Method for predicting occurrence of HFpEF of myocardial infarction patient based on epicardial fat volume

Abstract

The invention belongs to the field of intelligent medical treatment, and particularly relates to a method for predicting occurrence of HFpEF of a myocardial infarction patient based on epicardial fat volume. The method comprises the following steps: acquiring a CMR image of a myocardial infarction patient; the EAT entropy and the left ventricle epicardial fat volume are extracted on the basis of the CMR image, and the EAT entropy is the information entropy obtained through calculation on the basis of an EAT area in the CMR image; and inputting the EAT entropy and the left ventricular epicardial fat volume into a classifier to obtain the HFpEF generation risk of the myocardial infarction patient. The application finds that the left ventricular epicardial volume and the EAT entropy are independent predictive factors for the occurrence of HFpEF of a myocardial infarction patient; and the prediction value of the combination of the two to the HFpEF is found.

Description

Technical Field

[0001] The present invention relates to the field of intelligent medical care, and more specifically, to a method, device, medium and program product for predicting the occurrence of HFpEF in patients with myocardial infarction based on epicardial fat volume. Background Art

[0002] As an endocrine organ, pericardial adipose tissue is closely related to the occurrence and development of coronary artery disease (CAD). In particular, epicardial adipose tissue (EAT) can regulate perivascular inflammation and vascular remodeling through proinflammatory signals formed by paracrine pathways and vasa vasorum secretion mechanisms, thereby affecting the progression of coronary atherosclerosis and cardiac function. It is an important imaging indicator for cardiovascular risk classification. Therefore, non-invasive imaging analysis of pericardial fat helps to identify high-risk cardiovascular patients and is of great significance for the clinical diagnosis and treatment of CAD.

[0003] When studying epicardial fat tissue on cardiac MRI, because epicardial fat is thinly attached to the outside of the myocardium and difficult to identify on the image, the pericardial fat close to the epicardium is often analyzed together during the study, which is called paracardial fat. Currently, the study of paracardial fat is generally done by users manually outlining the pericardial contour and manually determining the high-signal fat area within the limited pericardial contour by adjusting the image signal threshold.

[0004] But in fact, EAT contains three fat components, and different components have different effects on cardiovascular diseases and will have different changes in the course of the disease, thus bringing about the heterogeneity of the epicardium. Among them, the excessive activation of inflammatory cells may lead to chronic inflammation, promote atherosclerosis and heart disease, but its role in the occurrence of heart failure with preserved ejection fraction (HFpEF) in patients with myocardial infarction (MI) is still unclear. Summary of the invention

[0005] In view of the above problems, the present invention provides a method for predicting the occurrence of HFpEF in patients with myocardial infarction based on epicardial fat volume, using cardiac magnetic resonance (CMR) to quantify the total and paraventricular volumes of epicardial fat (EAT) and evaluate their heterogeneity, and explore the heterogeneity of EAT and the predictive value of different EAT volumes combined with inflammatory cells for the occurrence of HFpEF in MI patients with normal left ventricular ejection fraction (LVEF).

[0006] The present application (first aspect) discloses a method for predicting the occurrence of HFpEF in patients with myocardial infarction based on epicardial fat volume, comprising:

[0007] Acquire CMR images of patients with myocardial infarction; extract EAT entropy and left ventricular epicardial fat volume based on the CMR images, wherein the EAT entropy is information entropy calculated based on the EAT region in the CMR images; input the EAT entropy and left ventricular epicardial fat volume into a classifier to obtain the risk of HFpEF in patients with myocardial infarction.

[0008] Furthermore, the EAT entropy is the Shannon entropy calculated based on the distribution of pixel values ​​in the EAT region after segmenting the CMR image into the EAT region.

[0009] Further, the calculation method of the EAT entropy is:

[0010] Among them, i∈[0,255] represents the pixel grayscale value, and p(i) represents the probability of pixel value i appearing in the EAT area.

[0011] Furthermore, the image that best reflects the target area is called the key frame, and the EAT entropy is calculated based on the information entropy of the EAT area in the key frame; Further, the EAT entropy of each frame is first calculated in the CMR image, and the EAT entropy of all frames is averaged to obtain the EAT entropy; Furthermore, the left ventricular epicardial fat volume is the left ventricular epicardial fat volume demarcated by the ventricular septum.

[0012] Furthermore, clinical data of myocardial infarction patients are obtained at the same time, and the clinical data include one or more of the following: whether the patient suffers from diabetes and BMI; the clinical data, EAT entropy, and left ventricular epicardial fat volume are input into the classifier to obtain the risk of HFpEF in myocardial infarction patients.

[0013] Further, the method for extracting EAT entropy and left ventricular epicardial fat volume based on the CMR image is: Step 1: Input the CMR image into the epicardial fat segmentation model to output the first segmentation result, which includes the total epicardial fat area, the left ventricular epicardial fat area, and the right ventricular epicardial fat area; Step 2, using morphological processing to eliminate the segmentation error area in the first segmentation result to obtain a second segmentation result, wherein the morphological processing includes closing operation and connected domain analysis; Step 3: Use conditional random fields to optimize the segmentation boundary of the second segmentation result to obtain the third segmentation result; Step 4: Perform three-dimensional reconstruction based on the third segmentation result to obtain three-dimensional epicardial fat; Step 5: Calculate the EAT entropy and EAT volume of epicardial fat based on the three-dimensional epicardial fat.

[0014] Furthermore, the LVEF of the myocardial infarction patient is normal.

[0015] Furthermore, the method for constructing the classifier includes: Acquire EAT volume, EAT entropy and labels of myocardial infarction patients in a training set, wherein the labels are HFpEF and non-HFpEF; The EAT volume and EAT entropy are input into the machine learning model to output a predicted label, and the predicted label is compared with the label, and then the machine learning model is optimized and iterated to the stopping condition to obtain the classifier.

[0016] Furthermore, the machine learning model includes any one or more of the following: support vector machine, logistic regression, and XGBoost.

[0017] Furthermore, the epicardial fat segmentation model is constructed as follows: Obtain CMR images and annotations of the training set, including epicardial fat, left ventricular epicardial fat, and right ventricular epicardial fat; The CMR images and annotations of the training set are input into the neural network model and then iteratively trained to obtain the epicardial fat segmentation model; Furthermore, the neural network model includes any one or more of the following: U-Net, SwinTransformer, LTSM.

[0018] The second aspect of the present application discloses a prediction system for the occurrence of HFpEF in myocardial infarction patients based on epicardial fat volume, comprising: Acquisition module: used to acquire CMR images of patients with myocardial infarction; Feature extraction module: used to extract EAT entropy and left ventricular epicardial fat volume based on the CMR image, wherein the EAT entropy is information entropy calculated based on the EAT region in the CMR image; Prediction module: used to input EAT entropy and left ventricular epicardial fat volume into the classifier to obtain the risk of HFpEF in patients with myocardial infarction.

[0019] The third aspect of the present application discloses a computer device, which includes: a memory and a processor; the memory is used to store program instructions; the processor is used to call the program instructions, and when the program instructions are executed, it is used to execute the steps of the above method.

[0020] A fourth aspect of the present application discloses a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a processor.

[0021] A fifth aspect of the present application discloses a computer program product, including a computer program, which implements the steps of the above method when executed by a processor.

[0022] This application has the following beneficial effects: (1) This application is the first to perform fine segmentation and study of epicardial fat and found that left ventricular epicardial volume is an independent predictor of HFpEF in patients with myocardial infarction; (2) This application introduced Shannon entropy into the calculation of EAT area and found that EAT entropy is an independent predictor of HFpEF in patients with myocardial infarction; (3) The application found that the left ventricular epicardial volume and EAT entropy have better prediction effect when combined; (4) This application proposes a model and method for automatically segmenting CMR and calculating the left ventricular epicardial volume and EAT entropy through an artificial intelligence model, which optimizes the measurement process of the left ventricular epicardial volume and EAT entropy. It first proposes to use an artificial intelligence deep learning algorithm model to realize automatic segmentation and automatic reconstruction of the three-dimensional model of paracardial fat. The segmented epicardial fat is analyzed for parameters such as volume, energy, and entropy, which provides intuitive quantitative parameter support for the analysis of epicardial fat and its impact on heart disease. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 is a schematic diagram of a method flow chart provided by the first aspect of an embodiment of the present invention; Figure 2 is a schematic diagram of a program product provided by the second aspect of an embodiment of the present invention; Figure 3 is a schematic diagram of a computer device provided by an embodiment of the present invention; Figure 4 is a schematic diagram of the architecture of an exemplary computing device provided by an embodiment of the present invention; Figure 5 is a schematic diagram of a storage medium provided by an embodiment of the present invention; Figure 6 It is a schematic diagram of a method for marking or measuring overall and local EAT provided by an embodiment of the present invention; Figure 7 is a schematic diagram of a method for measuring EAT entropy provided by an embodiment of the present invention; Figure 8is a flow chart of a model training provided by an embodiment of the present invention; Fig. 9 is a schematic diagram of a comparison of EAT parameters between a No-HFpEF group and a HFpEF group provided by an embodiment of the present invention; Fig.10 This is a schematic diagram of correlation analysis between EAT and inflammatory cells provided by an embodiment of the present invention; Fig.11 It is a schematic diagram of ROC curve analysis of an EAT parameter for predicting HFpEF in MI provided by an embodiment of the present invention; Fig.12 It is a schematic diagram of a Kaplan-Meier survival curve provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0026] In some of the processes described in the specification and claims of the present invention and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The sequence numbers of the operations, such as S101, S102, etc., are only used to distinguish between different operations, and the sequence numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0028] Figure 1 1 is a flow chart of a method for predicting the occurrence of HFpEF in patients with myocardial infarction based on the evaluation of epicardial fat volume provided by an embodiment of the present invention. Specifically, the method comprises the following steps: S101 obtains CMR images of patients with myocardial infarction; S102: extracting EAT entropy and left ventricular epicardial fat volume based on the CMR image, wherein the EAT entropy is information entropy calculated based on the EAT region in the CMR image; S103: Input EAT entropy and left ventricular epicardial fat volume into the classifier to obtain the risk of HFpEF in myocardial infarction patients. Since this application needs to extract the volume of different components of epicardial fat, firstly, an automatic segmentation calculation model of epicardial fat volume is established through a training model.

[0029] 1. Model Construction This application first manually annotates cardiac MRI images and uses an artificial intelligence deep learning algorithm model for training. It can realize automatic segmentation and reconstruction of epicardial fat on the short-axis movie sequence of cardiac MRI images, mark the area of ​​epicardial fat with color on the original short-axis movie sequence image, reconstruct a three-dimensional model of epicardial fat that can be rotated 360 degrees, and analyze parameters such as the volume of epicardial fat.

[0030] The present invention automatically segments epicardial fat in CMR (cardiac magnetic resonance) images, using deep learning + medical image processing to improve segmentation accuracy and automation. Traditional manual segmentation methods require doctors to outline frame by frame, which is not only time-consuming but also easily affected by subjectivity and difficult to ensure consistency. The present invention is based on an improved U-Net / Transformer segmentation network, combined with morphological post-processing and three-dimensional reconstruction algorithms, to achieve efficient and accurate automatic segmentation of paracardiac fat and provide fat quantitative analysis.

[0031] The process of the automatic segmentation method of paracardiac fat used in the present invention is as follows: Figure 8 As shown, the following steps are included: Step 1: Obtain CMR image dataset In some embodiments, the imaging dataset includes a dataset of healthy people.

[0032] In some embodiments, the image dataset further includes an image dataset of a patient with myocardial infarction.

[0033] Image preprocessing: Due to the low contrast of CMR images, the grayscale value of epicardial fat is close to that of surrounding tissues, which can easily lead to missegmentation. Therefore, the following preprocessing is required: Normalization: Normalize the grayscale value of CMR images to [0,1] to reduce the impact of different scanning parameters; use non-local mean filtering (NLM) to remove noise while retaining fat boundary features; use adaptive Gaussian filtering to reduce artifact interference and improve model input quality.

[0034] Step 2: Label the training set In order to ensure that the deep learning model can accurately identify paracardiac fat, a high-quality training dataset needs to be constructed. The process is as follows: Doctor annotation: Experienced cardiovascular imaging experts manually outline the CMR images to determine the precise boundaries of the paracardiac fat.

[0035] The annotations are as follows Figure 6 As shown, Figure 6 -Figure A is the original CMR image, Figure 6 -The red coil in B delineates the visceral epicardium, the green coil delineates the parietal fat layer, and the yellowish area between the green and red coils is the total epicardial fat; Figure 6 -The gray coil in C delineates the total epicardial fat minus the left ventricular epicardial fat, and the light yellow area between the green and red coils minus the gray coil represents the right ventricular epicardial fat; Figure 6 The gray coil in -D delineates the total epicardial fat minus the ventricular epicardial fat, and the light yellow area between the green and red coils minus the gray coil represents the left ventricular epicardial fat; Three types of EAT were identified by such annotation: ① total epicardial fat, ② right ventricular epicardial fat, and ③ left ventricular epicardial fat.

[0036] The above annotations of the CMR images in the training set serve as supervision during model training.

[0037] Semi-automatic annotation tools can be used to assist doctors during annotation, thereby improving annotation efficiency and reducing subjective errors.

[0038] Step 3: Initial Segmentation (Deep Learning Model) In some embodiments, the segmentation backbone network selects a U-Net network: it is suitable for local feature learning and can better handle fat areas with regular shapes.

[0039] In some embodiments, the segmentation backbone network selects Transformer (Swin Transformer): used to capture long-distance dependencies and enhance the recognition ability of complex morphological fat areas.

[0040] In some embodiments, a hybrid loss function is used when training the segmentation model: Dice Loss (to improve the ability to recognize small objects) + Focal Loss (to reduce the impact of difficult-to-segment areas) In some embodiments, the training process uses two-stage training: the first stage (coarse segmentation): using global context information to locate the approximate area of ​​​​paracardiac fat. The second stage (fine segmentation): using high-resolution feature maps to optimize boundaries and improve segmentation accuracy.

[0041] Step 4: Error optimization and post-processing after model segmentation The segmentation results may contain errors, such as: mis-segmentation (non-fat areas are mistaken for fat), missed segmentation (part of the fat is not identified); To this end, the following optimization strategies are used: 1. Morphological processing, such as closing operation: filling small holes to improve the coherence of fat areas; connected domain analysis: removing isolated small areas to ensure that the final result only contains pericardial fat. 2. CRF (conditional random field) optimization, combining the grayscale information of pixels with the spatial neighborhood information, optimizing the segmentation boundary and improving accuracy.

[0042] Step 5: 3D reconstruction In order to provide more intuitive analysis, the Marching Cubes algorithm is used to perform 3D reconstruction of the segmentation results: Input: post-processed 2D fat mask image, Output: 3D paracardiac fat / 3D epicardial fat model. It allows doctors to observe fat distribution in a 3D environment, which can be used for clinical applications such as surgical planning and disease assessment.

[0043] Step 6: Quantification of paracardiac fat After segmentation, the system automatically calculates the following fat quantitative indicators: ①Total fat volume (EAT Volume): the total volume of epicardial fat; ②LV EAT: left ventricular epicardial volume; ③RV EAT: right ventricular epicardial volume; ④Fat density (HU value): Calculate the average gray value of fat and analyze its biological activity.

[0044] ⑤Fat entropy value: reflects the uniformity of fat distribution and can be used to identify abnormal fat tissue.

[0045] ⑥ Cardiac function parameters: Combined with left ventricular ejection fraction (LVEF), the effect of paracardiac fat on cardiac function was analyzed.

[0046] 2. Research Methods: 2.1 Research subjects This is a historical cohort study. Patients who were clinically and CMR-diagnosed with MI but normal LVEF in the Second Affiliated Hospital of Kunming Medical University between January 2015 and July 2023 were enrolled. They were followed up with the occurrence of HFpEF as the endpoint event and divided into a non-HFpEF group and a HFpEF group.

[0047] 2.2 Image processing: For the acquired CMR images, the parameters such as cardiac structure, function, EAT volume (including total epicardial fat volume, ② right ventricular epicardial fat, ③ left ventricular epicardial fat), and infarct volume were obtained using the CMR post-processing software CVI-42, and the EAT heterogeneity parameters were obtained using Python software.

[0048] EAT heterogeneity parameters include: LADSV, LAESV, EDVI, ESVI, CO, CI, infarct size, GRS, GCS, and GLS.

[0049] In some embodiments, the acquired CMR images are input into the constructed model for automatic segmentation and post-processing optimization to output parameters such as EAT volume (including total epicardial fat volume, ② right ventricular epicardial fat, ③ left ventricular epicardial fat).

[0050] Among them, the calculation formula of EAT entropy extracted by Python software at one level is as follows:

[0051]

[0052] For grayscale images (8-bit, pixel value range 0~255), the calculation formula of information entropy H is:

[0053]

[0054] Where: p(i) represents the probability of pixel value i appearing in the image (that is, the frequency of the pixel value divided by the total number of pixels).

[0055] The calculation steps of EAT entropy in this application (Python example) are:

[0056] Step 1: Count pixel histogram

[0057] Calculate the EAT area segmented from the image ( Figure 7 -A, where Figure 7 -A is corresponding Figure 6 The frequency of occurrence of each pixel value in the EAT area between the green line and the red line in the middle is normalized to obtain the probability distribution p(i) ( Figure 7 -B).

[0058] Step 2: Calculate entropy

[0059] Traverse all possible pixel values ​​(0~255) and calculate the EAT entropy of the current slice according to the following formula.

[0060] In some embodiments, a CMR frame that best represents the heart region is selected as a key frame, and the EAT entropy in the key frame is calculated.

[0061] In some embodiments, the method of extracting EAT entropy is to save the original and outlined png images of each outline layer at the same time based on the original EAT collection. The images of each layer can be imported into python software to automatically quantify the entropy value of each layer, and finally take the average value.

[0062] In some embodiments, the delineated image is imported into Python, which combines all pixel values ​​at each level to calculate the frequency of occurrence of each pixel value of the EAT region segmented in the image, and normalizes it to obtain a probability distribution p ( i ), and further based on all levels of E-A-T p ( i ) calculated EAT entropy.

[0063] In some embodiments, LV EAT is also obtained by removing RVEAT based on the original EAT collection with the ventricular septum as the boundary, wherein the standard EAT collection method is as follows: the visceral layer of the epicardium and the parietal layer of the fat wall are manually marked at the end of diastole in the two-chamber short-axis sequence, and the high-signal fat tissue between the two is marked with a signal intensity threshold, while avoiding the coronary arteries and pericardial fat.

[0064] 2.3 Statistical analysis Independent sample t test, nonparametric test and chi-square test were used to analyze the clinical baseline data and CMR indicators of the two groups of patients. Spearman rank correlation was used to analyze the correlation between EAT parameters and inflammatory cells in patients. Univariate and multivariate Cox regression was further performed to analyze the predictive value of each indicator for HFpEF in MI patients. ROC curves were drawn to evaluate the efficacy of each parameter in predicting HFpEF. Finally, the Kaplan-Meier event survival curve was used to display the cumulative incidence curve according to the optimal critical value.

[0065] 3 Results 3.1 Comparison of baseline data In the HFpEF group, BMI was 24.40 (22.23, 26.73) vs 23.40 (21.60, 25.55) (kg / m2), diabetes n (%) 29 (39.19) vs 27 (20.93), renal failure 7 (9.46) vs 3 (2.33) n (%), white blood cell count 7.38 (6.15, 9.02) vs 7.12 (5.96, 7.95) (10*9 / L), neutrophil count 4.58 (3.59, 5.77) vs 4.06 (3.22, 4.81) (10*9 / L), monocyte count 0.51 (0.38, 0.65) vs 0.45 (0.37, 0.53) (10*9 / L), total EAT volume 69.43±21.21 vs. 62.18±19.93 (ml), LV EAT 24.81±7.39 vs. 18.45 (13.03, 25.30) (ml), RV EAT 44.61±15.03 vs. 40.81±13.68 (ml), LAESV 56.18 (42.13, 73.00) vs 51.24 (39.46, 64.44) (ml) were all higher than those in the group without HFpEF; The EAT entropy was 6.40 (6.22, 6.86) vs 6.75 (6.26, 7.15), and TG was 1.40 (0.86, 2.05) vs 1.63 (1.20, 2.40), which were lower than those in the group without HFpEF (p<0.05).

[0066] There were no significant differences in gender, age, smoking history, history of hypertension, systolic blood pressure, diastolic blood pressure, pulse pressure, NYHA classification, LDL, HDL, TC, lymphocytes, eosinophils, basophils, red blood cells, urea, creatinine, RV EAT, LVEF, LADSV, EDVI, ESVI, CO, CI, infarct size, LV global strain, mitral regurgitation, tricuspid regurgitation, aortic regurgitation, pulmonary hypertension, diseased vascular branches, and diseased vessels between the two groups (P>0.05) (as shown in Table 1 and Fig. 9 as shown).

[0067] Table 1 Comparison of clinical baseline data between the two groups

[0068] 3.2 Correlation analysis between EAT and inflammatory cells

[0069] Total EAT, LV EAT, and RV EAT were all slightly to moderately positively correlated with leukocytes and monocytes. Among them, total EAT had the strongest correlation with monocytes (r=0.658, p<0.001), and RV EAT had the strongest correlation with leukocytes (r=0.469, p<0.001). (Table 2 and Fig.10 )

[0070] Table 2 Correlation analysis between EAT and inflammatory cells

[0071] 3.3 Analysis of independent risk factors for HFpEF in MI patients

[0072] The indicators that were different between the two groups and were considered to predict the occurrence of HFpEF were used as independent variables, and HFpEF was used as the dependent variable; univariate Cox regression analysis showed that the risk factors for the occurrence of HFpEF were renal failure, diabetes, BMI, monocytes, total EAT, LV EAT, and EAT entropy. (p < 0.05). In the multivariate Cox analysis model, LV EAT (HR: 1.094, 95% CI: 1.019-1.174), EAT entropy (HR: 0.398, 95% CI: 0.220-0.721), diabetes (HR: 2.320, 95% CI: 1.398-3.849) and BMI (HR: 1.093, 95% CI: 1.025-1.165) were independent predictors of the occurrence of HFpEF (Table 3). ROC analysis was performed to obtain: AUC for BMI was 0.581, AUC for LV EAT was 0.628, AUC for EAT entropy was 0.405, and the combined predictive value of LV EAT and EAT entropy was the highest, with AUC of 0.740 (Table 4 and Fig.11 After a median follow-up of 27 years (range, 2-120 months), Kaplan-Meier survival curves showed that LV EAT greater than 16.42 ml was associated with the development of HFpEF, while EAT entropy was not ( Fig.12 ).

[0073] Table 3 Univariate and multivariate Cox regression analysis

[0074] Table 4 ROC analysis

[0075]

[0076] Table Notes: * Indicates p<0.05.

[0077] A total of 203 eligible MI patients were included, of which 74 patients developed HFpEF and 129 patients did not develop HFpEF. There were no differences in age, gender, and infarct volume between the two groups; however, there were significant statistical differences in BMI, diabetes, renal failure, leukocytes, neutrophils, monocytes, total EAT, EAT entropy, left ventricular EAT (LV EAT), LAESV, and TG (P < 0.05). Total and local EAT were positively correlated with leukocytes and monocytes. Univariate and multivariate Cox regression analysis showed that BMI, diabetes, LV EAT, and EAT entropy were independent risk factors for HFpEF. Further ROC analysis showed that the AUC of BMI was 0.581, the AUC of LV EAT was 0.628, and the AUC of EAT entropy was 0.405. The combined predictive value of LV EAT and EAT entropy was the highest, with an AUC of 0.740. After a median follow-up of 27 years (range, 2-120 months), Kaplan-Meier survival curves showed that LV EAT greater than 16.42 ml was associated with incident HFpEF, regardless of EAT entropy.

[0078] Conclusion: In MI patients with normal LVEF, the occurrence of HFpEF was not correlated with MI volume, but BMI, diabetes, LV EAT and EAT entropy were independent risk factors for HFpEF and had good predictive value. The combination of EAT entropy and LV EAT had the highest predictive efficiency. In addition, both total and local EAT were moderately positively correlated with leukocytes and monocytes. This suggests that in addition to traditional indicators, clinical attention should be paid to EAT heterogeneity and left ventricular EAT for MI patients with normal LVEF to improve the prediction of the risk of HFpEF in MI patients.

[0079] In some embodiments, a CMR image of the patient is obtained, and based on the steps in the first part, EAT entropy and LVEAT (volume) are extracted and input into a classifier model. According to the data, the prediction result of the patient developing HFpEF is 1, which means that the risk of developing HFpEF is high and targeted prevention should be carried out as early as possible.

[0080] Figure 3 is a schematic diagram of a computer device provided by an embodiment of the present invention, such as Figure 3 As shown, the device 2000 may include: one or more processors 2010, and one or more memories 2020; wherein the memories store computer-readable codes, and when the computer-readable codes are run by the one or more processors, the method described above may be executed.

[0081] The processor in this embodiment can be an integrated circuit chip with signal processing capabilities. The above processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The disclosed methods, operations and logic block diagrams in the embodiments of the present disclosure can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc., which can be an X86 architecture or an ARM architecture.

[0082] In general, various example embodiments of the present disclosure may be implemented in hardware or dedicated circuits, software, firmware, logic, or any combination thereof. Certain aspects may be implemented in hardware, while other aspects may be implemented in firmware or software that may be executed by a controller, microprocessor, or other computing device. When various aspects of the disclosed embodiments are illustrated or described as block diagrams, flow charts, or using some other graphical representation, it will be understood that the blocks, devices, systems, techniques, or methods described herein may be implemented in hardware, software, firmware, dedicated circuits or logic, general purpose hardware or controllers or other computing devices, or some combination thereof as non-limiting examples.

[0083] For example, the method or device according to the embodiment of the present disclosure may also be implemented by Figure 4 The architecture of the computing device 3000 shown in FIG. Figure 4 As shown, the computing device 3000 may include a bus 3010, one or more CPUs 3020, a read-only memory (ROM) 3030, a random access memory (RAM) 3040, a communication port 3050 connected to a network, an input / output component 3060, a hard disk 3070, etc. The storage device in the computing device 3000, such as ROM 3030 or hard disk 3070, may store various data or files used for processing and / or communication of the method provided by the present disclosure and program instructions executed by the CPU. The computing device 3000 may also include a user interface 3080. Of course, Figure 4 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 4 One or more components of a computing device are shown.

[0084] The embodiment of the present invention also provides a computer-readable storage medium, such as Figure 5As shown, it is a schematic diagram of a storage medium 4000 provided in an embodiment of the present invention, and a computer readable instruction 4010 is stored on the computer storage medium 4020. When the computer readable instruction 4010 is executed by a processor, the method according to the embodiment of the present disclosure described with reference to the above figures can be executed. The computer readable storage medium in the embodiment of the present disclosure can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. The non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM) or a flash memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus random access memory (DR RAM). It should be noted that the memory of the methods described herein is intended to include, but is not limited to, these and any other suitable types of memory. It should be noted that the memory of the methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0085] The present disclosure also provides a computer program product or a computer program, which implements the steps of the above method when executed by a processor, such as Figure 2 As shown, the computer program product or computer program comprises: Acquisition module 201: used to acquire CMR images of patients with myocardial infarction; Feature extraction module 202: used to extract EAT entropy and left ventricular epicardial fat volume based on the CMR image, wherein the EAT entropy is information entropy calculated based on the EAT region in the CMR image; Prediction module 203: used to input EAT entropy and left ventricular epicardial fat volume into a classifier to obtain the risk of HFpEF in patients with myocardial infarction.

[0086] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment, or a part of a code, and the module, program segment, or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0087] In general, various example embodiments of the present disclosure may be implemented in hardware or dedicated circuits, software, firmware, logic, or any combination thereof. Certain aspects may be implemented in hardware, while other aspects may be implemented in firmware or software that may be executed by a controller, microprocessor, or other computing device. When various aspects of the disclosed embodiments are illustrated or described as block diagrams, flow charts, or using some other graphical representation, it will be understood that the blocks, devices, systems, techniques, or methods described herein may be implemented in hardware, software, firmware, dedicated circuits or logic, general purpose hardware or controllers or other computing devices, or some combination thereof as non-limiting examples.

[0088] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0089] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0090] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0091] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0092] The exemplary embodiments of the present disclosure described in detail above are merely illustrative and not restrictive. It should be understood by those skilled in the art that various modifications and combinations may be made to these embodiments or their features without departing from the principles and spirit of the present disclosure, and such modifications should fall within the scope of the present disclosure.

Claims

1. A method for predicting HFpEF in patients with myocardial infarction based on epicardial fat volume, characterized in that: The method comprises: Obtain CMR images of patients with myocardial infarction; Extracting image features based on the CMR image, the image features including: EAT entropy and left ventricular epicardial fat volume, wherein the EAT entropy is information entropy calculated based on the epicardial fat area in the CMR image; The image features are input into a classifier to obtain a result of whether the myocardial infarction patient will develop HFpEF.

2. The method for predicting HFpEF in patients with myocardial infarction based on epicardial fat volume according to claim 1, characterized in that: The EAT entropy is the Shannon entropy calculated based on the distribution of pixel values ​​in the epicardial fat region after the epicardial fat region is segmented from the CMR image.

3. The method for predicting HFpEF in patients with myocardial infarction based on epicardial fat volume according to claim 1, characterized in that: The calculation method of the EAT entropy is: Among them, i∈[0,255] represents the pixel grayscale value, and p(i) represents the probability of pixel value i appearing in the epicardial fat area.

4. The method for predicting HFpEF in patients with myocardial infarction based on epicardial fat volume according to claim 1, characterized in that: The left ventricular epicardial fat volume is the left ventricular epicardial fat volume demarcated by the interventricular septum.

5. The method for predicting HFpEF in patients with myocardial infarction based on epicardial fat volume according to claim 1, characterized in that: At the same time, clinical data of the myocardial infarction patient is obtained, and the clinical data includes one or more of the following: whether the patient suffers from diabetes and BMI; the clinical data and imaging features are input into the classifier to obtain the result of whether HFpEF will occur.

6. The method for predicting HFpEF in patients with myocardial infarction based on epicardial fat volume according to claim 1, characterized in that: The method for extracting image features based on the CMR image is: Step 1: Input the CMR image into the epicardial fat segmentation model to output the first segmentation result, which includes the total epicardial fat area, the left ventricular epicardial fat area, and the right ventricular epicardial fat area; Step 2, using morphological processing to eliminate the segmentation error area in the first segmentation result to obtain a second segmentation result, wherein the morphological processing includes closing operation and connected domain analysis; Step 3: Use conditional random fields to optimize the segmentation boundary of the second segmentation result to obtain the third segmentation result; Step 4: Perform three-dimensional reconstruction based on the third segmentation result to obtain three-dimensional epicardial fat; Step 5: Calculate the EAT entropy and EAT volume of epicardial fat based on the three-dimensional epicardial fat.

7. The method for predicting HFpEF in patients with myocardial infarction based on epicardial fat volume according to claim 1, characterized in that: The method for constructing the classifier comprises: Acquire EAT volume, EAT entropy and labels of myocardial infarction patients in a training set, wherein the labels are HFpEF and non-HFpEF; The EAT volume and EAT entropy are input into the machine learning model to output a predicted label, and the predicted label is compared with the label, and then the machine learning model is optimized and iterated to the stopping condition to obtain the classifier.

8. A computer device, characterized in that: The device comprises: a memory and a processor; the memory is used to store a computer program; the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method described in any one of claims 1 to 7 are implemented.

Images