking QRS duration ≥120 ms or ≥150 ms as the key threshold. Based on these relatively single characteristics, the guidelines classify patients into categories such as recommended CRT implantation, possible benefit or not recommended.

[0049] The inventors have found through in-depth research that the existing guideline-based CRT recommendation method has significant accuracy problems, because the QRS complex duration and morphological characteristics are too single to fully reflect the patient's cardiac physiological state. Moreover, in clinical practice, there are also cases where patients recommended for CRT implantation according to the current guideline criteria do not obtain the expected benefits.

[0050] In summary, the existing guideline-based CRT recommendation method in the prior art cannot accurately classify patients, and the technical assistance information provided to doctors for reference is limited.

[0051] To solve the above technical problems in the prior art, the embodiments of the present application provide a patient classification method, device, equipment, computer readable storage medium and computer program product based on magnetic resonance parameters. First, the patient classification method based on magnetic resonance parameters provided by the embodiments of the present application will be introduced.

[0052] Figure 1 is a flowchart of the patient classification method based on magnetic resonance parameters provided by an embodiment of the present application. As shown in Figure 1 The patient classification method based on magnetic resonance parameters provided by an embodiment of the present application is executed by a patient classification device based on magnetic resonance parameters, and includes steps 101 to 103.

[0053] Step 101, obtaining a target magnetic resonance parameter of a target patient; the target magnetic resonance parameter is selected from a plurality of preset magnetic resonance parameters of a sample patient based on a cardiac resynchronization label of the sample patient and a true result of occurrence of a preset endpoint event; the target magnetic resonance parameter includes at least one of the following: a myocardial strain parameter, a gadolinium contrast agent delayed enhancement result, and a left ventricular torsion parameter.

[0054] In this embodiment, the cardiac resynchronization label is used to label whether the patient has undergone cardiac resynchronization, for example, the cardiac resynchronization label can be implanted or not implanted. Implantation means that the patient has implanted a cardiac pacing device for cardiac resynchronization, and non-implantation means that the patient has not implanted a cardiac pacing device for cardiac resynchronization.

[0055] The target magnetic resonance parameter can be a target magnetic resonance parameter received by a user, can be obtained by analyzing and calculating a magnetic resonance image of a target patient, and can also be obtained from a preset database according to an identification of the target patient.

[0056] The preset magnetic resonance parameters include a left ventricular ejection fraction LVEF, a right ventricular ejection fraction RVEF, a myocardial strain parameter, a gadolinium contrast agent delayed enhancement LGE result, and a left ventricular torsion parameter Torsion. The LGE result is LGE negative or LGE positive, LGE negative indicating that LGE does not exist, and LGE positive indicating that LGE exists. When LGE is positive, there is a LGE percentage. The myocardial strain parameter includes a global longitudinal strain GLS, a global circumferential strain GCS, and a global radial strain GRS.

[0057] The preset magnetic resonance parameters can be obtained by analyzing and calculating a cardiac magnetic resonance image of a sample patient. The cardiac magnetic resonance image can be obtained by performing a magnetic resonance scan on the heart by a magnetic resonance scanner. The cardiac magnetic resonance image can include the following three types: a conventional scout image, a conventional cine sequence, and a delayed enhancement sequence.

[0058] The left ventricular short-axis cine is analyzed and calculated by using a heart function special analysis software, and parameters such as left ventricular ejection fraction, right ventricular ejection fraction, and left ventricular mass can be obtained.

[0059] The left ventricular short-axis section is analyzed for delayed enhancement by using a full-width-half-maximum method, and abnormal delayed enhancement is defined as a threshold of a region of interest being more than half of the maximum signal intensity in myocardial scar. The area or volume of the region of interest and the relative percentage to the left ventricular mass can be calculated.

[0060] The endocardial and epicardial boundaries of the short-axis, two-chamber, three-chamber, and four-chamber cine sequences identified at the end of diastole are tracked throughout the cardiac cycle, and myocardial strain parameters can be calculated. The long-axis cine sequence is used to calculate the global longitudinal strain, and the short-axis cine sequence is used to calculate the global radial strain, the global circumferential strain, and the ventricular mechanics parameter torsion.

[0061] The calculation of the ventricular torsion value uses a standardized formula: torsion = (peak apical rotation angle-peak basal rotation angle) / ventricular long axis length. The quantification of ventricular torsion is based on the characteristic difference between the horizontal rotation movements of the apex and the base: from the apex direction, the apex part shows counterclockwise rotation, and the base part shows clockwise rotation.

[0062] The main magnetic field strength of the magnetic resonance scanner can be 1.5T, and has vector electrocardiogram gating, multi-channel coil combination and magnetic resonance compatible high-pressure injectors. When scanning, the head-advanced supine position scanning, wireless vector electrocardiogram single-gating scanning, respiratory navigation electrocardiogram gating method scanning and the like can be adopted.

[0063] The number of sample patients is multiple, and for the sample patients implanted with a cardiac resynchronization label, the preset magnetic resonance parameters refer to the parameters of the sample patients before implanting a cardiac pacing device for cardiac resynchronization.

[0064] The preset endpoint event includes heart failure death, sudden cardiac death (SCD), heart transplantation and implantation of a left ventricular assist device.

[0065] The true result of the sample patient occurring the preset endpoint event refers to whether the sample patient occurs the preset endpoint event within a preset time length. That is, the true result of the sample patient occurring the preset endpoint event can be occurrence or non-occurrence.

[0066] The target magnetic resonance parameter can be a parameter in the preset magnetic resonance parameters that has a significant correlation with the occurrence of the preset endpoint event, can be a parameter that has a positive effect on cardiac resynchronization, can be a parameter that has a significant correlation with the occurrence of the preset endpoint event or has a positive effect on cardiac resynchronization, or can be a parameter that has a significant correlation with the occurrence of the preset endpoint event and has a positive effect on cardiac resynchronization.

[0067] In an embodiment of the present application, the parameter in the preset magnetic resonance parameters that has a positive effect on cardiac resynchronization is selected as the target magnetic resonance parameter.

[0068] In an embodiment of the present application, the univariate Cox proportional hazards regression model and the multivariate Cox proportional hazards regression model can be used to determine whether each preset magnetic resonance parameter has a significant correlation with the occurrence of the preset endpoint event. For ease of description, the univariate Cox proportional hazards regression model is referred to as a univariate regression model, and the multivariate Cox proportional hazards regression model is referred to as a multivariate regression model.

[0069] For example, if the significance value p of the univariate regression model is less than a preset significance threshold, the preset magnetic resonance parameter corresponding to the univariate regression model has a significant correlation with the occurrence of the preset endpoint event. The preset significance threshold can be 0.05. That is, the preset magnetic resonance parameter corresponding to the univariate regression model with p<0.05 has a significant correlation with the occurrence of the preset endpoint event.

[0070] In one embodiment of the present application, Kaplan-Meier survival curve can be used to determine the influence of the preset magnetic resonance parameters on cardiac resynchronization, which can be seen from the description in step 2033 and will not be repeated here.

[0071] In one embodiment of the present application, the specific method for determining whether each preset magnetic resonance parameter has a significant correlation with the occurrence of the preset endpoint event can be as follows.

[0072] First, the cutoff values of each prediction magnetic resonance parameter are determined, and then the correlation between the cutoff values of each preset magnetic resonance parameter and the occurrence of the preset endpoint event, and the correlation between the plurality of clinical parameters and the occurrence of the preset endpoint event are analyzed by using a univariate regression model, which is referred to as univariate regression analysis hereinafter, and the results are shown in Table 1 below. Wherein, how to determine the cutoff values of each preset magnetic resonance parameter can be seen from the description in step 202 and will not be repeated here.

[0073] The plurality of clinical parameters are: gender, age, QRS waveform, systolic blood pressure, NYHA cardiac function classification, NT-proBNP level, creatinine level. Wherein, the QRS waveform refers to the QRS wave group morphological characteristics, including left bundle branch block, right bundle branch block and non-specific conduction delay.

[0074] Table 1 Univariate regression analysis results

[0075] According to the univariate regression analysis results shown in Table 1, there is a significant correlation between the occurrence of the preset endpoint event and the QRS waveform, systolic blood pressure, NYHA cardiac function classification, NT-proBNP level, creatinine level, biventricular ejection fraction, presence of LGE, LGE tripartition variable, GLS≥-5.9%, GCS≥-6.3%, GRS≤8.0% and torsion≤0.34° / cm.

[0076] In one embodiment of the present application, in order to more accurately determine the parameters having a significant correlation with the occurrence of the preset endpoint event, the parameters with p<0.05 in the univariate regression analysis results, i.e. QRS waveform, systolic blood pressure, NYHA cardiac function classification, NT-proBNP level, creatinine level, RVEF, LGE tripartition variable, LVEF, GLS, GCS, GRS and torsion, are included in the multivariate Cox regression model, and the forward stepwise method is used for parameter screening, which is referred to as multivariate regression analysis hereinafter.

[0077] Before performing multivariate regression analysis, Pearson correlation analysis was used to calculate the correlation between parameters. The correlation coefficients between LVEF, GLS, GCS and GRS were greater than 0.6, indicating that there was a significant correlation between LVEF, GLS, GCS and GRS.

[0078] Therefore, multivariate regression analysis requires the establishment of multiple independent multiple factor regression models. That is, LVEF, GLS, GCS and GRS are used to establish multiple factor regression models with other p<0.05 parameters in the univariate regression analysis results, and the significance value of each multiple factor regression model is determined.

[0079] Multivariate regression analysis was performed, and the consistency index Harrell's C-statistic was used to evaluate the discriminant performance of each multiple factor regression model. The value range of Harrell's C-statistic is 0.5 to 1.0, and the higher the value, the stronger the model's discrimination ability. Harrell's C-statistic of 0.5 indicates that the model has no discrimination ability, and Harrell's C-statistic of 1.0 indicates that the model has complete discrimination ability.

[0080] A multiple factor Cox proportional hazards regression model was constructed between LVEF and QRS waveform, systolic blood pressure, NYHA cardiac function classification, NT-proBNP level, creatinine level, RVEF, LGE three-part variable and torsion, named model one, and the forward stepwise method was used for parameter screening.

[0081] A multiple factor Cox proportional hazards regression model was constructed between GLS and QRS waveform, systolic blood pressure, NYHA cardiac function classification, NT-proBNP level, creatinine level, RVEF, LGE three-part variable and torsion, named model two, and the forward stepwise method was used for parameter screening.

[0082] A multiple factor Cox proportional hazards regression model was constructed between GCS and QRS waveform, systolic blood pressure, NYHA cardiac function classification, NT-proBNP level, creatinine level, RVEF, LGE three-part variable and torsion, named model three, and the forward stepwise method was used for parameter screening.

[0083] A multiple factor Cox proportional hazards regression model was constructed between GRS and QRS waveform, systolic blood pressure, NYHA cardiac function classification, NT-proBNP level, creatinine level, RVEF, LGE three-part variable and torsion, named model four, and the forward stepwise method was used for parameter screening.

[0084] The multivariate regression analysis results of Model 1 and Model 2 are shown in Table 2 below.

[0085] Table 2 Multivariate regression analysis results of Model 1 and Model 2

[0086] The multivariate regression analysis results of Model 3 and Model 4 are shown in Table 3 below.

[0087] Table 3 Multivariate regression analysis results of Model 3 and Model 4

[0088] The results of the above multivariate regression analysis showed that creatinine, RVEF, and Torsion ≤ 0.34° / cm were not significantly correlated with the occurrence of the preset endpoint events. The parameters that were significantly correlated with the occurrence of the preset endpoint events were QRS waveform, systolic blood pressure, NYHA cardiac function class, lgNT-proBNP, LGE tertile variable, LVEF ≤ 24.6%, GLS ≥ -5.9%, GCS ≥ -6.3%, and GRS ≤ 8.0%.

[0089] In step 102, the target magnetic resonance parameters of the target patient are input into a target classification tree, and the target classification tree is used to classify the target patient to obtain a target classification result corresponding to the target patient; the target classification tree is obtained by extending a preset classification tree based on the cutoff values ​​of each target magnetic resonance parameter and the impact of meeting each cutoff value on cardiac resynchronization.

[0090] In one embodiment of the present application, a preset classification tree is pre-constructed for initial classification of target patients. The preset classification tree can be constructed based on the patient classification in the clinical guidelines published by the ACC, AHA, or ESC. The preset classification tree can also be constructed by recursively segmenting clinical parameters such as QRS complex duration, QRS complex morphological characteristics, systolic blood pressure, cardiac function classification, lgNT-proBNP level, creatinine level, etc. of the sample patients.

[0091] The target classification tree is obtained by extending the pre-set classification tree. For example, based on the cutoff values ​​of the target MRI parameters, a next-level node is constructed based on the initial classification results of the sample patient in the pre-set classification tree. Based on the impact of the cutoff values ​​of each target MRI parameter on cardiac resynchronization, branches are extended from the next-level node and connected to leaf nodes.

[0092] In an embodiment of the present application, before or simultaneously with inputting the target magnetic resonance parameter into the target classification tree, the target classification tree is also inputted with clinical parameters of the target patient, such as QRS complex time limit, QRS complex morphological characteristics, systolic pressure, cardiac function classification, lgNT-proBNP, creatinine, etc., so that the target classification tree determines an initial classification result of the target patient.

[0093] In step 103, the target classification result is outputted.

[0094] In the embodiment, the manner of outputting the target classification result can include visual display, voice broadcast, printing, sending to a target device, etc.

[0095] The patient classification method based on magnetic resonance parameters provided in the embodiment is performed by obtaining a target magnetic resonance parameter of a target patient. The target magnetic resonance parameter is selected from a plurality of preset magnetic resonance parameters of sample patients based on the cardiac resynchronization labels of the sample patients and the true results of occurrence of a preset endpoint event. The target magnetic resonance parameter includes at least one of a myocardial strain parameter, a gadolinium contrast agent delayed enhancement result, and a left ventricular torsion parameter. Therefore, the target magnetic resonance parameter can comprehensively evaluate whether the physiological state of the patient is suitable for cardiac resynchronization. The target magnetic resonance parameter of the target patient is inputted into a target classification tree, and the target classification tree is used to classify the target patient to obtain a target classification result corresponding to the target patient. The target classification tree is obtained by extending a preset classification tree based on the cutoff values of the target magnetic resonance parameters and the influence of meeting each cutoff value on cardiac resynchronization. Therefore, the target classification tree can accurately classify the patient according to the effect of the target patient on cardiac resynchronization under different target magnetic resonance parameters. The target classification result is then outputted to achieve accurate classification of the target patient.

[0096] Figure 2 FIG. 1 is a flowchart of a patient classification method based on magnetic resonance parameters according to another embodiment of the present application. As shown in FIG. 1, in order to accurately classify a target patient, before step 101, steps 201 to 203 are further included. Figure 2

[0097] In step 201, a control patient is determined according to a cardiac resynchronization label of a sample patient. The cardiac resynchronization label of the control patient is not implanted.

[0098] In an embodiment of the present application, in order to determine the cutoff values of the preset magnetic resonance parameters, the sample patient with the cardiac resynchronization label of not implanted is determined as a control patient. Since the control patient does not implant a cardiac pacing device for cardiac resynchronization, the true result of occurrence of a preset endpoint event of the control patient can be used to determine the correlation between the preset magnetic resonance parameters and the occurrence of the preset endpoint event, and further determine the cutoff values of the preset magnetic resonance parameters.​

[0099] In step 202, the cut-off value of each preset magnetic resonance parameter is determined according to the real result of the preset endpoint event of the control patient.

[0100] In an embodiment of the present application, the real result of the preset endpoint event of the control patient can be obtained from a preset database according to the identification of the control patient. The real result of the preset endpoint event of the control patient in the preset database can be written by relevant staff through calling the patient or family members, checking the medical record or asking the attending physician, etc.

[0101] In an embodiment of the present application, in order to accurately determine the cut-off value of the preset magnetic resonance parameter, step 202 includes steps 2021 to 2022.

[0102] In step 2021, a receiver operating characteristic curve between each preset magnetic resonance parameter and the real result of the preset endpoint event of the control patient is drawn according to the real result of the preset endpoint event of the control patient.

[0103] In an embodiment of the present application, for a certain preset magnetic resonance parameter, the step of drawing the ROC curve can be as follows.

[0104] In order, the values of the preset magnetic resonance parameter are taken as thresholds in the order from high to low.

[0105] The control patient with the preset magnetic resonance parameter greater than the threshold is judged to have occurred the preset endpoint event, and the control patient with the preset magnetic resonance parameter less than the threshold is judged to have not occurred the preset endpoint event. The control patient with the preset magnetic resonance parameter equal to the threshold is judged to have not occurred the preset endpoint event or to have occurred the endpoint event.

[0106] The true positive rate TPR and the false discovery rate FDR under each threshold are calculated. The TPR, also known as sensitivity, represents the proportion of control patients with the real result of the preset endpoint event being correctly judged. The FDR represents the proportion of control patients with the real result of the preset endpoint event being incorrectly judged.

[0107] The FPR is taken as the abscissa, the TPR is taken as the ordinate, a coordinate system is constructed, the coordinate point (0, 0), the coordinate point (1, 1), and the point corresponding to the FPR and the TPR calculated for each threshold in the coordinate system are drawn, and the balance line segment is connected to obtain the ROC curve between the preset magnetic resonance parameter and the real result of the preset endpoint event of the control patient. The coordinate point (0, 0) represents the case that all control patients are judged to have not occurred the preset endpoint event, and the coordinate point (1, 1) represents the case that all control patients are judged to have occurred the preset endpoint event.

[0108] At step 2022, the cut-off value of each preset magnetic resonance parameter is determined according to each receiver operating characteristic curve.

[0109] In an embodiment of the present application, the accuracy of the ROC curve is evaluated by the area under the curve (AUC), and the cut-off value is determined by the Youden index.

[0110] Specifically, the Youden index of each threshold value is calculated, and the threshold value with the maximum Youden index is determined as the cut-off value. The Youden index = sensitivity + specificity - 1. The specificity = 1-FPR.

[0111] The patient classification method based on magnetic resonance parameters provided in the embodiment can determine the true positive rate TDR and the false positive rate FDR under different threshold values, determine the Youden index of different threshold values, and then determine the cut-off value of each preset magnetic resonance parameter according to each receiver operating characteristic curve, so as to accurately determine the cut-off value of each preset magnetic resonance parameter and accurately classify the target patient.

[0112] In an embodiment of the present application, by drawing the receiver operating characteristic curve between each preset magnetic resonance parameter and the true result of the occurrence of the preset endpoint event of the control patient, it is known that the AUC of 1-LVEF is 0.68, the 95% confidence interval CI is 0.64-0.73, and p<0.001; the AUC of LGE is 0.63, the 95% confidence interval CI is 0.59-0.68, and p<0.001; the AUC of LGE percentage is 0.69, the 95% confidence interval CI is 0.64-0.73, and p<0.001; the AUC of GLS is 0.68, the 95% confidence interval CI is 0.63-0.73, and p<0.001; the AUC of GCS is 0.68, the 95% confidence interval CI is 0.63-0.72, and p<0.001; the AUC of 1-GRS is 0.69, the 95% confidence interval CI is 0.65-0.74, and p<0.001; the AUC of 1-torsion is 0.53, the 95% confidence interval CI is 0.48-0.58, and p=0.27. Therefore, the cut-off values of LVEF, GLS, GCS, GRS and Torsion are determined as 24.6%, -5.9%, -6.3%, 8.0% and 0.34° / cm, and the LGE result is a three-part variable: LGE negative, LGE∈[0-8.0%) and LGE≥8.0%.

[0113] Step 203, screening target magnetic resonance parameters from the plurality of preset magnetic resonance parameters of the sample patients according to the respective cutoff values, and the cardiac resynchronization labels and the true results of the occurrence of the preset endpoint events of the sample patients.

[0114] In some embodiments, step 203 comprises steps 2031 to 2033.

[0115] Step 2031, determining the first type of patients and the second type of patients corresponding to the respective preset magnetic resonance parameters from the sample patients according to the cardiac resynchronization labels of the sample patients and whether the respective preset magnetic resonance parameters meet the corresponding cutoff values; the first type of patients meet the cutoff values of the preset magnetic resonance parameters and the cardiac resynchronization labels are implanted; the second type of patients meet the cutoff values of the preset magnetic resonance parameters and the cardiac resynchronization labels are not implanted.

[0116] In the present embodiment, meeting the cutoff value of the preset magnetic resonance parameter means that the preset magnetic resonance parameter is in the range judged as not occurring the preset endpoint event in the ROC curve of the cutoff value. Exemplarily, for the LGE enhancement result, meeting the cutoff value means LGE negative or LGE∈[0-8.0%], and not meeting the cutoff value means LGE≥8.0%; for LVEF, meeting the cutoff value means LVEF≤24.6%, and not meeting the cutoff value means LVEF>24.6%.

[0117] In an embodiment of the present application, the sample patients with the cardiac resynchronization label of implanted can be determined as experimental patients, and the sample patients with the cardiac resynchronization label of not implanted can be determined as control patients. Further, the first type of patients corresponding to the respective preset magnetic resonance parameters can be determined from the experimental patients, and the second type of patients corresponding to the respective preset magnetic resonance parameters can be determined from the control patients.

[0118] In an embodiment of the present application, the sample patients with the cardiac resynchronization label of implanted meet the following parameter conditions: LVEF≤35%; the internal diameter or volume of the left ventricular end diastole is greater than the normal reference value corrected by body surface area, gender and / or age by 2 standard deviations; QRS complex time limit≥120ms; QRS morphological characteristics are complete left bundle branch block, complete right bundle branch block or non-specific intraventricular conduction delay.

[0119] Wherein, the internal diameter or volume of the left ventricular end diastole can be calculated by the biplane method. Left atrial maximum volume index=left atrial maximum volume / body surface area. Left atrial maximum volume=0.85×(two-chamber heart plane left atrial area×four-chamber heart plane left atrial area) / left atrial shorter length (two-chamber heart or four-chamber heart plane: midpoint of mitral annulus plane to upper side of left atrium).

[0120] Step 2032, determining whether there is a significant difference between the real results of the occurrence of the preset endpoint event of the first type of patients and the second type of patients corresponding to each of the preset magnetic resonance parameters.

[0121] Step 2033, determining the preset magnetic resonance parameter corresponding to the significant difference between the real results of the occurrence of the preset endpoint event of the first type of patients and the second type of patients as the target magnetic resonance parameter.

[0122] In the embodiment, when there is a significant difference between the real results of the occurrence of the preset endpoint event of the first type of patients and the second type of patients corresponding to the preset magnetic resonance parameter, it indicates that when the preset magnetic resonance parameter threshold is met, the cardiac resynchronization has a significant correlation with the occurrence of the preset endpoint event, and the threshold of the preset magnetic resonance parameter has a positive effect on the cardiac resynchronization. Conversely, when there is no significant difference between the real results of the occurrence of the preset endpoint event of the first type of patients and the second type of patients corresponding to the preset magnetic resonance parameter, it indicates that when the preset magnetic resonance parameter threshold is met, the cardiac resynchronization has no significant correlation with the occurrence of the preset endpoint event, and the threshold of the preset magnetic resonance parameter has no positive effect on the cardiac resynchronization. Therefore, in order to accurately classify the patients based on the target magnetic resonance parameter, the preset magnetic resonance parameter corresponding to the significant difference between the real results of the occurrence of the preset endpoint event of the first type of patients and the second type of patients is determined as the target magnetic resonance parameter.

[0123] The patient classification method based on magnetic resonance parameters provided in the embodiment can determine the patients meeting each threshold from the sample patients by determining the first type of patients and the second type of patients corresponding to each preset magnetic resonance parameter from the sample patients according to the cardiac resynchronization label of the sample patients and whether each preset magnetic resonance parameter meets the corresponding threshold, and then classifying the patients meeting each threshold into the first type of patients and the second type of patients according to the cardiac resynchronization label, determining whether there is a significant difference between the real results of the occurrence of the preset endpoint event of the first type of patients and the second type of patients corresponding to each of the preset magnetic resonance parameters, and then determining the relationship between the cardiac resynchronization and the occurrence of the preset endpoint event when the preset magnetic resonance parameter threshold is met, accurately determining the target magnetic resonance parameter having an effect on the cardiac resynchronization, and accurately classifying the target patients.

[0124] In an embodiment of the present application, in order to verify whether the target magnetic resonance parameter is accurate, third patients and fourth patients corresponding to each preset magnetic resonance parameter can also be determined from the sample patients according to the cardiac resynchronization labels of the sample patients and whether each preset magnetic resonance parameter meets the corresponding cutoff value; the third patients do not meet the cutoff value of the preset magnetic resonance parameter and the cardiac resynchronization label is implanted; the fourth patients do not meet the cutoff value of the preset magnetic resonance parameter and the cardiac resynchronization label is not implanted; the survival rates corresponding to each third patient and each fourth patient are calculated according to the true results of the occurrence of the preset endpoint event of each third patient and each fourth patient; the control survival rate difference of each preset magnetic resonance parameter is calculated according to the survival rates corresponding to each third patient and each fourth patient. The control survival rate difference can be the difference between the survival rates corresponding to each third patient and each fourth patient. If the true survival rate difference corresponding to the preset magnetic resonance parameter is greater than the preset difference value, and the corresponding control survival rate difference is less than the preset difference value, it can be determined that the cutoff value of the target magnetic resonance parameter has an impact on cardiac resynchronization, and the target magnetic resonance parameter is accurate.

[0125] In an embodiment of the present application, step 2032 comprises steps 501 to 502.

[0126] Step 501, according to the true results of the occurrence of the preset endpoint event of the first patients and the second patients corresponding to each of the preset magnetic resonance parameters, determine the significance value of the Kaplan-Meier survival curve of the first patients and the second patients corresponding to each of the preset magnetic resonance parameters; In some embodiments, the true results of the occurrence of the preset endpoint event of the first patients and the second patients can be input into the preset analysis software, and the significance value of the Kaplan-Meier survival curve of the first patients and the second patients is output by the preset analysis software.

[0127] In some embodiments, the significance value of the Kaplan-Meier survival curve can be determined by plotting the Kaplan-Meier survival curve of the first patients and the second patients.

[0128] Step 502, the first patients and the second patients with a significance value of the Kaplan-Meier survival curve less than a preset threshold are determined as the true results of the occurrence of the preset endpoint event having a significant difference.

[0129] In some embodiments, the preset threshold can be 0.05. When the significance value of the Kaplan-Meier survival curve of the first patients and the second patients is p<0.05, it represents that there is a significant difference between the survival curves of the first patients and the second patients. Wherein, this significant difference can be a statistically significant difference.

[0130] If there is a significant difference between the survival curves of the first type of patients and the second type of patients, it indicates that cardiac resynchronization has a significant correlation with the occurrence of the preset endpoint event when the preset magnetic resonance parameter cutoff value is met, and the corresponding preset magnetic resonance parameter is determined as the target magnetic resonance parameter. Conversely, if there is no significant difference between the survival curves of the first type of patients and the second type of patients, such as almost overlapping, it indicates that cardiac resynchronization has no significant correlation with the occurrence of the preset endpoint event when the preset magnetic resonance parameter cutoff value is met, and the corresponding preset magnetic resonance parameter is not determined as the target magnetic resonance parameter.

[0131] In some embodiments, the true result of the occurrence of the preset endpoint event of the patient includes: a follow-up duration, and a result of the occurrence of the preset endpoint event within the follow-up duration. The result of the occurrence of the preset endpoint event within the follow-up duration includes: no occurrence of the preset endpoint event within the follow-up duration, and occurrence of the preset endpoint event at a time point within the follow-up duration. The true results of the occurrence of the preset endpoint event of different types of patients are shown below by Kaplan-Meier survival curves.

[0132] Figure 3 and Figure 4 is the Kaplan-Meier survival curve of the COR class I patients in another embodiment of the present application.

[0133] In Figure 3 and Figure 4 In the corresponding embodiment, the COR class I patients are patients with QRS complex time≥150ms and QRS complex morphological characteristics of left bundle branch block. In the COR class I patients, the patients with implanted cardiac resynchronization tags are determined as the COR class I CRT group, and the patients with non-implanted cardiac resynchronization tags are determined as the COR class I non-CRT group, and the Kaplan-Meier survival curves are shown in Figure 3 and Figure 4 In the Kaplan-Meier survival curves shown in

[0134] As Figure 3 and Figure 4The Kaplan-Meier survival curve shown, compared with the COR class I non-CRT group, the risk of occurrence of the pre-set endpoint event of the COR class I CRT group of patients was significantly reduced, and the survival rate was significantly improved, p=0.04. When the LGE, GLS, GCS, GRS, torsion meet the cutoff value, that is, LGE negative, LGE between 0-8.0%, GLS≥-5.9%, GCS≥-6.3%, GRS≤8.0%, torsion≤0.34° / cm, compared with the COR class I non-CRT group, the risk of occurrence of the pre-set endpoint event of the COR class I CRT group of patients was significantly reduced, and the survival rate was significantly improved, and the significance values p were p=0.02, p=0.04, p<0.001, p<0.001, p<0.001, p<0.001, respectively. When the LVEF meets the cutoff value, that is, LVEF≤24.6%, compared with the COR class I non-CRT group, the risk of occurrence of the pre-set endpoint event of the COR class I CRT group of patients was not significantly reduced, and the survival rate was not significantly improved, p=0.009. When the LVEF, LGE, GLS, GCS, GRS, torsion do not meet the cutoff value, compared with the COR class I non-CRT group, the event-free survival rate of the COR class I CRT group of patients is close, and the significance values p are p=0.934, p=0.60, p=0.79, p=0.76, p=0.94, p=0.70, respectively. Reverse verification that the cutoff value of GLE, GLS, GCS, GRS, torsion has a positive effect on cardiac resynchronization, and the cutoff value of LVEF does not have a positive effect on cardiac resynchronization, proves the accuracy of the target magnetic resonance parameters for patient classification.

[0135] Figure 5 The Kaplan-Meier survival curve of the COR class IIa patients and the COR class IIb patients in another embodiment of the present application.

[0136] In Figure 5In the corresponding embodiment, COR IIa patients are patients whose QRS complex duration is between 120-150ms and whose QRS complex morphology is characterized by left bundle branch block, or patients whose QRS complex duration is ≥150ms and whose QRS complex morphology is characterized by non-left bundle branch block. COR IIb patients are patients whose QRS complex duration is between 120-150ms and whose QRS complex morphology is characterized by non-left bundle branch block. Among COR IIa patients, those whose cardiac resynchronization labels are implanted are determined to be COR IIa CRT group, and those whose cardiac resynchronization labels are not implanted are determined to be COR IIa non-CRT group. Among COR IIb patients, those whose cardiac resynchronization labels are implanted are determined to be COR IIb CRT group, and those whose cardiac resynchronization labels are not implanted are determined to be COR IIb non-CRT group. In Figure 5 In the Kaplan-Meier survival curve shown, each point of the curve decreases indicates that one person in the group has a preset endpoint event, and the number of survivors represents the remaining number of people in the group as the follow-up time increases.

[0137] like Figure 5 The Kaplan-Meier survival curves shown show that the event-free survival rate was similar in the COR IIa CRT group compared with the COR IIa non-CRT group (with a significance value of p = 0.07); the event-free survival rate was similar in the COR IIb CRT group compared with the COR IIb non-CRT group (with a significance value of p = 0.93). When the degree of left ventricular genomic DNA (LGE) was high (i.e., LGE ≥ 8.0%) or when the left ventricular strain and torsion were less impaired, CRT implantation did not significantly improve clinical outcomes; the event-free survival rates were similar between the COR IIa CRT group and the COR IIa non-CRT group, and between the COR IIb CRT group and the COR IIb non-CRT group.

[0138] Figure 6 This is the Kaplan-Meier survival curve of COR IIa patients in another embodiment of the present application.

[0139] exist Figure 6 In the corresponding embodiment, the Figure 5 The same indicators in the corresponding embodiment are used to determine COR IIa patients, and COR IIa patients are divided into a COR IIa non-CRT group and a COR IIa CRT group. Figure 6As shown, when GLS≥-5.9%, GCS≥-6.3%, GRS≤8.0% or torsion≤0.34° / cm, the event-free survival rate of the COR IIa class CRT group is significantly improved compared with the COR IIa class non-CRT group, and the significance values p are p=0.005, p=0.01, p=0.01 and p=0.02 respectively.

[0140] Figure 7 is a schematic diagram of the probability of occurrence of a preset endpoint event of different types of patients in yet another embodiment of the present application.

[0141] In Figure 7 In the corresponding embodiment, the COR I class patients are determined by using the same indexes as in the Figure 3 and Figure 4 In the corresponding embodiment, the COR I class patients are determined by using the same indexes as in the Figure 5 In the corresponding embodiment, the COR IIa class patients are determined by using the same indexes as in the Figure 7 As shown, there are 241 COR I class patients and 255 COR IIa class patients. The probability of occurrence of a preset endpoint event of the COR I class patients is 2.2% when LGE is negative.

[0142] When LGE is negative, the probability of occurrence of a preset endpoint event of the COR I class CRT group is lower and the event-free survival rate is higher compared with the COR I class non-CRT group. The probability of occurrence of a preset endpoint event of the COR IIa class CRT group is lower and the event-free survival rate is higher compared with the COR IIa class non-CRT group. When LGE≥8.0%, GLS<-5.9%, GCS<-6.3%, GRS>8.0% or torsion>0.34° / cm is satisfied respectively, the probability of occurrence of a preset endpoint event of the COR I class CRT group is close to that of the COR I class non-CRT group, and the event-free survival rate of the COR I class CRT group is close to that of the COR I class non-CRT group. The probability of occurrence of a preset endpoint event of the COR IIa class CRT group is close to that of the COR IIa class non-CRT group, and the event-free survival rate of the COR IIa class CRT group is close to that of the COR IIa class non-CRT group. The event-free survival rate refers to the ratio of the number of patients who do not have a preset endpoint event to the total number of patients in the classification.

[0143] In summary, when the cutoff values of LGE, GLS, GCS, GRS and torsion are met, cardiac resynchronization can significantly reduce the probability of occurrence of the preset endpoint event; when the cutoff value of LVEF is met, cardiac resynchronization cannot significantly reduce the probability of occurrence of the preset endpoint event. That is, the cutoff values of LGE, GLS, GCS, GRS and torsion have a positive effect on cardiac resynchronization, and the cutoff value of LVEF does not have a positive effect on cardiac resynchronization. Figure 7 The probability of occurrence of the preset endpoint event of different types of patients in another embodiment of the application is shown, which again verifies the accuracy of the target magnetic resonance parameters determined by the application for classifying patients.

[0144] In some embodiments, in order to accurately classify the target patient, before step 103 "inputting the target magnetic resonance parameters of the target patient into the target classification tree", steps 501 to 505 are further included.

[0145] Step 501: acquiring a preset classification tree; the preset classification tree is used to determine the initial classification result corresponding to the target patient according to the clinical parameters of the target patient; the clinical parameters include at least one of the following: left ventricular ejection fraction LVEF, QRS complex duration, QRS complex morphological characteristics, cardiac function classification; the initial classification result includes a first initial classification and a second initial classification.

[0146] In this embodiment, the preset classification tree can divide the target patient into the first initial classification and the second initial classification according to the QRS complex duration and the QRS complex morphological characteristics of the target patient.

[0147] In some embodiments, the preset classification tree divides the patients in the class I recommendation standard in the guideline issued by the American Heart Association into the first initial classification, and divides the patients in the class IIa and IIb recommendation standard into the second initial classification.

[0148] Illustratively, the preset classification tree divides the target patient with QRS complex duration ≥ 150 ms and left bundle branch block into the first initial classification; divides the target patient with QRS complex duration between 120-150 ms and left bundle branch block, and the target patient with QRS complex duration ≥ 150 ms and non-left bundle branch block into the second initial classification; and divides the target patient with QRS complex duration between 120-150 ms and non-left bundle branch block into the second initial classification.

[0149] Step 502: determining the initial classification result as an initial intermediate node of the target classification tree; the branches of the initial intermediate node include the first initial classification and the second initial classification.

[0150] In the embodiment, the classification result of the target patient by the preset classification tree includes two cases: the target patient is the first initial classification and the target patient is the second initial classification, and therefore the branches of the initial intermediate node include the first initial classification and the second initial classification.

[0151] In step 503, the cutoff value of the gadolinium contrast agent delayed enhancement result is determined as the first intermediate node connected by the branch of the first initial classification, and whether the cutoff value of the gadolinium contrast agent delayed enhancement result is satisfied is determined as the branch of the first intermediate node.

[0152] In the embodiment, for the patient of the first initial classification, since the cardiac resynchronization can reduce the probability of occurrence of the preset endpoint event when the gadolinium contrast agent delayed enhancement result is satisfied, the cutoff value of the gadolinium contrast agent delayed enhancement result is determined as the first intermediate node connected by the branch of the first initial classification, and whether the cutoff value of the gadolinium contrast agent delayed enhancement result is satisfied is determined as the branch of the first intermediate node.

[0153] The first intermediate node has three branches: LGE negative, LGE∈[0-8.0%) and LGE≥8.0%.

[0154] In step 504, the cutoff value of the ventricular strain parameter and the cutoff value of the left ventricular torsion parameter are determined as the second intermediate node connected by the branch of the second initial classification, and whether the first preset condition is satisfied is determined as the branch of the second intermediate node; the first preset condition is that the cutoff value of the myocardial strain parameter or the cutoff value of the left ventricular torsion parameter is satisfied.

[0155] In the embodiment, for the patient of the second initial classification, since the cardiac resynchronization can reduce the probability of occurrence of the preset endpoint event when the cutoff value of the myocardial strain parameter and the cutoff value of the left ventricular torsion parameter is satisfied, the cutoff value of the myocardial strain parameter and the cutoff value of the left ventricular torsion parameter are determined as the second intermediate node connected by the branch of the second initial classification, and whether the first preset condition is satisfied is determined as the branch of the second intermediate node.

[0156] The second intermediate node has two branches: satisfying the first preset condition and not satisfying the first preset condition. Among them, satisfying the first preset condition has the following four cases: GLS≥-5.9%; GCS≥-6.3%; GRS≤8.0%; torsion≤0.34° / cm. Not satisfying the first preset condition has the following one case: GLS>-5.9%, GCS<-6.3%, GRS>8.0% and torsion>0.34° / cm.

[0157] In step 505, the branches of the first intermediate node and the branches of the second intermediate node are connected to the first classification result or the second classification result to obtain the target classification tree according to the influence of satisfying each cutoff value on the cardiac resynchronization.

[0158] In some embodiments, the step 505 "connecting the branches of the first intermediate node and the branches of the second intermediate node to the first classification result or the second classification result according to the influence of meeting the pairs of cutoff values on cardiac resynchronization" comprises steps 5051 to 5052.

[0159] The step 5051 connects the branches having positive influence on cardiac resynchronization to one of the first classification result and the second classification result.

[0160] The step 5052 connects the branches having no positive influence on cardiac resynchronization to the other of the first classification result or the second classification result.

[0161] In this embodiment, since meeting the pair of cutoff values of gadolinium contrast agent delayed enhancement result has positive influence on cardiac resynchronization, LGE negative and LGE∈[0-8.0%) have positive influence on cardiac resynchronization.

[0162] Since meeting the cutoff value of myocardial strain parameter or the cutoff value of left ventricular torsion parameter both have positive influence on cardiac resynchronization, meeting the first preset condition has positive influence on cardiac resynchronization.

[0163] Therefore, the branches LGE negative, LGE∈[0-8.0%) and meeting the first preset condition are connected to the same classification result, and the branches LGE≥8.0% and not meeting the first preset condition are connected to the same classification result. For example, the branches LGE negative, LGE∈[0-8.0%) and meeting the first preset condition are connected to the first classification result, and the branches LGE≥8.0% and not meeting the first preset condition are connected to the second classification result.

[0164] The patient classification method based on magnetic resonance parameters provided in this embodiment can extend a target classification tree on a preset classification tree according to the influence of the pairs of cutoff values of each preset magnetic resonance parameter on cardiac resynchronization, and the target classification tree determined can accurately classify target patients.

[0165] Figure 8 is a logical diagram of the target classification tree classifying target patients in another embodiment of the present application. As shown in Figure 8As shown, in yet another embodiment of the present application, the cardiac function classification comprises first to fourth classifications in ascending order, and the classification is positively correlated with the degree of activity limitation of the patient; the QRS complex morphological feature is left bundle branch block (LBBB) or non-LBBB; the first initial classification satisfies LVEF less than or equal to a first threshold value, QRS complex time interval greater than or equal to a second threshold value, LBBB, and cardiac function classification being second to fourth classifications; the second initial classification satisfies LVEF less than or equal to the first threshold value, QRS complex time interval greater than or equal to the second threshold value, non-LBBB, and cardiac function classification being second to fourth classifications, or LVEF less than or equal to the first threshold value, QRS complex time interval less than the second threshold value and greater than a third threshold value, LBBB, and cardiac function classification being second to fourth classifications, and the step 102 of "classifying the target patient by using the target classification tree to obtain a target classification result corresponding to the target patient" comprises steps 1021 to 1026.

[0166] In the present embodiment, the first threshold value can be 35%, the second threshold value can be 150 ms, and the third threshold value can be 120 ms. Before inputting the target magnetic resonance parameter into the target classification tree, the initial classification result of the target patient is input into the target classification tree. The target classification tree first determines whether the initial classification result of the target patient is the first initial classification or the second initial classification.

[0167] In step 1021, if the initial classification result of the target patient is the first initial classification, it is determined whether the gadolinium contrast agent delayed enhancement result satisfies a cutoff value; the gadolinium contrast agent delayed enhancement result satisfying the cutoff value includes: the gadolinium contrast agent delayed enhancement result being negative, the gadolinium contrast agent delayed enhancement result being positive and the enhancement percentage being ∈ [0, 8%].

[0168] In step 1022, if it is determined that the gadolinium contrast agent delayed enhancement result satisfies the cutoff value, it is determined that the target classification result is one of the first classification result and the second classification result.

[0169] In step 1023, if it is determined that the gadolinium contrast agent delayed enhancement result does not satisfy the cutoff value, it is determined that the target classification result is the other of the first classification result and the second classification result.

[0170] In step 1024, if the initial classification result of the target patient is the second initial classification, it is determined whether a first preset condition is satisfied.

[0171] In step 1025, if it is determined that the first preset condition is satisfied, it is determined that the target classification result is one of the first classification result and the second classification result.

[0172] In step 1026, if it is determined that the first preset condition is not satisfied, it is determined that the target classification result is the other of the first classification result and the second classification result.

[0173] The patient classification method based on the magnetic resonance parameters provided in the embodiment can accurately classify patients.

[0174] In another embodiment of the present application, the step 103 of outputting the target classification result comprises at least one of the step 1031 to the step 1033.

[0175] The step 1031 is to visually display the target classification result.

[0176] The electronic device can include a display device, and the manner of outputting the target classification result can be to visually display the target classification result in the display device. The visually displayed target classification result can be provided to a doctor to assist the doctor in formulating a treatment plan for the target patient.

[0177] The step 1032 is to generate a cardiac resynchronization indication recommendation report according to the target classification result, and output the cardiac resynchronization indication recommendation report.

[0178] The electronic device can be integrated in a magnetic resonance scanning instrument, and can directly obtain the target magnetic resonance parameters from the scanning results of the magnetic resonance scanning instrument. The electronic device can pre-store a template of the cardiac resynchronization indication recommendation report. After classifying the target patient according to the target magnetic resonance parameters, the cardiac resynchronization indication recommendation report can be output according to the target classification result and the pre-stored template of the cardiac resynchronization indication recommendation report.

[0179] The step 1033 is to generate a cardiac resynchronization preoperative evaluation report by using a preset artificial intelligence prediction model, and output the cardiac resynchronization preoperative evaluation report.

[0180] The electronic device can be configured with a preset artificial intelligence prediction model. After the target classification tree determines the target classification result corresponding to the target patient, the target classification result is input into the preset artificial intelligence prediction model. The cardiac resynchronization preoperative evaluation report of the target patient is generated by using the preset artificial intelligence prediction model, and the cardiac resynchronization preoperative evaluation report is output.

[0181] The patient classification method based on the magnetic resonance parameters provided in the embodiment can be used in a cardiology department or an imaging department of a hospital to assist a doctor in formulating a treatment plan for a target patient.

[0182] According to the patient classification method based on the magnetic resonance parameters provided in the above embodiments, the present application also provides a specific implementation manner of a patient classification device based on the magnetic resonance parameters. Please refer to the following embodiments.

[0183] Figure 9 The structure diagram of the patient classification device based on the magnetic resonance parameters provided in another embodiment of the present application is shown inFigure 9 In still another embodiment of the present application, the patient classification device 60 based on magnetic resonance parameters comprises the following units: The first acquisition module 61 is configured to acquire a target magnetic resonance parameter of a target patient; the target magnetic resonance parameter is selected from a plurality of preset magnetic resonance parameters of a sample patient based on a cardiac resynchronization label of the sample patient and a true result of occurrence of a preset endpoint event; the target magnetic resonance parameter comprises at least one of the following: a myocardial strain parameter, a gadolinium contrast agent delayed enhancement result, and a left ventricular torsion parameter; The classification module 62 is configured to input the target magnetic resonance parameter of the target patient into a target classification tree, and classify the target patient by using the target classification tree to obtain a target classification result corresponding to the target patient; the target classification tree is obtained by extending a preset classification tree based on a cutoff value of each target magnetic resonance parameter and an influence of meeting each cutoff value on cardiac resynchronization; The output module 63 is configured to output the target classification result.

[0184] The patient classification device based on magnetic resonance parameters provided in the embodiment can comprehensively evaluate whether the physiological state of the heart of the patient is suitable for cardiac resynchronization by acquiring a target magnetic resonance parameter of a target patient; since the target magnetic resonance parameter is selected from a plurality of preset magnetic resonance parameters of a sample patient based on a cardiac resynchronization label of the sample patient and a true result of occurrence of a preset endpoint event; the target magnetic resonance parameter comprises at least one of the following: a myocardial strain parameter, a gadolinium contrast agent delayed enhancement result, and a left ventricular torsion parameter; the target classification tree can accurately classify the patient according to the effect of cardiac resynchronization of the target patient under different target magnetic resonance parameters by inputting the target magnetic resonance parameter of the target patient into a target classification tree, and classifying the target patient by using the target classification tree to obtain a target classification result corresponding to the target patient; since the target classification tree is obtained by extending a preset classification tree based on a cutoff value of each target magnetic resonance parameter and an influence of meeting each cutoff value on cardiac resynchronization; and the target classification result can be outputted, so that the target patient can be accurately classified.

[0185] As an implementation manner of the present application, in order to accurately classify the target patient, the above-mentioned device can further comprise: The first determination module is configured to determine a control patient according to the cardiac resynchronization label of the sample patient; the cardiac resynchronization label of the control patient is not implanted; The second determination module is configured to determine a cutoff value of each preset magnetic resonance parameter according to the true result of occurrence of the preset endpoint event of the control patient; The screening module is configured to screen target magnetic resonance parameters from the plurality of preset magnetic resonance parameters of the sample patients according to the cutoff values, and the true results of the occurrence of the preset endpoint events of the sample patients and the cardiac resynchronization labels of the sample patients.

[0186] As another implementation manner of the present application, in order to accurately determine the cutoff values of the plurality of preset magnetic resonance parameters, and further accurately classify the target patients, the second determining module can be specifically configured to: According to the true results of the occurrence of the preset endpoint events of the control patients, draw a receiver operating characteristic curve between each preset magnetic resonance parameter and the true results of the occurrence of the preset endpoint events of the control patients; and determine the cutoff value of each preset magnetic resonance parameter according to each receiver operating characteristic curve.

[0187] As another implementation manner of the present application, in order to accurately determine the target magnetic resonance parameter which has a positive impact on the cardiac resynchronization of the patients, so as to accurately classify the target patients, the screening module can be specifically configured to: determine the first type of patients and the second type of patients corresponding to each preset magnetic resonance parameter from the sample patients according to the cardiac resynchronization labels of the sample patients, and whether each preset magnetic resonance parameter meets the corresponding cutoff value; the first type of patients meet the cutoff value of the preset magnetic resonance parameter and the cardiac resynchronization label is implanted; the second type of patients meet the cutoff value of the preset magnetic resonance parameter and the cardiac resynchronization label is not implanted; determine whether there is a significant difference in the true results of the occurrence of the preset endpoint events of the first type of patients and the second type of patients corresponding to each preset magnetic resonance parameter; and determine the preset magnetic resonance parameter for which the true results of the occurrence of the preset endpoint events of the first type of patients and the second type of patients are significantly different, as the target magnetic resonance parameter.

[0188] As another implementation manner of the present application, in order to quickly determine whether there is a significant difference in the true results of the occurrence of the preset endpoint events of the first type of patients and the second type of patients corresponding to each preset magnetic resonance parameter, the screening module can be specifically further configured to: determine the significance value of the Kaplan-Meier survival curve of the first type of patients and the second type of patients corresponding to each preset magnetic resonance parameter according to the true results of the occurrence of the preset endpoint events of the first type of patients and the second type of patients corresponding to each preset magnetic resonance parameter; and determine the first type of patients and the second type of patients with a significance value of the Kaplan-Meier survival curve less than a preset threshold value, as the true results of the occurrence of the preset endpoint events having a significant difference.

[0189] As another implementation manner of the present application, in order to determine the target classification tree which can accurately classify the target patients, the apparatus can further include: The second acquisition module is configured to acquire a preset classification tree; the preset classification tree is used to determine an initial classification result corresponding to the target patient according to a clinical parameter of the target patient; the clinical parameter comprises at least one of the following: left ventricular ejection fraction (LVEF), QRS complex time, QRS complex morphological characteristics, and cardiac function classification; the initial classification result comprises a first initial classification and a second initial classification; The third determination module is configured to determine the initial classification result as an initial intermediate node of the target classification tree; branches of the initial intermediate node comprise the first initial classification and the second initial classification; The fourth determination module is configured to determine a cutoff value of the gadolinium contrast agent delayed enhancement result as a first intermediate node connected by a branch of the first initial classification, and determine whether the cutoff value of the gadolinium contrast agent delayed enhancement result is met as a branch of the first intermediate node. The fifth determination module is configured to determine a cutoff value of the ventricular strain parameter and a cutoff value of the left ventricular torsion parameter as a second intermediate node connected by a branch of the second initial classification, and determine whether a first preset condition is met as a branch of the second intermediate node; the first preset condition is that the cutoff value of the myocardial strain parameter or the cutoff value of the left ventricular torsion parameter is met. The connection module is configured to connect the branches of the first intermediate node and the second intermediate node to one of the first classification result or the second classification result according to an influence of meeting each cutoff value on cardiac resynchronization, to obtain the target classification tree.

[0190] As another implementation manner of the present application, in order to determine the target classification tree capable of accurately classifying the target patient, the connection module can be specifically configured to: connect a branch having a positive influence on cardiac resynchronization to one of the first classification result and the second classification result; and connect a branch not having a positive influence on cardiac resynchronization to the other one of the first classification result or the second classification result.

[0191] As another implementation form of the present application, in order to accurately classify the target patient, the cardiac function classification comprises first to fourth classifications in ascending order, and the classification is positively correlated with the degree of activity limitation of the patient; the QRS complex morphological feature is left bundle branch block (LBBB) or non-LBBB; the first initial classification satisfies LVEF less than or equal to a first threshold, QRS complex time limit greater than or equal to a second threshold, LBBB, and cardiac function classification is the second to fourth classifications; the second initial classification satisfies LVEF less than or equal to the first threshold, QRS complex time limit greater than or equal to the second threshold, non-LBBB, and cardiac function classification is the second to fourth classifications, or satisfies LVEF less than or equal to the first threshold, QRS complex time limit less than the second threshold and greater than a third threshold, LBBB, and cardiac function classification is the second to fourth classifications, and the classification module is specifically configured to: if the initial classification result of the target patient is the first initial classification, determine whether the gadolinium contrast agent delayed enhancement result satisfies a cutoff value; the gadolinium contrast agent delayed enhancement result satisfying the cutoff value includes: the gadolinium contrast agent delayed enhancement result being negative, the gadolinium contrast agent delayed enhancement result being positive and the enhancement percentage being in [0, 8%); if it is determined that the gadolinium contrast agent delayed enhancement result satisfies the cutoff value, determine that the target classification result is one of the first classification result and the second classification result; if it is determined that the gadolinium contrast agent delayed enhancement result does not satisfy the cutoff value, determine that the target classification result is the other of the first classification result and the second classification result; if the initial classification result of the target patient is the second initial classification, determine whether a first preset condition is satisfied; if it is determined that the first preset condition is satisfied, determine that the target classification result is one of the first classification result and the second classification result; if it is determined that the first preset condition is not satisfied, determine that the target classification result is the other of the first classification result and the second classification result.

[0192] As another implementation form of the present application, the output module is specifically configured to at least one of the following: visually display the target classification result; generate a cardiac resynchronization indication recommendation report according to the target classification result, and output the cardiac resynchronization indication recommendation report; generate a cardiac resynchronization preoperative evaluation report by using a preset artificial intelligence prediction model, and output the cardiac resynchronization preoperative evaluation report Figure 7 is a hardware structure schematic diagram of a patient classification device based on magnetic resonance parameters provided by another embodiment of the present application. Referring to Figure 7 In another embodiment of the present application, the patient classification device based on magnetic resonance parameters can include a processor 71 and a memory 72 storing computer program instructions.

[0193] In another embodiment of the present application, the patient classification device based on magnetic resonance parameters can include a processor 71 and a memory 72 storing computer program instructions.

[0194] In particular, the processor 71 can include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to perform the operations of the embodiments of the application.

[0195] The memory 72 can include mass storage for data or instructions. As an example and not by way of limitation, the memory 72 can include a hard disk drive (HDD), a floppy disk drive, flash memory, an optical disc (e.g., a compact disc (CD) or DVD), a tape drive, a USB drive, or a combination of two or more of these. The memory 72 can include removable or non-removable (or fixed) media, where appropriate. The memory 72 can be internal or external to the integrated gateway disaster recovery device, where appropriate. In particular embodiments, the memory 72 is non-volatile, solid-state memory.

[0196] In particular embodiments, the memory 72 includes read-only memory (ROM). Where appropriate, this ROM can be mask programmed ROM, programmable ROM (PROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), electrically alterable ROM (EAROM), or flash memory or a combination of two or more of these. In particular embodiments, the memory 72 includes random-access memory (RAM). Where appropriate, this RAM can include single-data rate RAM, double-data rate RAM, or a combination of two or more of these. Where appropriate, this RAM can be volatile memory or non-volatile memory, or a combination of both. In particular embodiments, the memory 72 includes one or more non-transitory computer-readable storage media.

[0197] The memory can include read-only memory (ROM), random access memory (RAM), magnetic disk storage mediums, optical storage mediums, flash memory devices, electrical, optical, or other physical / tangible memory storage devices. Thus, in general, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., a memory device) encoded with software that, when executed (by one or more processors), is operable to perform operations described with reference to the methods according to the aspects of the present disclosure.

[0198] The processor 71 implements any one of the above-described patient classification methods based on magnetic resonance parameters by reading and executing computer program instructions stored in the memory 72.

[0199] In one example, the patient classification device based on magnetic resonance parameters can further include a communication interface 73 and a bus 74. As shown, the processor 71, the memory 72, and the communication interface 73 are connected through the bus 74 and complete communication between each other. Figure 7

[0200] The communication interface 73 is mainly used to realize the communication between the modules, devices, units and / or equipment in the embodiments of the application.​

[0201] Bus 74 includes hardware, software, or both, coupling components of the online data traffic metering device to each other. By way of example, and not limitation, the bus can include an accelerated graphics port (AGP) or other graphics bus, a peripheral component interconnect (PCI) bus, a PCI-Express bus, a serial advanced technology attachment (SATA) bus, a video electronics standards board (VLB) bus, or another suitable bus or a combination of two or more of these. Where appropriate, bus 74 can include one or more buses. Although the present embodiments describe and show a particular bus, the present embodiments contemplate any suitable bus or interconnect.

[0202] In addition, in combination with the above-mentioned embodiments of the patient classification method based on magnetic resonance, the embodiments of the present application can provide a computer readable storage medium to implement. The computer readable storage medium has computer program instructions stored thereon; the computer program instructions are executed by the processor to implement any one of the above-mentioned embodiments of the patient classification method based on magnetic resonance.

[0203] The embodiments of the present application also provide a computer program product, comprising a computer program, the computer program is executed by the processor to implement any one of the above-mentioned embodiments of the patient classification method based on magnetic resonance.

[0204] It needs to be clear that the present application is not limited to the specific configurations and processes described above and shown in the drawings. For the sake of simplicity, detailed descriptions of known methods are omitted here. In the above-mentioned embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between steps, after understanding the spirit of the present application.

[0205] The functional modules shown in the structural block diagram above can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a functional card, and the like. When implemented in software, the elements of the present application are program or code segments that are used to perform the required tasks. The program or code segments can be stored in a machine-readable medium, or transmitted through a data signal carried in a carrier wave over a transmission medium or communication link. The "machine-readable medium" can include any medium that can store or transfer information. Examples of the machine-readable medium include an electronic circuit, a semiconductor memory device, a ROM, a flash memory, an erasable ROM (EROM), a floppy diskette, a CD-ROM, an optical disk, a hard disk, a fiber optic medium, a radio frequency (RF) link, and the like. The code segments can be downloaded via a computer network, such as the Internet, an intranet, and the like.

[0206] It is also important to note that the examples mentioned in the present application describe some methods or systems based on a series of steps or devices. However, the present application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the examples, or in an order different from the examples, or several steps can be performed simultaneously.

[0207] The above describes the aspects of the present application with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each block in the flowchart and / or block diagram, and the combination of blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus, to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus enable the implementation of the functions / acts specified in one or more blocks of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It can also be understood that each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, can also be implemented by special hardware that performs the specified functions or acts, or can be implemented by a combination of special hardware and computer instructions.

[0208] The above is merely a specific implementation of the present application. As can be clearly understood by a person skilled in the art from the above description, for the convenience and brevity of description, the specific working process of the system, module and unit described above can refer to the corresponding process in the foregoing method embodiments, which will not be described herein again. It should be understood that the protection scope of the present application is not limited in this way, and any person skilled in the art can easily think of various equivalent modifications or replacements within the technical range disclosed in the present application, and these modifications or replacements should be covered in the protection scope of the present application.

Claims

1. A patient classification method based on magnetic resonance parameters, characterized in that: include: acquiring target magnetic resonance parameters of a target patient; The target magnetic resonance parameter is selected from a plurality of preset magnetic resonance parameters of the sample patient based on the cardiac resynchronization signature of the sample patient and the actual result of the occurrence of a preset endpoint event; the target magnetic resonance parameter includes at least one of the following: a myocardial strain parameter, a delayed enhancement result of a gadolinium contrast agent, and a left ventricular torsion parameter; Inputting the target magnetic resonance parameters of the target patient into a target classification tree, and classifying the target patient using the target classification tree to obtain a target classification result corresponding to the target patient; The target classification tree is obtained by extending a preset classification tree based on the cutoff values ​​of the target magnetic resonance parameters and the impact of meeting the cutoff values ​​on cardiac resynchronization; Output the target classification result.

2. The patient classification method based on magnetic resonance parameters according to claim 1, characterized in that: Before acquiring the target magnetic resonance parameters of the target patient, the method further includes: Determining a control patient based on the cardiac resynchronization label of the sample patient; wherein the cardiac resynchronization label of the control patient is not implanted; Determining the cutoff value of each of the preset magnetic resonance parameters according to the actual results of the occurrence of the preset endpoint event in the control patient; Target magnetic resonance parameters are screened from a plurality of preset magnetic resonance parameters of the sample patient according to the cutoff values, the cardiac resynchronization signature of the sample patient, and actual results of occurrence of preset endpoint events.

3. The patient classification method based on magnetic resonance parameters according to claim 2, characterized in that: Determining the cutoff value of each of the preset magnetic resonance parameters based on the actual results of the control patients having the preset endpoint events includes: According to the actual results of the preset endpoint events occurring in the control patients, a receiver operating characteristic curve is drawn between each of the preset magnetic resonance parameters and the actual results of the preset endpoint events occurring in the control patients; The cutoff value of each of the preset magnetic resonance parameters is determined according to each of the receiver operating characteristic curves.

4. The patient classification method based on magnetic resonance parameters according to claim 2, characterized in that: The method of screening a target magnetic resonance parameter from a plurality of preset magnetic resonance parameters of the sample patient according to each of the cutoff values, the cardiac resynchronization signature of the sample patient, and the actual result of the occurrence of a preset endpoint event comprises: Determining, from the sample patients, based on the cardiac resynchronization labels of the sample patients and whether each of the preset magnetic resonance parameters meets a corresponding cutoff value, a first category of patients and a second category of patients corresponding to each of the preset magnetic resonance parameters; the first category of patients meeting the cutoff value of the preset magnetic resonance parameter and having a cardiac resynchronization label implanted; the second category of patients meeting the cutoff value of the preset magnetic resonance parameter and having not had a cardiac resynchronization label implanted; Determining whether there is a significant difference in the actual results of the first category of patients and the second category of patients in the occurrence of the preset endpoint event corresponding to each of the preset magnetic resonance parameters; The preset magnetic resonance parameters corresponding to the actual results of the preset endpoint events occurring in the first category of patients and the second category of patients are determined as the target magnetic resonance parameters.

5. The patient classification method based on magnetic resonance parameters according to claim 4, characterized in that: Determining whether there is a significant difference between the actual results of the preset endpoint events in the first category of patients and the second category of patients corresponding to each of the preset magnetic resonance parameters includes: Determining the significance value of the Kaplan-Meier survival curve of the first category of patients and the second category of patients corresponding to each of the preset magnetic resonance parameters according to the actual results of the preset endpoint events occurring in the first category of patients and the second category of patients; The patients in the first category and the patients in the second category whose significance values ​​of the Kaplan-Meier survival curve were less than the preset threshold were determined to have significant differences in the actual results of the occurrence of the preset endpoint events.

6. The patient classification method based on magnetic resonance parameters according to claim 1, characterized in that: Before inputting the target magnetic resonance parameters of the target patient into the target classification tree, the method further includes: Obtaining a preset classification tree; the preset classification tree is used to determine an initial classification result corresponding to the target patient based on clinical parameters of the target patient; the clinical parameters include at least one of the following: left ventricular ejection fraction (LVEF), QRS complex duration, QRS complex morphological characteristics, and cardiac function classification; the initial classification result includes a first initial classification and a second initial classification; Determining the initial classification result as an initial intermediate node of the target classification tree; the branches of the initial intermediate node include a first initial classification and a second initial classification; determining the cutoff value of the delayed enhancement result of the gadolinium contrast agent as a first intermediate node connected to the first initial classification branch, and determining whether the cutoff value of the delayed enhancement result of the gadolinium contrast agent is met as a branch of the first intermediate node; determining the cutoff value of the ventricular strain parameter and the cutoff value of the left ventricular torsion parameter as a second intermediate node connected to the second initial classification branch, and determining whether a first preset condition is met as a branch of the second intermediate node; the first preset condition is meeting the cutoff value of the myocardial strain parameter or the cutoff value of the left ventricular torsion parameter; According to the influence of satisfying each cutoff value on cardiac resynchronization, the branch of the first intermediate node and the branch of the second intermediate node are connected to the first classification result or the second classification result to obtain the target classification tree.

7. The patient classification method based on magnetic resonance parameters according to claim 6, characterized in that: The step of connecting the branch of the first intermediate node and the branch of the second intermediate node to the first classification result or the second classification result according to the influence of satisfying each cutoff value on cardiac resynchronization comprises: connecting a branch having a positive influence on cardiac resynchronization to one of the first classification result and the second classification result; The branch that does not have a positive influence on performing cardiac resynchronization is connected to the other of the first classification result or the second classification result.

8. The patient classification method based on magnetic resonance parameters according to claim 6, characterized in that: The cardiac function classification includes the first to fourth grades with increasing grades, and the grades are positively correlated with the degree of activity limitation of the patient; the QRS complex morphological characteristics are left bundle branch block LBBB or non-LBBB; the first initial classification satisfies LVEF less than or equal to the first threshold, QRS complex duration greater than or equal to the second threshold, LBBB, and cardiac function classification of the second to fourth grades; the second initial classification satisfies LVEF less than or equal to the first threshold, QRS complex duration greater than or equal to the second threshold, non-LBBB, and cardiac function classification of the second to fourth grades, or satisfies LVEF less than or equal to the first threshold, QRS complex duration less than the second threshold and greater than the third threshold, LBBB, and cardiac function classification of the second to fourth grades; The target patient is classified by using the target classification tree to obtain a target classification result corresponding to the target patient, including: If the initial classification result of the target patient is the first initial classification, determining whether a delayed enhancement result of the gadolinium contrast agent meets a cutoff value; the cutoff value that meets the delayed enhancement result of the gadolinium contrast agent includes: a negative delayed enhancement result of the gadolinium contrast agent, a positive delayed enhancement result of the gadolinium contrast agent and an enhancement percentage ∈ [0, 8%); If it is determined that the delayed enhancement result of the gadolinium contrast agent meets the cutoff value, determining the target classification result to be one of the first classification result and the second classification result; If it is determined that the delayed enhancement result of the gadolinium contrast agent does not meet the cutoff value, determining the target classification result to be the other of the first classification result and the second classification result; If the initial classification result of the target patient is the second initial classification, determining whether the first preset condition is met; If it is determined that the first preset condition is satisfied, determining that the target classification result is one of the first classification result and the second classification result; If it is determined that the first preset condition is not satisfied, the target classification result is determined to be the other of the first classification result and the second classification result.

9. The patient classification method based on magnetic resonance parameters according to claim 1, characterized in that: Outputting the target classification result includes at least one of the following: Visually displaying the target classification results; generating a cardiac resynchronization indication recommendation report according to the target classification result, and outputting the cardiac resynchronization indication recommendation report; A preset artificial intelligence prediction model is used to generate a cardiac resynchronization preoperative evaluation report, and the cardiac resynchronization preoperative evaluation report is output.

10. A patient classification device based on magnetic resonance parameters, characterized in that: The device comprises: an acquisition module, configured to acquire target magnetic resonance parameters of a target patient; the target magnetic resonance parameters are screened from a plurality of preset magnetic resonance parameters of the sample patient based on the cardiac resynchronization signature of the sample patient and the actual occurrence of a preset endpoint event; the target magnetic resonance parameters include at least one of the following: a myocardial strain parameter, a delayed enhancement result of a gadolinium contrast agent, and a left ventricular torsion parameter; a classification module, configured to input target magnetic resonance parameters of a target patient into a target classification tree, and classify the target patient using the target classification tree to obtain a target classification result corresponding to the target patient; the target classification tree is obtained by extending a preset classification tree based on cutoff values ​​of each target magnetic resonance parameter and the impact of meeting each cutoff value on cardiac resynchronization; An output module is used to output the target classification result.

11. A patient classification device based on magnetic resonance parameters, characterized in that: The patient classification based on magnetic resonance parameters includes: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, the patient classification method based on magnetic resonance parameters according to any one of claims 1 to 9 is implemented.

12. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the patient classification method based on magnetic resonance parameters according to any one of claims 1 to 9 is implemented.

13. A computer program product, characterized in that When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device is caused to perform the patient classification method based on magnetic resonance parameters according to any one of claims 1 to 9.

Images