Heart failure prognosis evaluation instrument and evaluation method

By combining Doppler ultrasound and diffusion-related spectroscopy modules with a non-invasive detection technique and a machine learning model, changes in intrathoracic temperature are measured, solving the problems of accuracy and invasiveness in heart failure prognosis assessment and enabling accurate assessment and early warning of heart failure prognosis.

CN118948332AInactive Publication Date: 2024-11-15FUXING HOSPITAL OF CAPITAL MEDICAL UNIV
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411022467.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-29
Publication Date
2024-11-15
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing prognostic assessment methods for heart failure are mostly invasive procedures, making them difficult to use outside of coronary care units and intensive care units. Furthermore, their accuracy and consistency are insufficient, making them unable to effectively predict the prognosis of heart failure patients.

Method used

This non-invasive detection technology, which combines a Doppler ultrasound module and a diffusion-related spectroscopy module, measures lung water volume, pleural temperature, and blood flow. By incorporating a machine learning model, it calculates the increase in heat within the pleural cavity, enabling accurate assessment of heart failure prognosis.

Benefits of technology

It enables non-invasive and accurate assessment of heart failure prognosis, provides early warning of heart failure deterioration, improves patient outcomes, and reduces the risk of death.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118948332A_ABST
    Figure CN118948332A_ABST
Patent Text Reader

Abstract

The invention discloses a heart failure prognosis evaluation instrument and an evaluation method, and relates to the field of heart failure prognosis evaluation, a Doppler ultrasound module in the heart failure prognosis evaluation instrument is used for sending an ultrasonic signal to the thoracic cavity part of a target human body, and determining the pulmonary water volume of the target human body according to an echo signal; the diffusion correlation spectrum module is used for sending near-infrared laser to the lung area of the target human body and determining the blood flow volume of the lung area according to the detected scattered light spots of the lung area and a light intensity self-correlation function; the central automatic control device is used for determining the temperature in the thoracic cavity according to the blood flow of the lung area; calculating heat increased in the target human chest according to the temperature in the target human chest, the surface layer temperature and the pulmonary water volume; the pulmonary water volume of the target human body, the blood flow volume of the lung area, the temperature in the thoracic cavity, the surface layer temperature and the heat increased in the thoracic cavity are input into the heart failure prognosis model, a heart failure prognosis evaluation result is obtained, and accurate evaluation of heart failure prognosis can be achieved in a non-invasive mode.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of heart failure prognosis assessment, and in particular to a heart failure prognosis assessment instrument and assessment method. Background Art

[0002] At present, patients with worsening heart failure are usually examined by bedside right heart floating catheter (e.g., Swan-Ganz catheter) and dynamic monitoring by pulse indicator continuous cardiac output (PICCO) method. This examination and monitoring method can better determine the hemodynamic state and thus evaluate heart failure. However, the above methods are invasive operations and are usually only applicable to coronary care units (CCU) and intensive care units (ICU) with conditions. They are difficult to apply to the vast majority of other heart failure patients, and their role in prognostic evaluation is also relatively limited.

[0003] At present, in order to improve the accuracy of prognosis assessment of heart failure, a comprehensive assessment is usually carried out by combining clinical symptoms, signs, cardiac function classification, biochemical markers (e.g., BNP and NT-proBNP) detection, echocardiography, imaging examination, and cardiopulmonary exercise test, because the evaluation efficiency of using a single predictive factor is poor. For example, the left ventricular ejection fraction (LVEF), a commonly used clinical predictive indicator, is no longer reliable to assess the severity of heart failure due to the existence of a special type of heart failure called heart failure with preserved ejection fraction, that is, heart failure with normal ejection fraction, because people with normal ejection fraction may also be heart failure patients. Another example is BNP, another commonly used predictive indicator. Although it has important reference value in clinical practice, it is affected by many factors. This means that in some cases, the BNP level cannot fully and accurately reflect the severity and prognosis of heart failure. In fact, it is not uncommon for patients who have improved and been discharged from the hospital to have a higher BNP level than those who died in the hospital. However, although comprehensive assessment can provide more comprehensive information, in actual operation the accuracy and consistency of the assessment results may be affected due to factors such as problems with the connection between various examination methods, the complexity of data integration, and differences in doctors' interpretation capabilities. Summary of the invention

[0004] The purpose of this application is to provide a heart failure prognosis assessment instrument and assessment method to achieve accurate assessment of heart failure prognosis in a non-invasive manner.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides a heart failure prognosis assessment instrument, the heart failure prognosis assessment instrument comprising: a central automatic control device, a Doppler ultrasound module and a diffusion correlation spectrum module; the Doppler ultrasound module and the diffusion correlation spectrum module are both connected to the central automatic control device; the Doppler ultrasound module is connected to the diffusion correlation spectrum module;

[0007] The Doppler ultrasound module is used for:

[0008] Sending ultrasonic signals to the chest area of ​​the target human body and receiving echo signals;

[0009] generating an ultrasonic image according to the echo signal;

[0010] Determine the position and boundary of the lung in the chest cavity according to the ultrasonic image to obtain the lung region of the target human body;

[0011] Determine the lung water volume of the target person according to the lung area of ​​the target person;

[0012] The diffusion correlation spectroscopy module is used to:

[0013] Sending near-infrared laser light to the lung area of ​​the target person;

[0014] Detecting the response signal of the lung area to the near-infrared laser to obtain a scattered light spot;

[0015] Determine the blood flow in the target human lung area according to the scattered light spots and the light intensity autocorrelation function;

[0016] The central automatic control device is used for:

[0017] Determine the temperature in the chest cavity of the target human body according to the blood flow in the lung area of ​​the target human body and the correlation between the blood flow and the temperature;

[0018] Calculating the heat added in the chest cavity of the target human body according to the temperature in the chest cavity of the target human body, the surface temperature of the target human body and the lung water volume of the target human body;

[0019] The target human body's lung water volume, the target human body's lung area blood flow, the target human body's chest cavity temperature, the target human body's surface temperature and the target human body's chest cavity heat increase are input into the heart failure prognosis model to obtain the target human body's heart failure prognosis assessment result; the heart failure prognosis model is constructed based on a machine learning model.

[0020] In a second aspect, the present application provides a method for evaluating the prognosis of heart failure, the method comprising:

[0021] Sending ultrasonic signals to the chest area of ​​the target human body and receiving echo signals;

[0022] generating an ultrasonic image according to the echo signal;

[0023] Determine the position and boundary of the lung in the chest cavity according to the ultrasonic image to obtain the lung region of the target human body;

[0024] Determine the lung water volume of the target person according to the lung area of ​​the target person;

[0025] Sending near-infrared laser light to the lung area of ​​the target person;

[0026] Detecting the response signal of the lung area to the near-infrared laser to obtain a scattered light spot;

[0027] Determine the blood flow in the target human lung area according to the scattered light spots and the light intensity autocorrelation function;

[0028] Determine the temperature in the chest cavity of the target human body according to the blood flow in the lung area of ​​the target human body and the correlation between the blood flow and the temperature;

[0029] Calculating the heat added in the chest cavity of the target human body according to the temperature in the chest cavity of the target human body, the surface temperature of the target human body and the lung water volume of the target human body;

[0030] The target human body's lung water volume, the target human body's lung area blood flow, the target human body's chest cavity temperature, the target human body's surface temperature and the target human body's chest cavity heat increase are input into the heart failure prognosis model to obtain the target human body's heart failure prognosis assessment result; the heart failure prognosis model is constructed based on a machine learning model.

[0031] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0032] The present application provides a heart failure prognosis assessment instrument and assessment method, which measures the lung water volume of a target human body through a Doppler ultrasound module, and determines the blood flow in the lung area of ​​the target human body through a diffusion-related spectrum module; a central automatic control device determines the temperature in the chest cavity according to the blood flow in the lung area, calculates the increased heat in the chest cavity of the target human body according to the temperature in the chest cavity, the surface temperature and the lung water volume of the target human body, and uses a heart failure prognosis model to predict the heart failure prognosis assessment result. The present application assesses the prognosis of heart failure based on the lung water volume of the target human body, the blood flow in the lung area, the temperature in the chest cavity, the surface temperature, the increased heat in the chest cavity and the heart failure prognosis model, thereby achieving accurate assessment of the prognosis of heart failure in a non-invasive manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0034] Figure 1 A schematic diagram of the structure of a heart failure prognosis assessment instrument provided in one embodiment of the present application;

[0035] Figure 2 A flowchart of a method for evaluating the prognosis of heart failure provided in another embodiment of the present application.

[0036] Explanation of symbols: central automatic control device—1, Doppler ultrasound module—2, diffusion correlation spectroscopy module—3. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0038] This application aims to solve two core problems in the field of heart failure: poor prognosis of heart failure and inaccurate prognosis assessment of heart failure. These two problems are closely related and together affect the treatment effect and quality of life of heart failure patients.

[0039] Heart failure is the terminal stage of cardiovascular disease. Although there have been advances in treatment, the prognosis has not improved significantly, and the mortality rate remains high. According to epidemiological data, the 5-year mortality rate of patients with heart failure is as high as 50%, which is far higher than the mortality rate of cancers such as breast cancer and colorectal cancer. Faced with the current situation of prevention and treatment that is difficult to break through, new diagnosis and treatment strategies are urgently needed to break the dilemma. The 2023ESC-HFA Clinical Consensus Statement pointed out that preventing the worsening of heart failure can ultimately improve prognosis, but this requires early warning of the worsening of heart failure, and accurately assessing the severity and prognosis of heart failure is the key. Unfortunately, the current evaluation methods are not accurate, and new evaluation methods need to be explored urgently.

[0040] Clinically, patients with worsening heart failure often complain of feeling extremely hot in their chests, and they crave cold drinks or even ice cubes to relieve their symptoms. Such patients often have a poor prognosis and a very high risk of death in the short term. This unique clinical phenomenon suggests that there is a high correlation between the feeling of heat in the patient's chest and the prognosis of heart failure.

[0041] In order to fully explore the relationship between the patient's chest heat sensation and prognosis, we need to first deeply understand the pathophysiological process of heart failure. The core problem of heart failure is that the decline in cardiac pump function causes a decrease in forward blood flow and backward blood congestion. Among them, pulmonary congestion increases the amount of lung water that should be pumped to the epidermis by the heart for heat dissipation. This part of the heat is therefore not effectively dissipated, resulting in an increase in the patient's chest heat. The increased heat is not only related to the amount of lung water, but also closely related to the difference between the chest temperature and the surface temperature. In heart failure, the patient's chest temperature changes in the opposite direction to the surface temperature, which increases the difference between the chest temperature and the surface temperature, further increasing the patient's chest heat, and thus feeling a heat sensation in the chest. If the patient's chest heat changes can be monitored, the severity and prognosis of heart failure can be accurately judged.

[0042] Based on this, the present application provides a heart failure prognosis assessment instrument based on chest congestion heat detection technology, which non-invasively measures lung water volume through ultrasound and non-invasively measures the temperature in the chest through diffusion correlation spectroscopy technology, thereby monitoring the changes in chest heat in real time. By observing the characteristics and laws of heat changes, early warning of heart failure deterioration can be given and medical intervention can be carried out to avoid heart failure deterioration and ultimately improve prognosis.

[0043] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0044] See also Figure 1 The heart failure prognosis assessment instrument provided in the embodiment of the present application includes: a central automatic control device 1, a Doppler Ultrasound (DU) module 2 and a Diffuse Correlation Spectroscopy (DCS) module 3; the Doppler Ultrasound module 2 and the Diffuse Correlation Spectroscopy module 3 are both connected to the central automatic control device 1; the Doppler Ultrasound module 2 is connected to the Diffuse Correlation Spectroscopy module 3.

[0045] The Doppler ultrasound module 2 is used to: send an ultrasonic signal to the chest cavity of the target human body and receive an echo signal; generate an ultrasonic image according to the echo signal; determine the position and boundary of the lungs in the chest cavity according to the ultrasonic image to obtain the lung area of ​​the target human body; and determine the lung water volume of the target human body according to the lung area of ​​the target human body.

[0046] The diffusion correlation spectroscopy module 3 is used to: send near-infrared laser to the lung area of ​​the target human body; detect the response signal of the lung area to the near-infrared laser to obtain a scattered light spot; and determine the blood flow in the lung area of ​​the target human body according to the scattered light spot and the light intensity autocorrelation function.

[0047] The central automatic control device 1 is used to: determine the temperature in the chest cavity of the target human body according to the blood flow in the lung area of ​​the target human body and the correlation between the blood flow and the temperature; calculate the increased heat in the chest cavity of the target human body according to the temperature in the chest cavity of the target human body, the surface temperature of the target human body and the lung water volume of the target human body; input the lung water volume of the target human body, the blood flow in the lung area of ​​the target human body, the temperature in the chest cavity of the target human body, the surface temperature of the target human body and the increased heat in the chest cavity of the target human body into the heart failure prognosis model to obtain the heart failure prognosis evaluation result of the target human body; the heart failure prognosis model is constructed based on the machine learning model.

[0048] This embodiment uses ultrasound-guided near-infrared tissue blood flow detection technology to measure chest heat. Specifically, in view of the fact that the increase in chest heat is closely related to the progression of heart failure, Doppler ultrasound technology and diffusion correlation spectroscopy technology are organically combined, and the detection of chest tissue blood flow is achieved in combination with diffusion correlation spectroscopy under ultrasound guidance, thereby achieving multi-modal signal synchronous detection of lung water volume, chest temperature and blood flow, and then measuring the heat change in the chest cavity to obtain the increased heat in the chest cavity. The heart failure prognosis model is used to achieve heart failure prognosis assessment. This embodiment evaluates the prognosis of heart failure by measuring the increased heat in the chest cavity, thereby achieving accurate assessment of the prognosis of heart failure in a non-invasive manner.

[0049] In another exemplary embodiment of the present application, the Doppler ultrasound module 2 includes: an ultrasound probe and an image processing unit.

[0050] The ultrasonic probe is used to send ultrasonic signals to the chest area of ​​the target human body and receive echo signals. The image processing unit is used to generate an ultrasonic image according to the echo signals; determine the position and boundary of the lungs in the chest cavity according to the ultrasonic image to obtain the lung area of ​​the target human body; identify the echo information and blood flow information of the lung area; and determine the lung water volume of the target human body according to the echo information and the blood flow information.

[0051] The Doppler ultrasound module 2 in this embodiment is implemented based on the ultrasound processing technology of the Doppler effect, which can measure the speed and direction of moving objects (such as blood flow). For the determination of lung water volume, the Doppler ultrasound module 2 indirectly evaluates the lung water content by identifying and analyzing the ultrasound image of the lung tissue in the chest cavity. The Doppler ultrasound module 2 uses the penetration and reflectivity of ultrasound to accurately locate the chest cavity and lung tissue and identify the anatomical structure, thereby achieving accurate determination of lung water volume.

[0052] Specifically, based on the results of the positioning and identification of the chest cavity and lung tissue, the system can determine the position and boundary of the lung in the chest cavity. Then, the ultrasound image of the lung area is analyzed by the Doppler ultrasound module 2, and the degree of edema of the lung is determined by the echo information and blood flow information of the lung area, thereby indirectly evaluating the lung water volume.

[0053] When evaluating the amount of lung water, one possible way is as follows: by observing the echo density and distribution of the lung area, determine whether there are echo changes caused by fluid accumulation; at the same time, combined with the measurement of blood flow velocity and direction, further evaluate the state of pulmonary circulation, thereby indirectly inferring changes in lung water volume.

[0054] It should be noted that due to the complexity and individual differences of ultrasound images, the assessment of lung water volume is a relative process that requires comprehensive consideration of multiple factors and indicators. Therefore, in practical applications, the experience and expertise of clinicians can also be combined to carefully analyze and interpret ultrasound images to obtain more accurate lung water volume assessment results.

[0055] In another exemplary embodiment of the present application, the diffusion correlation spectroscopy module 3 includes: a near-infrared light source module, a detection module and an autocorrelation calculation module.

[0056] The near-infrared light source module is used to send near-infrared laser to the chest area of ​​the target human body; the detection module is used to detect the response signal of the lung area to the near-infrared laser to obtain a scattered light spot. Specifically, the near-infrared laser can be a long coherent laser in the near-infrared band.

[0057] The autocorrelation calculation module is used to: calculate the movement state of red blood cells in lung tissue according to the scattered light spots and the light intensity autocorrelation function; and determine the blood flow in the target human lung area according to the movement state.

[0058] The diffusion correlation spectroscopy module 3 of this embodiment utilizes near-infrared to irradiate the tissue surface, and infers the movement state of red blood cells in the tissue by calculating the light intensity autocorrelation function of the scattered light spot on the tissue surface, so as to realize quantitative detection of tissue blood flow changes. By measuring the blood flow of the surface layer of human skin and the deep tissue in the chest cavity, and establishing the relationship between tissue blood flow and temperature, non-invasive measurement of skin and chest cavity temperature can be realized.

[0059] In another exemplary embodiment of the present application, the central automatic control device 1 includes: a processor, a display terminal and a data interface; the Doppler ultrasound module 2 and the diffusion correlation spectroscopy module 3 are both connected to the processor via the data interface.

[0060] The processor is used to realize the fusion and analysis of multimodal data, and finally generate a heart failure prognosis evaluation result. The processor includes: a data storage module, a lung tissue temperature calculation module, a chest congestion heat determination module and a heart failure prognosis evaluation module.

[0061] The data storage module is used to store the target human lung water volume, the target human lung area blood flow and the correlation between blood flow and temperature. The correlation between blood flow and temperature can be obtained by using the chest diffusion correlation spectrum module 3 for different heart failure patients to obtain the blood flow of the corresponding lung area, using the existing temperature measurement method to obtain the chest temperature of the corresponding heart failure patient in advance, fitting the blood flow of the lung area of ​​different heart failure patients with the corresponding chest temperature, obtaining the correlation between blood flow and temperature, and storing it in the data storage module.

[0062] The lung tissue temperature calculation module is used to determine the temperature corresponding to the blood flow in the lung area of ​​the target human body according to the correlation between the blood flow and the temperature, and obtain the temperature in the chest cavity of the target human body.

[0063] The chest congestion heat determination module is used to: calculate the difference between the temperature in the chest cavity of the target human body and the surface temperature of the target human body to obtain the temperature difference; determine the current lung water volume increase according to the lung water volume of the target human body; and calculate the increased heat in the chest cavity of the target human body according to the temperature difference and the current lung water volume increase. The specific calculation formula is as follows: ΔQ=Δm×c×ΔT, where ΔQ represents the increased heat in the chest cavity of the target human body; Δm represents the current lung water volume increase, c represents the specific heat capacity, and ΔT represents the temperature difference.

[0064] The heart failure prognosis assessment module is used to input the increased heat in the chest cavity of the target human body into the heart failure prognosis model to obtain the heart failure prognosis assessment result of the target human body.

[0065] The display terminal is used to display the heart failure prognosis assessment result; the heart failure prognosis assessment result includes: the risk level of heart failure prognosis, the predicted survival period of heart failure prognosis and the suggestion of medical intervention based on the current prognosis status.

[0066] It can be seen from the above calculation formula for the increased heat in the chest cavity that the present application not only measures the lung water volume, but also measures the chest temperature and surface temperature at the same time, taking into account various clinical characteristics of patients with worsening heart failure, such as "dry", "wet", "cold", and "warm", which is significantly better than the lung water measuring instruments on the market. According to the above formula, the "wet and cold" type of patients with the worst prognosis are precisely those with high lung water volume and low skin temperature. This type of patient has the largest Q value, that is, the heart failure prognosis assessment instrument of the present application has a strong ability to identify patients with the worst prognosis.

[0067] In another exemplary embodiment of the present application, the processor also includes: a model training module.

[0068] The model training module is used to:

[0069] Acquire training data; the training data include: lung water volume of different heart failure patients, blood flow in lung areas of different heart failure patients, temperature in chest cavity of different heart failure patients, surface temperature of different heart failure patients, increased heat in chest cavity of different heart failure patients and corresponding real results of heart failure prognosis; the real results of heart failure prognosis include: real risk level of heart failure prognosis, real survival period of heart failure prognosis and real suggestions for medical intervention based on current prognosis status.

[0070] The training data is input into a machine learning model, and training is performed with the goal of minimizing the sum of weighted cross entropy loss and time series loss, and the trained machine learning model is determined as a heart failure prognosis model; the machine learning model is an improved convolutional neural network or an integrated learning classifier; the improved convolutional neural network is a convolutional neural network that introduces an attention mechanism.

[0071] In this embodiment, a customized convolutional neural network is used to more effectively capture the spatial and temporal patterns of chest congestion heat characteristics. This network structure combines a multi-scale feature extractor and can simultaneously process heat information of different scales. An attention mechanism is introduced so that the model can automatically focus on key features closely related to the prognosis of heart failure and improve the accuracy of prediction. In order to further improve the predictive ability of the model, an integrated learning classifier, such as a Bagging model and a Boosting model, can also be used to integrate multiple base classifiers (such as convolutional neural networks, support vector machines, etc.) to produce more stable and reliable prediction results. The traditional mean square error loss function may not be sufficient to capture the complexity and uncertainty of heart failure prognosis. This embodiment uses a weighted cross entropy loss function and combines it with a regularization term to better balance the loss of samples of different categories and prevent overfitting. At the same time, considering the temporal characteristics of heart failure prognosis, a time series loss is also introduced to capture dependencies and trends in time series data.

[0072] A specific implementation process of the above-mentioned heart failure prognosis assessment instrument in practical application is as follows:

[0073] Step 1: Enter patient information.

[0074] The heart failure prognosis assessment instrument first inputs the patient's basic information through an input device (such as a touch screen, keyboard or external data interface), including name, age, gender, height, weight, and medical history, etc. At the same time, it records the patient's current heart failure symptoms and related medical indicators, such as heart function classification, etc.

[0075] Step 2: Determination of chest congestion heat.

[0076] Method 1: The patient lies flat on the measuring bed of the heart failure prognosis assessment instrument, and the increased heat in the patient's chest cavity is obtained by using the central automatic control device 1, the Doppler ultrasound module 2, and the diffusion correlation spectroscopy module 3.

[0077] Method 2: The patient lies flat on the measurement bed of the heart failure prognosis assessment instrument, and the diffusion correlation spectroscopy module and the central automatic control device are used to determine the temperature in the patient's chest cavity. The temperature sensor array is used to contact the surface of the patient's chest cavity to collect the skin temperature as the patient's surface temperature. The central automatic control device preprocesses the temperature data, removes noise and interference signals, and extracts characteristic parameters reflecting the heat congestion in the chest cavity. The characteristic parameters include: the rate of change of chest cavity temperature, the average or difference of chest cavity temperature in a specific time period, and the heat parameters related to the change of lung water volume.

[0078] Step 3: Analysis of heart failure prognostic model.

[0079] The built-in heart failure prognosis model of the heart failure prognostic assessment instrument takes the patient's lung water volume, the blood flow in the patient's lung area, the temperature in the patient's chest cavity, the patient's surface temperature and the heat added in the patient's chest cavity as input, or takes the processed chest congestion heat characteristic parameters as input. The model is built based on big data analysis and machine learning algorithms, and its effectiveness has been pre-trained and verified. Based on the input parameters, the model outputs the patient's heart failure prognosis assessment results through complex calculations and analysis.

[0080] Step 4: Output and display of evaluation results.

[0081] The heart failure prognosis assessment instrument displays the assessment results to patients and doctors through output devices (such as display screens or printers). The assessment results include the risk level of heart failure prognosis, the predicted survival period of heart failure prognosis, and recommendations for medical intervention based on the current prognosis status. Doctors and patients can develop more accurate treatment and management plans based on this information.

[0082] Based on the same inventive concept, the embodiment of the present application also provides a heart failure prognosis assessment method for implementing the heart failure prognosis assessment instrument involved above. The implementation scheme for solving the problem provided by the method is similar to the implementation scheme recorded in the above heart failure prognosis assessment instrument, so the specific limitations in one or more heart failure prognosis assessment instrument method embodiments provided below can refer to the limitations on the heart failure prognosis assessment instrument above, and will not be repeated here.

[0083] In an exemplary embodiment, Figure 2 As shown, a method for evaluating the prognosis of heart failure is provided, including:

[0084] Step 101, determining the lung water volume of the target human body.

[0085] Specifically, an ultrasonic signal is sent to the chest cavity of a target human body, and an echo signal is received; an ultrasonic image is generated according to the echo signal; the position and boundary of the lungs in the chest cavity are determined according to the ultrasonic image to obtain the lung area of ​​the target human body; and the lung water volume of the target human body is determined according to the lung area of ​​the target human body.

[0086] Step 102, determining the temperature in the chest cavity of the target human body.

[0087] Specifically, a near-infrared laser is sent to the lung area of ​​the target human body; a response signal of the lung area to the near-infrared laser is detected to obtain a scattered light spot; the blood flow in the lung area of ​​the target human body is determined according to the scattered light spot and the light intensity autocorrelation function; and the temperature in the chest cavity of the target human body is determined according to the blood flow in the lung area of ​​the target human body and the correlation between the blood flow and the temperature.

[0088] Step 103, calculating the heat added in the chest cavity of the target human body according to the temperature in the chest cavity of the target human body, the surface temperature of the target human body and the lung water volume of the target human body.

[0089] Step 104, inputting the target human body's lung water volume, the target human body's lung area blood flow, the target human body's chest cavity temperature, the target human body's surface temperature and the target human body's chest cavity increased heat into the heart failure prognosis model to obtain the target human body's heart failure prognosis assessment result; the heart failure prognosis model is constructed based on a machine learning model.

[0090] As an optional implementation, in step 101, determining the lung water volume of the target human body according to the lung area of ​​the target human body specifically includes: identifying echo information and blood flow information of the lung area; and determining the lung water volume of the target human body according to the echo information and the blood flow information.

[0091] As an optional implementation, in step 102, the blood flow in the lung area of ​​the target human body is determined based on the scattered light spot and the light intensity autocorrelation function, specifically including: calculating the movement state of red blood cells in the lung tissue based on the scattered light spot and the light intensity autocorrelation function; and determining the blood flow in the lung area of ​​the target human body based on the movement state.

[0092] As an optional implementation, step 103 specifically includes: calculating the difference between the temperature in the chest cavity of the target human body and the surface temperature of the target human body to obtain the temperature difference; determining the current increase in lung water volume according to the lung water volume of the target human body; and calculating the increased heat in the chest cavity of the target human body according to the temperature difference and the current increase in lung water volume.

[0093] As an optional implementation, the method for determining the heart failure prognosis model specifically includes:

[0094] Acquire training data; the training data include: lung water volume of different heart failure patients, blood flow in lung areas of different heart failure patients, temperature in chest cavity of different heart failure patients, surface temperature of different heart failure patients, increased heat in chest cavity of different heart failure patients and corresponding real results of heart failure prognosis; the real results of heart failure prognosis include: real risk level of heart failure prognosis, real survival period of heart failure prognosis and real suggestions for medical intervention based on current prognosis status.

[0095] The training data is input into a machine learning model, and training is performed with the goal of minimizing the sum of weighted cross entropy loss and time series loss, and the trained machine learning model is determined as a heart failure prognosis model; the machine learning model is an improved convolutional neural network or an integrated learning classifier; the improved convolutional neural network is a convolutional neural network that introduces an attention mechanism.

[0096] In this embodiment, the measurement of the increased heat in the chest cavity is a core step, which is directly related to the accuracy of the assessment of the prognosis of heart failure. Through Doppler ultrasound technology and diffusion correlation spectroscopy technology, the stagnant heat information reflecting the blood circulation state and metabolic conditions in the chest cavity of heart failure patients is obtained; the heart failure prognosis model analysis in this embodiment is another key step, which uses advanced machine learning algorithms to conduct in-depth analysis of chest stagnant heat information, thereby achieving accurate prediction of the prognosis of heart failure. Among them, the training and optimization of the model are the key to ensuring the accuracy and reliability of the assessment results.

[0097] This application proposes that measuring chest congestion heat can predict the prognosis of heart failure patients. This innovative theory has a solid theoretical and practical basis.

[0098] In theory, lung water volume and temperature difference are closely related to the severity and prognosis of heart failure. The evaluation index ΔQ of this embodiment is calculated based on the two key indicators ΔT and lung water volume. Compared with a single ΔT or lung water volume index, ΔQ can more comprehensively reflect the pathophysiological state of heart failure patients. Therefore, the index ΔQ can provide a more accurate and reliable method for the prognosis assessment of heart failure patients.

[0099] At the practical level, this application is closely integrated with clinical practice, starting from the actual symptoms of patients, and deeply exploring the nature of heart failure. Patients with heart failure often have a dynamic imbalance between the body's overall heat production and heat dissipation, especially the phenomenon of heat stagnation in the chest cavity. Therefore, this application starts with the effect of heat stagnation in the chest cavity, accurately measures lung water volume and ΔT through Doppler ultrasound and near-infrared tissue blood flow detection technology, and calculates a new evaluation index ΔQ, thereby achieving an accurate assessment of the prognosis of heart failure.

[0100] The logical chain of this application is clear. The decline in cardiac pump function leads to pulmonary congestion and increased lung water. Under normal circumstances, this lung water should participate in the heat dissipation process on the skin surface, but pulmonary congestion inhibits this heat dissipation effect, causing heat to accumulate in the chest cavity. On the other hand, the reverse change between the chest temperature and the surface temperature of patients with heart failure provides another important observation parameter. Specifically, the decrease in surface temperature is related to factors such as sympathetic nerve excitement, vasodilatory dysfunction, and decreased cardiac output. These factors work together to cause skin vasoconstriction and reduced blood flow; while the increase in chest temperature may be caused by a variety of factors, including reduced heat dissipation, increased sympathetic nerve excitement, increased heat production caused by increased respiratory muscle exercise, and possible edema insulation. This reverse change in chest temperature and surface temperature further leads to an increase in ΔT, exacerbating the accumulation of heat in the patient's chest. The heat in the chest felt by the patient is not only a reflection of pathophysiological changes, but also an indicator of the severity of heart failure.

[0101] Therefore, measuring the increased heat ΔQ in the chest cavity has a high clinical value. The larger the ΔQ value, the more severe the patient's heart failure and the worse the prognosis. This application regards ΔQ as the "death heat" of heart failure patients. It is not just a numerical value, but also a quantitative expression of the severity and prognosis of heart failure. By monitoring the changes in the ΔQ value, not only can the condition of patients in the worsening stage of heart failure be accurately assessed, but also signs of worsening heart failure can be detected in advance, which provides the possibility for early warning and intervention, thereby avoiding the worsening of heart failure and ultimately improving patient prognosis.

[0102] This application proposes ultrasound-guided near-infrared tissue blood flow detection technology for the first time, and realizes the organic combination of DU technology and DCS technology for the first time, and is used to realize the synchronous detection of multimodal signals of lung water volume, intrathoracic temperature and blood flow. It also independently develops heart failure prognosis assessment equipment based on the proposed technology, which essentially solves the pain points of existing heart failure prognosis assessment and assesses the condition of heart failure patients by quantitatively analyzing tissue blood flow and temperature changes.

[0103] This application integrates and analyzes multimodal data, successfully realizing the different collection of multimodal data such as ultrasound images, blood flow data and temperature data, and combines deep learning algorithms such as multidimensional attention mechanism to extract effective features that can reflect the essence of heart failure, and uses effective information to improve the accuracy and reliability of heart failure prognosis assessment.

[0104] This application fills the technical gap in the field of heart failure prognosis assessment. Compared with existing heart failure prognosis assessment instruments and assessment methods, this application has higher accuracy and reliability and can provide more accurate early warning for heart failure deterioration.

[0105] In addition, there are some alternatives to the above-mentioned heart failure prognosis assessment instrument and method that can also achieve the purpose of this application, that is, to accurately assess the prognosis of heart failure patients. The following are several possible alternatives:

[0106] (1) Heart failure prognosis assessment based on multimodal biological signals.

[0107] In addition to chest congestion heat, other biological signals of heart failure patients (such as electrocardiogram, blood pressure, respiratory rate, etc.) also contain rich prognostic information. Therefore, a heart failure prognosis assessment system based on multimodal biological signals can be designed. The system comprehensively judges the patient's heart failure condition and prognosis by simultaneously collecting and analyzing multiple biological signals.

[0108] (2) Heart failure prognosis assessment based on medical imaging.

[0109] Medical images (such as echocardiogram, X-ray, MRI, etc.) can intuitively display the structural and functional changes of the heart and are an important means of diagnosing and evaluating the prognosis of heart failure. By developing advanced image processing and analysis algorithms, key information related to the prognosis of heart failure can be extracted from medical images, thereby achieving accurate prognostic evaluation.

[0110] (3) Heart failure prognosis assessment based on blood biochemical indicators.

[0111] The blood biochemical indicators of patients with heart failure (such as creatinine, urea nitrogen, electrolytes, etc.) can also reflect the severity and progression of heart failure. Therefore, the prognosis of heart failure can be evaluated by regularly testing the patient's blood biochemical indicators and combining them with other clinical information.

[0112] (4) Comprehensive evaluation based on artificial intelligence and big data.

[0113] With the continuous development of artificial intelligence and big data technology, these advanced technologies can be used to build a comprehensive evaluation model to achieve accurate prediction of heart failure prognosis. The model can integrate multi-source data (including biological signals, medical images, blood biochemical indicators, etc.), extract key features through deep learning and data mining algorithms, and train and optimize based on a large amount of historical data to improve the accuracy and reliability of the evaluation.

[0114] It should be noted that although these alternative options can realize the assessment of heart failure prognosis, in actual work, the accuracy and consistency of the assessment results are affected by factors such as the connection problems between various examination methods, the complexity of data integration, and differences in doctors' interpretation capabilities. Most importantly, the above-mentioned alternative options are difficult to achieve the requirements of real-time non-invasive monitoring, and therefore have little effect in providing early warning of heart failure deterioration and conducting medical intervention, avoiding heart failure deterioration, and ultimately improving prognosis.

[0115] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0116] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A heart failure prognosis assessment instrument, characterized in that: The heart failure prognosis assessment instrument comprises: a central automatic control device, a Doppler ultrasound module and a diffusion correlation spectrum module; the Doppler ultrasound module and the diffusion correlation spectrum module are both connected to the central automatic control device; the Doppler ultrasound module is connected to the diffusion correlation spectrum module; The Doppler ultrasound module is used for: Sending ultrasonic signals to the chest area of ​​the target human body and receiving echo signals; generating an ultrasonic image according to the echo signal; Determine the position and boundary of the lung in the chest cavity according to the ultrasonic image to obtain the lung region of the target human body; Determine the lung water volume of the target person according to the lung area of ​​the target person; The diffusion correlation spectroscopy module is used to: Sending near-infrared laser light to the lung area of ​​the target person; Detecting the response signal of the lung area to the near-infrared laser to obtain a scattered light spot; Determine the blood flow in the target human lung area according to the scattered light spots and the light intensity autocorrelation function; The central automatic control device is used for: Determine the temperature in the chest cavity of the target human body according to the blood flow in the lung area of ​​the target human body and the correlation between the blood flow and the temperature; Calculating the heat added in the chest cavity of the target human body according to the temperature in the chest cavity of the target human body, the surface temperature of the target human body and the lung water volume of the target human body; The target human body's lung water volume, the target human body's lung area blood flow, the target human body's chest cavity temperature, the target human body's surface temperature and the target human body's chest cavity heat increase are input into the heart failure prognosis model to obtain the target human body's heart failure prognosis assessment result; the heart failure prognosis model is constructed based on a machine learning model.

2. The heart failure prognosis assessment instrument according to claim 1, characterized in that: The Doppler ultrasound module includes: an ultrasound probe and an image processing unit; The ultrasonic probe is used for: Sending ultrasonic signals to the chest area of ​​the target human body and receiving echo signals; The image processing unit is used for: generating an ultrasonic image according to the echo signal; Determine the position and boundary of the lung in the chest cavity according to the ultrasonic image to obtain the lung region of the target human body; Identify the echogenicity and blood flow information of lung regions; The lung water volume of the target human body is determined according to the echo information and the blood flow information.

3. The heart failure prognosis assessment instrument according to claim 1, characterized in that: The diffusion correlation spectrum module includes: a near-infrared light source module, a detection module and an autocorrelation calculation module; The near-infrared light source module is used for: Sending near-infrared laser light to the chest area of ​​the target person; The detection module is used for: Detecting the response signal of the lung area to the near-infrared laser to obtain a scattered light spot; The autocorrelation calculation module is used for: Calculating the movement state of red blood cells in lung tissue according to the scattered light spots and the light intensity autocorrelation function; The blood flow of the target human lung area is determined according to the motion state.

4. The heart failure prognosis assessment instrument according to claim 1, characterized in that: The central automatic control device comprises: a processor, a display terminal and a data interface; the Doppler ultrasound module and the diffusion correlation spectrum module are both connected to the processor via the data interface; The processor includes: a data storage module, a lung tissue temperature calculation module, a chest congestion heat determination module and a heart failure prognosis assessment module; The data storage module is used for: storing the lung water volume of the target human body, the blood flow in the lung area of ​​the target human body, and the correlation between the blood flow and the temperature; The lung tissue temperature calculation module is used for: According to the correlation between blood flow and temperature, the temperature corresponding to the blood flow in the lung area of ​​the target human body is determined to obtain the temperature in the chest cavity of the target human body; The chest congestion heat determination module is used for: Calculate the difference between the temperature in the chest cavity of the target human body and the temperature on the surface of the target human body to obtain the temperature difference; Determine the current lung water volume increase amount according to the lung water volume of the target human body; Calculating the heat added in the chest cavity of the target human body according to the temperature difference and the current increase in lung water volume; The heart failure prognosis assessment module is used to input the increased heat in the chest cavity of the target human body into the heart failure prognosis model to obtain the heart failure prognosis assessment result of the target human body; The display terminal is used to display the heart failure prognosis assessment result; the heart failure prognosis assessment result includes: the risk level of heart failure prognosis, the predicted survival period of heart failure prognosis and the suggestion of medical intervention based on the current prognosis status.

5. The heart failure prognosis assessment instrument according to claim 4, characterized in that: The processor also includes: a model training module; The model training module is used to: Acquire training data; the training data includes: lung water volume of different heart failure patients, blood flow of different lung regions of heart failure patients, chest temperature of different heart failure patients, surface temperature of different heart failure patients, increased heat in the chest of different heart failure patients and corresponding real results of heart failure prognosis; the real results of heart failure prognosis include: real risk level of heart failure prognosis, real survival period of heart failure prognosis and real suggestions for medical intervention based on current prognosis status; The training data is input into a machine learning model, and training is performed with the goal of minimizing the sum of weighted cross entropy loss and time series loss, and the trained machine learning model is determined as a heart failure prognosis model; the machine learning model is an improved convolutional neural network or an integrated learning classifier; the improved convolutional neural network is a convolutional neural network that introduces an attention mechanism.

6. A method for evaluating the prognosis of heart failure, characterized in that: The heart failure prognosis assessment method comprises: Sending ultrasonic signals to the chest area of ​​the target human body and receiving echo signals; generating an ultrasonic image according to the echo signal; Determine the position and boundary of the lung in the chest cavity according to the ultrasonic image to obtain the lung region of the target human body; Determine the lung water volume of the target person according to the lung area of ​​the target person; Sending near-infrared laser light to the lung area of ​​the target person; Detecting the response signal of the lung area to the near-infrared laser to obtain a scattered light spot; Determine the blood flow in the target human lung area according to the scattered light spots and the light intensity autocorrelation function; Determine the temperature in the chest cavity of the target human body according to the blood flow in the lung area of ​​the target human body and the correlation between the blood flow and the temperature; Calculating the heat added in the chest cavity of the target human body according to the temperature in the chest cavity of the target human body, the surface temperature of the target human body and the lung water volume of the target human body; The target human body's lung water volume, the target human body's lung area blood flow, the target human body's chest cavity temperature, the target human body's surface temperature and the target human body's chest cavity heat increase are input into the heart failure prognosis model to obtain the target human body's heart failure prognosis assessment result; the heart failure prognosis model is constructed based on a machine learning model.

7. The heart failure prognosis assessment method according to claim 6, characterized in that: The target person's lung water volume is determined based on the target person's lung area, including: Identify the echogenicity and blood flow information of lung regions; The lung water volume of the target human body is determined according to the echo information and the blood flow information.

8. The heart failure prognosis assessment method according to claim 6, characterized in that: Determining the blood flow rate in the target human lung area according to the scattered light spot and the light intensity autocorrelation function specifically includes: Calculating the movement state of red blood cells in lung tissue according to the scattered light spots and the light intensity autocorrelation function; The blood flow of the target human lung area is determined according to the motion state.

9. The heart failure prognosis assessment method according to claim 6, characterized in that: The heat added in the chest cavity of the target human body is calculated according to the temperature in the chest cavity of the target human body, the surface temperature of the target human body and the lung water volume of the target human body, specifically including: Calculate the difference between the temperature in the chest cavity of the target human body and the temperature on the surface of the target human body to obtain the temperature difference; Determine the current lung water volume increase amount according to the lung water volume of the target human body; The increased heat in the chest cavity of the target human body is calculated based on the temperature difference and the current increase in lung water volume.

10. The heart failure prognosis assessment method according to claim 6, characterized in that: The method for determining the heart failure prognosis model specifically includes: Acquire training data; the training data includes: lung water volume of different heart failure patients, blood flow of different lung regions of heart failure patients, chest temperature of different heart failure patients, surface temperature of different heart failure patients, increased heat in the chest of different heart failure patients and corresponding real results of heart failure prognosis; the real results of heart failure prognosis include: real risk level of heart failure prognosis, real survival period of heart failure prognosis and real suggestions for medical intervention based on current prognosis status; The training data is input into a machine learning model, and training is performed with the goal of minimizing the sum of weighted cross entropy loss and time series loss, and the trained machine learning model is determined as a heart failure prognosis model; the machine learning model is an improved convolutional neural network or an integrated learning classifier; the improved convolutional neural network is a convolutional neural network that introduces an attention mechanism.

Citation Information

Cited By

  • Paster type cardiac function evaluation device based on ultrasonic and DCS fusion probe

    CN224655343U