CN-121987177-A - Noninvasive heart displacement estimation method and noninvasive heart displacement estimation device

CN121987177ACN 121987177 ACN121987177 ACN 121987177ACN-121987177-A

Abstract

The application provides a noninvasive cardiac output estimating method and device, the method comprises the following steps of obtaining a first physiological parameter signal and a second physiological parameter of a target user, wherein the first physiological parameter signal comprises an electrocardiogram signal and a cardiogram signal, and the second physiological parameter comprises a body mass index and blood oxygen saturation; the first data matrix and the second data matrix are input into a target heart displacement estimation model, and the target heart displacement of a target user is output. The application acquires a plurality of physiological parameters which are strongly related to the heart displacement, and uses the pre-estimation model to pre-estimate the heart displacement corresponding to the physiological parameters, thereby realizing accurate, real-time and noninvasive heart displacement monitoring.

Inventors

LI ZHONGKAI
YANG YUZHUO
YU SHUNZHOU

Assignees

核心医疗科技(香港)有限公司

Dates

Publication Date: 20260508
Application Date: 20260108

Claims (10)

1. A method of noninvasive cardiac displacement estimation for use with a ventricular assist device, the method comprising: Acquiring a first physiological parameter signal and a second physiological parameter of a target user, wherein the first physiological parameter signal comprises an electrocardiogram signal and a cardiogram signal, and the second physiological parameter comprises a body mass index and blood oxygen saturation; preprocessing the first physiological parameter signal and the second physiological parameter signal respectively to obtain a first data matrix and a second data matrix; And inputting the first data matrix and the second data matrix into a target heart displacement estimation model, and outputting the target heart displacement of the target user.
2. The method of claim 1, wherein the preprocessing the first physiological parameter signal and the second physiological parameter signal to obtain a first data matrix and a second data matrix, respectively, comprises: respectively filtering the first physiological parameter signals to obtain first signals of a target frequency band; Segmenting the first signal, wherein each segment of the first signal is provided with p sampling points, and p is a positive integer; forming n segments of the first signals into the first data matrix; And forming n second physiological parameters into the second data matrix.
3. The method of claim 2, wherein the target cardiac output estimation model comprises a first feature extraction module, a second feature extraction module and an estimation module, wherein the first feature extraction module consists of four convolutional layers, three active layers and one normalization layer, the second feature extraction module consists of two linear layers and one active layer, and the estimation module consists of two linear layers and one active layer; Inputting the first data matrix and the second data matrix into a target heart displacement estimation model to obtain the target heart displacement of the target user, wherein the method comprises the following steps: inputting the first data matrix into the first feature extraction module to obtain a first feature vector; Inputting the second data matrix into the second feature extraction module to obtain a second feature vector; splicing the first characteristic vector and the second characteristic vector to obtain a target characteristic vector; And inputting the target feature vector into the pre-estimating module and outputting the target heart displacement.
4. The method of claim 3, wherein the training method of the target cardiac output estimation model comprises: Acquiring r first training data sets, wherein each first training data set comprises the first physiological parameter signals and the second physiological parameters of a plurality of users, and r is a positive integer; Determining a first verification data set, wherein the first verification data set is an ith first training data set in the r first training data sets, and i is a positive integer less than or equal to r; Inputting the remaining r-1 first training data sets except the first verification data set into a heart displacement estimation model to be trained to train to obtain the target heart displacement estimation model, and determining an ith weight matrix of the target heart displacement estimation model according to the first verification data set; and (3) repeating the steps until i=i+1, and obtaining r weight matrixes of the target heart displacement estimation model.
5. The method of claim 4, wherein inputting the target feature vector into the predictive module and outputting the target heart displacement comprises: Inputting the target feature vector into an estimation module of an ith weight matrix to obtain an ith core displacement; let i=i+1, repeat the above steps until i=r; and determining the average value of the r ith heart displacement as the target heart displacement.
6. The method of claim 4, wherein inputting the remaining r-1 first training data sets except the first verification data set into a heart displacement estimation model to be trained to obtain the target heart displacement estimation model, determining an i-th weight matrix of the target heart displacement estimation model according to the first verification data set, and comprising: Preprocessing the r-1 first training data sets to obtain r-1 first data matrixes and r-1 second data matrixes; the r-1 first data matrixes and the r-1 second data matrixes are input into a kth heart displacement estimation model for training after the phases are randomly moved, and a weight matrix of the kth+1 heart displacement estimation model is obtained; Inputting the first verification data set into the k+1st heart displacement estimation model, outputting estimated heart displacement, and calculating root mean square error between the estimated heart displacement and actual heart displacement; Let k=k+1, repeat the above steps until k=q, where q is a positive integer greater than n; And determining a weight matrix of the k+1st heart displacement estimation model with the minimum root mean square error as the ith weight matrix.
7. The method according to claim 1, wherein the method further comprises: Acquiring a plurality of cardiac parameters of the target user, the plurality of cardiac parameters including heart rate, left ventricular end-systole volume, left ventricular end-diastole volume, right ventricular end-systole volume and right ventricular end-diastole volume; inputting the plurality of heart parameters into a target atrial pressure prediction model to obtain a left atrial pressure maximum value and a left atrial pressure minimum value; A pulmonary capillary wedge pressure is calculated based on the left atrial pressure maximum and the left atrial pressure minimum.
8. The method of claim 7, wherein the training method of the target atrial pressure predictive model comprises: Obtaining m second training data sets, wherein each second training data set comprises a plurality of heart parameters of a plurality of users, and m is a positive integer; Determining a second verification data set, wherein the second verification data set is a j second training data set in the m second training data sets, and j is a positive integer less than or equal to m; inputting the remaining r-1 second training data sets except the second verification data set into an atrial pressure pre-estimated model to be trained to train to obtain the target atrial pressure pre-estimated model, and determining a j weight matrix of the target atrial pressure pre-estimated model according to the second verification data set; And (3) repeating the steps until j=j+1, and obtaining m weight matrixes of the target atrial pressure estimation model.
9. A non-invasive cardiac displacement estimation apparatus, comprising one or more processors configured to perform the steps of: Acquiring a first physiological parameter signal and a second physiological parameter of a target user, wherein the first physiological parameter signal comprises an electrocardiogram signal and a cardiogram signal, and the second physiological parameter comprises a body mass index and blood oxygen saturation; preprocessing the first physiological parameter signal and the second physiological parameter signal respectively to obtain a first data matrix and a second data matrix; And inputting the first data matrix and the second data matrix into a target heart displacement estimation model, and outputting the target heart displacement of the target user.
10. A medical device comprising a processor, a memory and a communication interface, the memory storing one or more programs, and the one or more programs being executed by the processor, the one or more programs comprising instructions for performing the steps in the method of any of claims 1-8.

Description

Noninvasive heart displacement estimation method and noninvasive heart displacement estimation device Technical Field The application relates to the technical field of medical equipment, in particular to a noninvasive heart displacement estimation method and device. Background Ventricular assist devices are devices that provide support or assist functions for patients suffering from heart related diseases, such as heart failure, to assist the heart in pumping blood to other parts of the body. Cardiac Output (CO), which refers to the amount of blood ejected per minute from the left ventricle, is the most important parameter that characterizes the health of the cardiovascular system, and can reflect the current amount of blood perfused by the ventricular assist device to the user. For users implanted ventricular assist devices, cardiac output is used as an overall measure of ventricular assist device performance and cardiac pumping performance. Accordingly, there is a need for a cardiac output estimation scheme that accurately estimates cardiac output of the heart with the support of ventricular assist devices. Disclosure of Invention The embodiment of the application provides a noninvasive cardiac output estimating method and device, which can noninvasively and accurately estimate cardiac output of a heart under the support of a ventricular assist device. In a first aspect, an embodiment of the present application provides a method for non-invasive cardiac output estimation, applied to a ventricular assist device, the method including: Acquiring a first physiological parameter signal and a second physiological parameter of a target user, wherein the first physiological parameter signal comprises an electrocardiogram signal and a cardiogram signal, and the second physiological parameter comprises a body mass index and blood oxygen saturation; preprocessing the first physiological parameter signal and the second physiological parameter signal respectively to obtain a first data matrix and a second data matrix; And inputting the first data matrix and the second data matrix into a target heart displacement estimation model, and outputting the target heart displacement of the target user. In a second aspect, an embodiment of the present application provides a non-invasive cardiac output estimation device, including one or more processors, where the one or more processors are configured to perform the following steps: Acquiring a first physiological parameter signal and a second physiological parameter of a target user, wherein the first physiological parameter signal comprises an electrocardiogram signal and a cardiogram signal, and the second physiological parameter comprises a body mass index and blood oxygen saturation; preprocessing the first physiological parameter signal and the second physiological parameter signal respectively to obtain a first data matrix and a second data matrix; And inputting the first data matrix and the second data matrix into a target heart displacement estimation model, and outputting the target heart displacement of the target user. In a third aspect, embodiments of the present application provide a medical device comprising a processor, a memory, a communication interface, and one or more programs stored in the memory and configured to be executed by the processor, the programs comprising instructions for performing part or all of the steps described in the method of the first aspect above. The technical scheme includes that a first physiological parameter signal and a second physiological parameter of a target user are obtained, the first physiological parameter signal comprises an electrocardiogram signal and a heart shock map signal, the second physiological parameter comprises a body mass index and blood oxygen saturation, the first physiological parameter signal and the second physiological parameter are preprocessed to obtain a first data matrix and a second data matrix respectively, the first data matrix and the second data matrix are input into a target heart displacement estimation model, and target heart displacement of the target user is output. The application acquires a plurality of physiological parameters which are strongly related to the heart displacement, and uses the pre-estimation model to pre-estimate the heart displacement corresponding to the physiological parameters, thereby realizing accurate, real-time and noninvasive heart displacement monitoring. Drawings In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. FIG. 1 is a schematic illustration of a ventricular assist device according to an e