Hemodialysis patient vascular health assessment and arteriovenous fistula intelligent nursing system

Abstract

The present application relates to the technical field of blood monitoring, in particular to a vascular health assessment and intelligent nursing system for hemodialysis patients. The degree of venous reflux obstruction is determined based on the change trend and degree of venous pressure data within each dialysis period, and the degree of inconsistency between the change trends of venous pressure data and transmembrane pressure data. The degree of arterial supply obstruction is determined based on the change trend and degree of arterial pressure data within each dialysis period. The suspected internal fistula obstruction period is determined based on the change trend of the degree of venous reflux obstruction and the degree of arterial supply obstruction. The nursing intervention prompting degree of the patient to be tested is determined based on the relationship between the blood flow data and arterial pressure data of each dialysis stage of the patient to be tested within the suspected internal fistula obstruction period, and the blood flow data of the non-dialysis stage. The internal fistula nursing scheme is determined. The present application realizes early warning of internal fistula dysfunction based on the change of venous and arterial pressure of the blood circulation loop during dialysis.

Description

Technical Field

[0001] This invention relates to the field of blood monitoring technology, specifically to an intelligent nursing system for assessing vascular health and arteriovenous fistulas in hemodialysis patients. Background Technology

[0002] Arteriovenous fistulas (AVFs) are common vascular access points used in hemodialysis. They are created by surgically anastomosing an artery to a nearby vein, forming a blood flow pathway to provide a stable blood flow for hemodialysis. During the use of an AVF, hemodynamic changes may occur due to factors such as intimal hyperplasia, thrombosis, or vascular wall sclerosis. This manifests as a gradual decrease in blood flow, which may progress to fistula stenosis or thrombosis, leading to fistula dysfunction. Systematic and early health assessments of AVFs, including vascular ultrasound, blood flow monitoring, and observation of clinical symptoms, help to identify risks such as fistula stenosis in a timely manner, guiding healthcare professionals to intervene promptly. This effectively prevents further deterioration of fistula function, prolongs its lifespan, reduces the burden of secondary surgeries, and improves the quality of dialysis.

[0003] Existing technology, patent document CN120304871A, discloses an arteriovenous fistula (AVF) detection system for hemodialysis. Specifically, it determines the stenosis status of the AVF by judging whether the blood flow velocity and stenosis risk characterization parameter of the AVF before dialysis are successively greater than a preset stenosis risk characterization parameter and a preset blood flow velocity. The stenosis risk characterization parameter is the ratio of the standard deviation of the diameter of several sampling segments along the AVF centerline to the average diameter of the same sampling segments. Existing methods examine vascular structural parameters and a single blood flow velocity obtained through imaging and other means before dialysis, lacking the ability to monitor real-time, continuous changes in AVF function during dialysis treatment. Furthermore, early stenosis hemodynamic abnormalities gradually appear as the workload increases during dialysis treatment, and existing methods struggle to capture early functional impairment signals during treatment, leading to delayed vascular health warnings. Summary of the Invention

[0004] To address the technical problem that static structural parameter detection before dialysis cannot capture the dynamic situation of early arteriovenous fistula stenosis during dialysis, leading to delayed vascular health early warning, the present invention aims to provide an intelligent nursing system for vascular health assessment and arteriovenous fistula in hemodialysis patients. The specific technical solution adopted is as follows: This invention proposes an intelligent nursing system for vascular health assessment and arteriovenous fistula in hemodialysis patients, the system comprising: The data acquisition module is used to record the patients to be tested and the test patients as the analysis patients, and to acquire the venous pressure data, arterial pressure data and transmembrane pressure data at each moment during each dialysis process of the analysis patients during the use of the arteriovenous fistula. The venous end impact analysis module is used to determine the degree of venous return obstruction in each dialysis session based on the trend and degree of change of venous pressure data within each dialysis session, as well as the degree of inconsistency between the trend of venous pressure data and transmembrane pressure data at each moment within each dialysis session. The arterial impact analysis module is used to determine the degree of arterial supply obstruction during each dialysis session based on the trend and degree of change in the patient's arterial pressure data during each dialysis session. The arteriovenous fistula (AVF) obstruction analysis module is used to determine the suspected AVF obstruction period during the AVF usage phase based on the changing trends of the venous return obstruction and the arterial supply obstruction. The fistula assessment and care module is used to determine the nursing intervention alert level for the patient based on the relationship between blood flow and arterial pressure data during each dialysis stage in the suspected fistula failure period, as well as blood flow data during non-dialysis stages, using a trained neural network, and to determine the fistula care plan.

[0005] Furthermore, determining the degree of venous return obstruction during each dialysis session includes: Based on the trend and degree of change of venous pressure data during each dialysis session, the initial degree of venous return obstruction during each dialysis session is determined. The venous pressure and transmembrane pressure data at all times within each dialysis session are arranged in chronological order to obtain the venous pressure sequence and transmembrane pressure sequence, respectively. The correlation coefficient between the venous blood pressure sequence and the transmembrane pressure sequence in the same dialysis session is negatively correlated and normalized to obtain the blocked value of each dialysis session. The initial degree of venous return obstruction is weighted based on the obstruction value to obtain the degree of venous return obstruction for each dialysis session.

[0006] Furthermore, determining the initial venous return resistance during each dialysis session includes: Obtain the first-order difference sequence of the venous pressure sequence, and record the ratio of the number of elements with positive values ​​to the total number of elements in the first-order difference sequence as the trend indicator for each dialysis period. Calculate the mean of all elements in the first-order difference sequence, and multiply the mean by the trend indicator to obtain the strength of the upward trend for each dialysis process. Calculate the difference between the venous pressure data at the end and start times of each dialysis session, and use this as the venous pressure start-end difference; Based on the strength of the upward trend and the difference between the beginning and end of the venous pressure, the initial degree of venous return obstruction during each dialysis period is obtained.

[0007] Furthermore, determining the degree of arterial supply obstruction during each dialysis session includes: The venous pressure data obtained during the acquisition of the initial venous return obstruction is updated using the absolute value of the arterial pressure data at each moment within each dialysis session. The recalculated initial venous return obstruction is then used as the arterial supply obstruction for each dialysis session.

[0008] Furthermore, the determination of suspected arteriovenous fistula dysfunction periods during the arteriovenous fistula use phase includes: The degree of venous return obstruction and the degree of arterial supply obstruction during the same dialysis period are weighted and summed to obtain the degree of venous and arterial impact during each dialysis period; One dialysis process is randomly selected during the use of the arteriovenous fistula and recorded as an example process. The first set consists of the venous and arterial impact of the example process during the use of the arteriovenous fistula and all the dialysis processes before it. The second set consists of the venous and arterial impact of all the dialysis processes after the example process during the use of the arteriovenous fistula. Calculate the between-group variance between the first set and the second set, and denote it as the set difference significance of the example procedure; The dialysis process corresponding to the maximum value of the set difference significance among the remaining dialysis processes (excluding the first and last dialysis processes) during the use of the arteriovenous fistula is selected and denoted as the critical intervention period. The dialysis period following the critical intervention period within the arteriovenous fistula (AVF) usage phase is designated as the suspected AVF failure period.

[0009] Furthermore, determining the nursing intervention reminder level for the patient to be tested includes: Obtain average blood flow data for each dialysis session within the suspected period of arteriovenous fistula obstruction; A two-dimensional space is constructed using blood flow data as the horizontal axis and arterial pressure data as the vertical axis. The mean arterial pressure data and average blood flow data at all times during each dialysis period within the suspected arteriovenous fistula malfunction period are mapped to the two-dimensional space to obtain corresponding scatter points. Linear fitting is performed on all scatter points in the two-dimensional space to obtain the blood pressure and blood flow lines. The average blood flow data during non-dialysis periods is obtained, and the average blood flow data during non-dialysis periods, along with the blood pressure and flow rate line, are input into a trained neural network to obtain the nursing intervention reminder level for the patient to be tested.

[0010] Furthermore, the training method for the neural network includes: The ratio of the absolute difference between the slope of the blood pressure flow line of each analyzed patient and the preset normal slope to the preset normal slope threshold is used as the first barrier characteristic degree. The absolute difference between the average blood flow data of each analyzed patient during the non-dialysis period within the suspected arteriovenous fistula malfunction period and the preset normal flow threshold, and the ratio of the preset normal flow threshold, is used as the second malfunction characteristic. The first and second obstacle feature scores are weighted and summed to obtain the nursing intervention reminder score for each analyzed patient. The blood pressure and flow rate data of each analyzed patient during the suspected arteriovenous fistula obstacle period and the average blood flow data during the non-dialysis period are used as inputs to the neural network, and the nursing intervention reminder score is used as the output of the neural network to train the neural network.

[0011] Furthermore, determining the arteriovenous fistula care plan includes: The average blood flow data, average blood pressure data, and average blood flow data of patients in the pre-arteriovenous fistula stage and the mature arteriovenous fistula stage were obtained and analyzed, and standardized respectively. The three standardized data constituted a vascular feature sequence. Based on the distance between the vascular feature sequences of different analyzed patients, all analyzed patients were clustered to obtain several clusters; the arteriovenous fistula care plans of the other test patients in the cluster to be tested were used as the reference arteriovenous fistula care plans for the test patients.

[0012] Furthermore, the weight of the venous return obstruction is greater than the weight of the arterial supply obstruction.

[0013] Furthermore, the intensity of the upward trend is positively correlated with both the initial and final venous pressure and the initial degree of obstruction of venous return.

[0014] The present invention has the following beneficial effects: Firstly, by utilizing real-time hemodynamic data collected during each dialysis treatment of the arteriovenous fistula, compared to existing methods that rely on static imaging parameters before dialysis, this approach can detect the progressive deterioration of fistula function earlier during treatment.

[0015] Secondly, existing methods primarily rely on statistical variations in vascular morphology and single flow velocities. However, the impact of early stenosis is a chain reaction originating from the venous end and propagating to the arterial end. Assessing only a single location or static structure cannot comprehensively and sensitively reflect the pathophysiological process. Considering that the arteriovenous fistula and dialysis tubing form a closed-loop system, it is necessary to analyze the abnormal linkage patterns between veins and arteries, targeting the chain reaction during dialysis. By analyzing the dynamic trend and degree of change in venous pressure at the venous end during a single dialysis session, the dynamic development characteristics and ultimate impact intensity of arteriovenous fistula dysfunction (i.e., stenosis) can be analyzed. This allows for real-time and accurate capture of signs of deterioration in arteriovenous fistula function during dialysis, and simultaneous analysis of secondary changes at the arterial end. From the perspective of the entire circulatory loop, this approach can more sensitively and reliably capture the characteristic changes of early arteriovenous fistula stenosis. Considering that the deterioration of arteriovenous fistula function is a continuous pathophysiological process, trend analysis of the degree of venous return obstruction and arterial supply obstruction during multiple dialysis sessions can accurately divide the stable and abnormal periods of arteriovenous fistula function, identify the suspected period of arteriovenous fistula dysfunction, and achieve early warning of arteriovenous fistula dysfunction, even before clinical symptoms have changed.

[0016] Thirdly, based on the degree of inconsistency between the trends of venous pressure and transmembrane pressure data at different times during dialysis, it effectively distinguishes between physiological pressure increases caused by hemoconcentration and pathological pressure increases caused by stenosis, greatly improving the accuracy and reliability of early warning of arteriovenous fistula stenosis and reducing false positive alarms caused by hemoconcentration.

[0017] Fourthly, the relationship between blood flow and arterial pressure during periods of suspected arteriovenous fistula dysfunction can capture the subtle dynamic balance of functional compensation when the fistula is under dialysis load. Combined with blood flow data during non-dialysis phases that directly present indicators of fistula dysfunction, nursing intervention alerts can more accurately reflect the state of fistula dysfunction and generate personalized nursing intervention plans for patients. Attached Figure Description

[0018] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art 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 effort.

[0019] Figure 1 This is a system structure diagram of an intelligent nursing system for vascular health assessment and arteriovenous fistula in hemodialysis patients, provided in one embodiment of the present invention. Figure 2 This is a structural diagram of a vein end influence analysis module provided in one embodiment of the present invention; Figure 3 This is a schematic diagram of a computer device for intelligent nursing care of hemodialysis patients' vascular health assessment and arteriovenous fistula, provided in one embodiment of the present invention. Detailed Implementation

[0020] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an intelligent nursing system for vascular health assessment and arteriovenous fistula in hemodialysis patients proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0022] The following description, in conjunction with the accompanying drawings, details the specific solution of the intelligent nursing system for vascular health assessment and arteriovenous fistula of hemodialysis patients provided by the present invention.

[0023] Example 1: Please see Figure 1 The diagram illustrates a system block diagram of an intelligent nursing system for vascular health assessment and arteriovenous fistula in hemodialysis patients according to an embodiment of the present invention. The system includes: a data acquisition module 110, a venous end influence analysis module 120, an arterial end influence analysis module 130, an arteriovenous fistula disorder analysis module 140, and an arteriovenous fistula assessment and nursing module 150.

[0024] The data acquisition module 110 is used to record the patient to be tested and the test patient as the analysis patient, and to acquire the venous pressure data, arterial pressure data and transmembrane pressure data at each moment during each dialysis process of the analysis patient during the use of the arteriovenous fistula.

[0025] Arteriovenous fistulas (AVFs) require a postoperative maturation period before they can be safely used for hemodialysis. According to clinical guidelines and general practice, the maturation period is typically 4 to 8 weeks post-surgery. To ensure the reliability of the assessment and eliminate interference from changes in the maturation process itself, the start date of the 9th week (the end of the 8th week post-surgery) is defined as the beginning of the AVF usage phase. The end date of the AVF usage phase for the test patient is the date when the patient discontinues hemodialysis treatment. The end date of the AVF usage phase for the candidate patient is the date of their last dialysis session using the AVF. The test patient is someone who has completed hemodialysis treatment, while the candidate patient is someone currently undergoing hemodialysis treatment.

[0026] Each dialysis session is defined as a dialysis interval. Through the hemodialysis machine's data interface, built-in pressure and flow sensors and other monitoring modules synchronously collect and analyze hemodynamic data at various times within each dialysis interval during the arteriovenous fistula (AVF) usage period. This includes venous pressure, arterial pressure, and transmembrane pressure data. It is important to note that each AVF usage period must comprise at least three dialysis intervals.

[0027] In this embodiment of the invention, the pressure sensor and the flow sensor have the same data acquisition frequency, both set to once every 1 second. The implementer can set it according to the specific situation.

[0028] The venous end impact analysis module 120 is used to determine the degree of venous return obstruction in each dialysis session based on the trend and degree of change of venous pressure data within each dialysis session, as well as the degree of inconsistency between the trend of venous pressure data and transmembrane pressure data at each moment within each dialysis session.

[0029] Ideally, the arteriovenous fistula (AVF) and dialysis tubing form a smooth, stable closed-loop system, with the arterial end supplying blood and the venous end providing return. When AVF dysfunction occurs at the venous end, the balance of this closed-loop system is disrupted. The venous end is usually affected first and most directly, with the impact then spreading to the arterial end, causing a continuous reaction. Specifically, AVF dysfunction at the venous end obstructs the outflow tract, causing increased pressure and hindering blood return. This obstruction leads to blood pooling downstream of the pump, and the resistance is transmitted upstream, increasing the pump's workload. To maintain the set blood flow rate, the pump needs to generate greater suction at the arterial end to overcome the increased overall system resistance.

[0030] In a healthy arteriovenous fistula (AVF), the blood flow path is unobstructed and the resistance is stable, meaning the pressure in the venous segment is relatively consistent. Early AVF dysfunction manifests primarily at the venous end, obstructing the outflow tract. This obstruction becomes increasingly pronounced with dialysis treatment, leading to a sustained increase in venous pressure, with the pressure at the end of treatment significantly higher than at the beginning. Therefore, based on the trend and degree of change in venous pressure data during each dialysis session, the dynamic development characteristics and ultimate impact of AVF dysfunction at the venous end are captured sequentially to assess the physiological extent of AVF dysfunction in the venous segment during that dialysis session.

[0031] Besides arteriovenous fistula dysfunction, factors affecting the rise in venous pressure during dialysis may also be related to hemoconcentration and a slight increase in viscosity. Specifically, as ultrafiltration removes water, the concentration of formed elements such as red blood cells and proteins in the blood increases, leading to a slow increase in blood viscosity. Consequently, venous pressure exhibits a gentle and gradual upward trend during ultrafiltration, which is a normal physiological process. To distinguish between these two causes, changes in transmembrane pressure can be observed simultaneously. Specifically, increased blood viscosity increases the resistance to blood flow through the dialyzer's hollow fibers, resulting in a synchronized, gentle increase in transmembrane pressure; that is, the trends of venous pressure and transmembrane pressure are relatively consistent. However, the pressure increase caused by arteriovenous fistula dysfunction is mainly characterized by an isolated and significant rise in venous pressure, meaning that the trends of venous pressure and transmembrane pressure are inconsistent. Therefore, combining the degree of inconsistency between the trends of venous pressure and transmembrane pressure data during dialysis can more accurately determine the main cause of the venous pressure increase, obtain the degree of venous return obstruction, greatly improve the accuracy and reliability of early warning of arteriovenous fistula stenosis, and reduce false positive alarms caused by hemoconcentration.

[0032] The arterial end impact analysis module 130 is used to determine the degree of arterial supply obstruction during each dialysis session based on the trend and degree of change in the patient's arterial pressure data during each dialysis session.

[0033] During dialysis, the blood pump generates negative pressure in the arterial segment at the proximal end of the arteriovenous fistula (AVF), drawing blood from the AVF into the arterial line via an arterial needle. When early AVF dysfunction occurs, blood flow resistance increases significantly, requiring the blood pump to increase suction force to maintain the set blood flow rate. This manifests as a sustained increase in the absolute negative value of arterial pressure, with the absolute negative pressure at the end of dialysis typically being much higher than at the start of treatment. Therefore, the pressure changes at the arterial and venous ends are similar during early AVF dysfunction, allowing assessment of the physiological extent of arterial segment AVF dysfunction during a single dialysis session and determining the degree of arterial supply obstruction.

[0034] The fistula failure analysis module 140 is used to identify suspected fistula failure periods during the use of the fistula based on the changing trends of venous return obstruction and arterial supply obstruction.

[0035] According to the pathophysiological mechanism, early arteriovenous fistula (AVF) dysfunction originates from obstruction of the venous outflow tract, making abnormal venous pressure the primary and most direct manifestation of hemodynamic changes. Abnormal arterial pressure is a secondary and indirect effect caused by abnormal venous resistance. In actual clinical monitoring, due to limitations such as sensor sensitivity and data sampling frequency, abnormal signals of venous and arterial pressure may be captured almost simultaneously, making it difficult to clearly distinguish their temporal order. Therefore, analyzing the linkage abnormal patterns between veins and arteries in response to the chain reaction during dialysis can more sensitively and reliably capture the characteristic changes of early AVF stenosis. The deterioration of AVF function is a continuous pathophysiological process, and the degree of venous and arterial impaction progressively worsens as AVF function deteriorates. Based on the changing trends of venous return obstruction and arterial supply obstruction, the stable and abnormal periods of AVF function can be accurately divided to identify suspected AVF dysfunction periods. Early detection of AVF dysfunction, even before clinical symptom recognition changes occur, can provide early warning of AVF dysfunction.

[0036] The fistula assessment and care module 150 is used to determine the nursing intervention alert level of the patient based on the relationship between blood flow data and arterial pressure data during each dialysis stage of the suspected fistula failure period, as well as blood flow data during non-dialysis stages, using a trained neural network, and to determine the fistula care plan.

[0037] The relationship between blood flow data and arterial pressure data reflects the functional changes of the arteriovenous fistula (AVF) during dialysis. Blood flow data during non-dialysis phases provides the most direct evidence of structural functional impairment of the AVF. The nursing intervention alert level obtained through the fusion analysis of these two types of information using neural networks can improve the accuracy of AVF health assessment and early warning capabilities. Then, personalized nursing intervention plans are generated for patients. This represents a shift from lagging, singular structural assessment to a proactive, multi-dimensional functional vascular health early warning and intelligent nursing decision-making approach.

[0038] Please see Figure 2 The diagram shows a structural diagram of a vein obstruction analysis module provided in an embodiment of the present invention. The vein obstruction analysis module includes: a vein obstruction initial analysis unit 121 and a vein obstruction final determination unit 122.

[0039] Initial analysis unit 121 for venous obstruction: used to determine the initial degree of venous return obstruction for each dialysis session based on the trend and degree of change of venous pressure data within each dialysis session.

[0040] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the initial obstruction of venous return includes: obtaining a first-order difference sequence of the venous pressure sequence; recording the ratio of the number of elements with positive values ​​to the total number of elements in the first-order difference sequence as a trend index for each dialysis period; calculating the mean of all elements in the first-order difference sequence; using the product of the mean and the trend index as the upward trend strength for each dialysis process; calculating the difference between the venous pressure data at the end and beginning times within each dialysis period as the venous pressure beginning-to-end difference; and obtaining the initial obstruction of venous return for each dialysis period based on the upward trend strength and the venous pressure beginning-to-end difference.

[0041] It should be noted that, taking venous pressure data as an example for analysis, the trend indicator reflects the persistence of the upward trend in venous pressure data, while the mean of all elements in the first-order difference sequence reflects the severity of the change in venous pressure data. A stronger upward trend indicates a more sustained and dramatic increase in venous pressure during dialysis, meaning a rapid and significant rise in venous pressure, and thus more severe early arteriovenous fistula dysfunction at the venous end. The difference between the initial and final venous pressure values ​​the net increase in venous pressure before and after treatment. A larger difference indicates a significant increase in reflux resistance during dialysis, meaning more severe outflow obstruction and more severe early arteriovenous fistula dysfunction at the venous end. Therefore, the strength of the upward trend and the difference between the initial and final venous pressure are both positively correlated with the initial degree of venous reflux obstruction.

[0042] In this embodiment of the invention, the intensity of the upward trend and the difference between the beginning and end of venous pressure are normalized respectively, and the arithmetic mean of the normalized result of the intensity of the upward trend and the normalized result of the difference between the beginning and end of venous pressure for each dialysis period is taken as the initial degree of obstruction of venous return.

[0043] In this embodiment of the invention, the upward trend intensity and the difference between the beginning and end of venous pressure data for all dialysis processes during the use of the arteriovenous fistula for all analyzed patients are normalized using the minimax normalization method for each dialysis process; alternatively, the Z-score normalization method can be used for normalization, which is not limited here.

[0044] The final determination unit 122 for venous obstruction is used to arrange the venous pressure data and transmembrane pressure data of all moments within each dialysis session in chronological order to obtain the venous pressure sequence and transmembrane pressure sequence in sequence; to perform negative correlation and normalization on the correlation coefficient between the venous blood pressure sequence and the transmembrane pressure sequence of the same dialysis session to obtain the obstruction value of each dialysis session; and to weight the initial obstruction of venous return based on the obstruction value to obtain the degree of venous return obstruction of each dialysis session.

[0045] It should be noted that a larger correlation coefficient indicates a more consistent trend between venous pressure and transmembrane pressure changes during dialysis, suggesting that the venous pressure changes are more likely due to hemoconcentration, and the lower the reference value of vascular access dysfunction. Conversely, a smaller correlation coefficient suggests that the venous pressure changes are more likely due to early arteriovenous fistula dysfunction, and the higher the reference value of vascular access dysfunction. Therefore, by using a negatively correlated and normalized correlation coefficient to weight the initial degree of venous return obstruction, the degree of venous return obstruction can accurately reflect the severity of early arteriovenous fistula dysfunction at the venous end.

[0046] In this embodiment of the invention, the product of the obstruction value of each dialysis period and the initial obstruction of venous return is used as the venous return obstruction degree of each dialysis period.

[0047] In this embodiment of the invention, since the correlation coefficient ranges from -1 to 1, half of the difference between the constant 1 and the correlation coefficient is calculated to achieve negative correlation and normalization of the correlation coefficient.

[0048] In one implementation of this invention, the correlation coefficient is the Pearson correlation coefficient.

[0049] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the degree of arterial supply obstruction includes: using the absolute value of the arterial pressure data at each moment within each dialysis period to update the venous pressure data in the process of obtaining the initial degree of venous return obstruction, and using the recalculated initial degree of venous return obstruction as the degree of arterial supply obstruction for each dialysis period.

[0050] It is important to note that, considering arterial pressure data primarily reflects the resistance of the blood pump, and blood concentration has a relatively small impact on vascular resistance, the interference from blood concentration should not be considered when analyzing the severity of arteriovenous fistula dysfunction at the arterial end during dialysis. Furthermore, the pressure changes at the arterial and venous ends are similar in the early stages of fistula dysfunction, and the normal range for negative arterial pressure during hemodialysis is generally between -150 mmHg and -50 mmHg. Therefore, the method for obtaining the initial venous return obstruction is similar to that for arterial supply obstruction; only the venous pressure data in the initial venous return obstruction data needs to be replaced with the absolute value of the arterial pressure data, while all other information remains unchanged.

[0051] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the suspected arteriovenous fistula (AVF) malfunction period includes: weighted summing of the venous return obstruction and the arterial supply obstruction during the same dialysis period to obtain the venous-arterial impact degree for each dialysis period; randomly selecting one dialysis process from the AVF usage phase as an example process; forming a first set by the venous-arterial impact degree of the example process and all previous dialysis processes within the AVF usage phase; forming a second set by the venous-arterial impact degree of all dialysis processes after the example process within the AVF usage phase; calculating the inter-group variance between the first set and the second set, and recording it as the set difference significance of the example process; selecting the dialysis process corresponding to the maximum value of the set difference significance among the remaining dialysis processes within the AVF usage phase excluding the first and last dialysis processes, and recording it as the critical intervention period; and recording the dialysis period after the critical intervention period within the AVF usage phase as the suspected AVF malfunction period.

[0052] It should be noted that, due to the hemodynamic changes in early arteriovenous fistula dysfunction, with abnormal venous pressure as the direct manifestation and abnormal arterial pressure as a secondary manifestation, the weight of venous return obstruction should be greater than that of arterial supply obstruction when weighted and fused to obtain the degree of venous-arterial impaction during the process of obtaining the degree of venous-arterial impaction through arterial supply obstruction. In one implementation of this invention, the weight of venous return obstruction is set to 0.6, and the weight of arterial supply obstruction is set to 0.4. Implementers can set these weights according to specific circumstances. The deterioration of arteriovenous fistula function is a continuous pathophysiological process, that is, the fistula gradually develops from a healthy state to functional deterioration, and the deterioration then shows a progressively worsening trend, resulting in a smaller degree of venous-arterial impaction in the healthy state and a larger degree in the functional deterioration state, with the degree of venous-arterial impaction gradually increasing as the deterioration progressively worsens. The greatest difference in aggregate significance corresponds to the largest difference in the degree of influence of the vein and artery during the two dialysis procedures before and after the dialysis session. This strongly indicates that the functional status of the arteriovenous fistula underwent the most significant and stable systemic shift after this dialysis session, meaning that the fistula transitioned from the mature phase to the stenotic phase during this dialysis session. Therefore, the dialysis session corresponding to the greatest difference in aggregate significance was selected as the key intervention period.

[0053] In one specific implementation of this invention, the set difference significance of the example process is... Expressed as a formula: In the formula, The number of elements in the first set of the example process; The number of elements in the second set of the example process; The mean of the elements in the first set of the example process; The mean of the elements in the second set of the example process; This represents the mean of all elements in the first and second sets of the example process. It should be noted that if... The larger the value, the more significant the difference in the degree of venous and arterial impact between the two parts during the example process division. It is important to note that the method for obtaining the venous and arterial impact is the same for all dialysis stages during the arteriovenous fistula use phase as for the example stage.

[0054] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the nursing intervention reminder score includes: obtaining the average blood flow data of each dialysis session within the suspected arteriovenous fistula (AVF) malfunction period; constructing a two-dimensional space with the blood flow data as the horizontal axis and the arterial pressure data as the vertical axis; mapping the mean arterial pressure data and average blood flow data at all times within each dialysis session within the suspected AVF malfunction period to the two-dimensional space to obtain corresponding scatter points; performing linear fitting on all scatter points in the two-dimensional space to obtain a blood pressure-flow rate line; obtaining the average blood flow data of non-dialysis sessions; inputting the average blood flow data of non-dialysis sessions and the blood pressure-flow rate line into a trained neural network to obtain the nursing intervention reminder score of the patient to be tested. Each dialysis session corresponds to one scatter point, and the blood pressure-flow rate line presents the relationship between the changes in blood flow data and arterial pressure data during the dialysis session.

[0055] It's important to note that the slope of the blood pressure-flow rate curve is essentially a reflection of vascular flow resistance. The more severe the arteriovenous fistula (AVF) dysfunction, the higher the pressure required to maintain the same blood flow, meaning a larger slope. Therefore, the slope of the blood pressure-flow rate curve can serve as a crucial feature for detecting AVF dysfunction. It is highly sensitive to early, non-occlusive luminal changes, such as intimal hyperplasia leading to wall stiffening and slight reduction in diameter, often showing abnormalities earlier than static flow rate values. Blood flow during non-dialysis periods is unaffected by dialysis equipment like blood pumps and better reflects the natural state of the AVF. When AVF dysfunction is severe, blood flow decreases, providing the most direct evidence of structural dysfunction.

[0056] In this embodiment of the invention, the training process of the neural network is as follows: the ratio of the absolute difference between the slope of the blood pressure flow line of each analyzed patient and a preset normal slope to a preset normal slope threshold is used as the first obstacle feature degree; the ratio of the absolute difference between the average blood flow data of each analyzed patient during the suspected arteriovenous fistula obstacle period and the non-dialysis period to a preset normal flow threshold is used as the second obstacle feature degree; the first obstacle feature degree and the second obstacle feature degree are weighted and summed to obtain the nursing intervention reminder degree for each analyzed patient; the blood pressure flow line of each analyzed patient during the suspected arteriovenous fistula obstacle period and the average blood flow data during the non-dialysis period are used as the input of the neural network, and the nursing intervention reminder degree is used as the output of the neural network to train the neural network. All test patients are divided into training set and validation set in a 7:3 ratio. The neural network is trained using the training set, the loss function is the cross-entropy function, and the gradient descent method is used to train until the loss function converges. The robustness of the training results is verified through the validation set to obtain the trained neural network.

[0057] It should be noted that the first symptom measure quantifies the degree of abnormality in the hemodynamic characteristics of the arteriovenous fistula, while the second symptom measure quantifies the degree of abnormality in the static blood flow of the fistula. The nursing intervention alertness score, which combines both, is an indicator for comprehensively assessing the severity of arteriovenous fistula dysfunction and the urgency of intervention. The higher the nursing intervention alertness score, the more severe the arteriovenous fistula dysfunction, i.e., the more severe the stenosis, and the higher the urgency of nursing intervention, which helps to achieve early warning and graded intervention.

[0058] It is important to note that the average blood flow data for each dialysis session is the average of the blood flow data at all times within that dialysis session. During periods of suspected fistula malfunction, excluding dialysis sessions, blood flow in the fistula venous segment (2-4 cm from the arterial anastomosis) is measured using Doppler ultrasound. Measurements are repeated at the same Doppler angle to ensure consistency with the position and angle data during dialysis, yielding the blood flow data for each measurement in these remaining periods. The average of these measurements is then calculated to obtain the average blood flow data for each patient during non-dialysis periods. In this embodiment, the data collection frequency for the blood flow in the fistula venous segment during non-dialysis periods is set to once daily; however, the implementer can adjust this frequency according to specific circumstances. Based on historical patient datasets, including abnormal blood pressure flow rate slopes, abnormal non-dialysis blood flow rates, and corresponding clinical intervention records, the necessity of actual nursing interventions, such as the urgency assessed by clinical experts or the occurrence of the final intervention event, is used as the training objective. Regression analysis or machine learning algorithms are employed for fitting to determine the optimal weighting coefficients for the first and second impairment characteristics during the weighted summation of these characteristics. In one implementation of this invention, a large number of blood pressure flow rate lines from healthy arteriovenous fistula (AVF) test patients during the AVF usage phase are obtained, and the average slope of these blood pressure flow rate lines is used as the preset normal slope. The method for obtaining blood pressure flow rate lines during the AVF usage phase is similar to that for suspected AVF impairment periods, except that the suspected AVF impairment period is replaced with the AVF usage phase, while other content remains unchanged.

[0059] In one implementation of this invention, clinical guidelines typically recommend that the blood flow rate of the arteriovenous fistula should be at least 500 to 600 ml per minute to ensure adequate dialysis, with 500 ml per minute serving as the preset normal flow rate threshold.

[0060] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the arteriovenous fistula (AVF) care plan for the patient to be tested includes: acquiring and analyzing the average blood flow data, average blood pressure data, and average blood flow data of the patient in the pre-AVF construction stage and the mature AVF stage, respectively, and performing standardization processing on each of them, thereby constructing a vascular feature sequence from the three standardized data; clustering all analyzed patients based on the distance between the vascular feature sequences of different analyzed patients to obtain several clusters; and using the AVF care plans of the other tested patients in the cluster where the patient to be tested is located as the reference AVF care plan for the patient to be tested.

[0061] It should be noted that the pre-defined period of one week before fistula establishment is designated as the pre-fistula stage. During both the pre-fistula stage and the fistula maturation stage, average blood flow data were collected using the same method during non-dialysis periods. In the pre-fistula stage, standard clinic blood pressure measurements were taken of the brachial arteries in both upper arms. The average of the bilateral measurements was recorded as a single blood pressure measurement. The average of all blood pressure measurements during the pre-fistula stage was used as the average blood pressure data for that stage. Blood pressure data was collected daily. Pre-fistula baseline physiological data reflects the patient's overall cardiovascular status and may affect fistula maturation and long-term function. Blood flow after fistula maturation directly reflects fistula function. Therefore, patients within the same cluster may have similar vascular conditions and fistula function, and historically successful nursing protocols may be valuable for reference. Based on the reference fistula nursing protocol for the target patient, the physician, considering the patient's specific circumstances such as age, past medical history, and other clinical characteristics, determined a safe and effective final nursing protocol through personalized adjustments.

[0062] In this embodiment of the invention, the average blood flow data and average blood pressure data in the pre-fistula stage and the average blood flow data in the mature stage of the fistula are recorded as characteristic indicators. Based on each characteristic indicator of all analyzed patients, the Z-score standardization method is used to standardize each characteristic indicator of each patient.

[0063] In one implementation of this invention, the distance between the vascular feature sequences of different analyzed patients is used as the distance index in the clustering process. The K-means clustering algorithm is selected to cluster the analyzed patients, where the K value is determined by the elbow method.

[0064] This invention is now complete.

[0065] Example 2: Figure 3 This is a schematic diagram of a computer device for intelligent nursing care of hemodialysis patients' vascular health assessment and arteriovenous fistula, provided as an embodiment of the present invention. Exemplary, such as... Figure 3 As shown, the computer device includes: a memory 201, a processor 202, and a computer program 203 stored in the memory 201 and running on the processor 202, wherein when the processor 202 executes the computer program 203, the computer device can execute any of the aforementioned intelligent nursing systems for vascular health assessment and arteriovenous fistulas of hemodialysis patients.

[0066] Furthermore, embodiments of this application also protect an apparatus that may include a memory and a processor, wherein the memory stores executable program code, and the processor is used to call and execute the executable program code to execute the intelligent nursing system for vascular health assessment and arteriovenous fistula of hemodialysis patients provided in embodiments of this application.

[0067] This embodiment can divide the device into functional modules based on the above method example. For example, each module can correspond to a separate function, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware. It should be noted that the module division in this embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0068] It should be understood that the device provided in this embodiment is used to perform the above-described intelligent nursing system for vascular health assessment and arteriovenous fistula of hemodialysis patients, and therefore can achieve the same effect as the above-described implementation method.

[0069] When using integrated units, the device may include a processing module and a storage module. When applied to a workpiece, the processing module can be used to control and manage the workpiece's operations. The storage module can be used to support the execution of program code by the workpiece.

[0070] The processing module may be a processor or a controller, which can implement or execute various exemplary logic blocks, modules, and circuits as disclosed in this application. The processor may also be a combination of computing functions, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and microprocessors, etc., and the storage module may be a memory.

[0071] Example 3: This embodiment also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer, the computer executes the above-described related method steps to realize the intelligent nursing system for vascular health assessment and arteriovenous fistula of hemodialysis patients provided in the above embodiment.

[0072] In this embodiment, the device and computer-readable storage medium are used to execute the corresponding system provided above. Therefore, the beneficial effects that can be achieved can be referred to the beneficial effects of the corresponding system provided above, and will not be repeated here.

[0073] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0074] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0075] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An intelligent nursing system for assessing vascular health and managing arteriovenous fistulas in hemodialysis patients, characterized in that: The system includes: The data acquisition module is used to record the patients to be tested and the test patients as the analysis patients, and to acquire the venous pressure data, arterial pressure data and transmembrane pressure data at each moment during each dialysis process of the analysis patients during the use of the arteriovenous fistula. The venous end impact analysis module is used to determine the degree of venous return obstruction in each dialysis session based on the trend and degree of change of venous pressure data within each dialysis session, as well as the degree of inconsistency between the trend of venous pressure data and transmembrane pressure data at each moment within each dialysis session. The arterial impact analysis module is used to determine the degree of arterial supply obstruction during each dialysis session based on the trend and degree of change in the patient's arterial pressure data during each dialysis session. The arteriovenous fistula (AVF) obstruction analysis module is used to determine the suspected AVF obstruction period during the AVF usage phase based on the changing trends of the venous return obstruction and the arterial supply obstruction. The fistula assessment and care module is used to determine the nursing intervention alert level for the patient based on the relationship between blood flow and arterial pressure data during each dialysis stage in the suspected fistula failure period, as well as blood flow data during non-dialysis stages, using a trained neural network, and to determine the fistula care plan.

2. The intelligent nursing system for vascular health assessment and arteriovenous fistula in hemodialysis patients according to claim 1, characterized in that, Determining the degree of venous return obstruction during each dialysis session includes: Based on the trend and degree of change of venous pressure data during each dialysis session, the initial degree of venous return obstruction during each dialysis session is determined. The venous pressure and transmembrane pressure data at all times within each dialysis session are arranged in chronological order to obtain the venous pressure sequence and transmembrane pressure sequence, respectively. The correlation coefficient between the venous blood pressure sequence and the transmembrane pressure sequence in the same dialysis session is negatively correlated and normalized to obtain the blocked value of each dialysis session. The initial degree of venous return obstruction is weighted based on the obstruction value to obtain the degree of venous return obstruction for each dialysis session.

3. The intelligent nursing system for vascular health assessment and arteriovenous fistula in hemodialysis patients according to claim 2, characterized in that, Determining the initial venous return obstruction for each dialysis session includes: Obtain the first-order difference sequence of the venous pressure sequence, and record the ratio of the number of elements with positive values ​​to the total number of elements in the first-order difference sequence as the trend indicator for each dialysis period. Calculate the mean of all elements in the first-order difference sequence, and multiply the mean by the trend indicator to obtain the strength of the upward trend for each dialysis process. Calculate the difference between the venous pressure data at the end and start times of each dialysis session, and use this as the venous pressure start-end difference; Based on the strength of the upward trend and the difference between the beginning and end of the venous pressure, the initial degree of venous return obstruction during each dialysis period is obtained.

4. The intelligent nursing system for vascular health assessment and arteriovenous fistula in hemodialysis patients according to claim 3, characterized in that, The determination of the degree of arterial supply obstruction during each dialysis session includes: The venous pressure data obtained during the acquisition of the initial venous return obstruction is updated using the absolute value of the arterial pressure data at each moment within each dialysis session. The recalculated initial venous return obstruction is then used as the arterial supply obstruction for each dialysis session.

5. The intelligent nursing system for vascular health assessment and arteriovenous fistula in hemodialysis patients according to claim 1, characterized in that, The period during which suspected arteriovenous fistula dysfunction is identified during the use of the fistula includes: The degree of venous return obstruction and the degree of arterial supply obstruction during the same dialysis period are weighted and summed to obtain the degree of venous and arterial impact during each dialysis period; One dialysis process is randomly selected during the use of the arteriovenous fistula and recorded as an example process. The first set consists of the venous and arterial impact of the example process during the use of the arteriovenous fistula and all the dialysis processes before it. The second set consists of the venous and arterial impact of all the dialysis processes after the example process during the use of the arteriovenous fistula. Calculate the between-group variance between the first set and the second set, and denote it as the set difference significance of the example procedure; The dialysis process corresponding to the maximum value of the set difference significance among the remaining dialysis processes (excluding the first and last dialysis processes) during the use of the arteriovenous fistula is selected and denoted as the critical intervention period. The dialysis period following the critical intervention period within the arteriovenous fistula (AVF) usage phase is designated as the suspected AVF failure period.

6. The intelligent nursing system for vascular health assessment and arteriovenous fistula in hemodialysis patients according to claim 1, characterized in that, The determination of the nursing intervention reminder level for the patient to be tested includes: Obtain average blood flow data for each dialysis session within the suspected period of arteriovenous fistula obstruction; A two-dimensional space is constructed using blood flow data as the horizontal axis and arterial pressure data as the vertical axis. The mean arterial pressure data and average blood flow data at all times during each dialysis period within the suspected arteriovenous fistula malfunction period are mapped to the two-dimensional space to obtain corresponding scatter points. Linear fitting is performed on all scatter points in the two-dimensional space to obtain the blood pressure and blood flow lines. The average blood flow data during non-dialysis periods is obtained, and the average blood flow data during non-dialysis periods, along with the blood pressure and flow rate line, are input into a trained neural network to obtain the nursing intervention reminder level for the patient to be tested.

7. The intelligent nursing system for vascular health assessment and arteriovenous fistula in hemodialysis patients according to claim 6, characterized in that, The training method for the neural network includes: The ratio of the absolute difference between the slope of the blood pressure flow line of each analyzed patient and the preset normal slope to the preset normal slope threshold is used as the first barrier characteristic degree. The absolute difference between the average blood flow data of each analyzed patient during the non-dialysis period within the suspected arteriovenous fistula malfunction period and the preset normal flow threshold, and the ratio of the preset normal flow threshold, is used as the second malfunction characteristic. The first and second obstacle feature scores are weighted and summed to obtain the nursing intervention reminder score for each analyzed patient. The blood pressure and flow rate data of each analyzed patient during the suspected arteriovenous fistula obstacle period and the average blood flow data during the non-dialysis period are used as inputs to the neural network, and the nursing intervention reminder score is used as the output of the neural network to train the neural network.

8. The intelligent nursing system for vascular health assessment and arteriovenous fistula in hemodialysis patients according to claim 1, characterized in that, The determination of the arteriovenous fistula care plan includes: The average blood flow data, average blood pressure data, and average blood flow data of patients in the pre-arteriovenous fistula stage and the mature arteriovenous fistula stage were obtained and analyzed, and standardized respectively. The three standardized data constituted a vascular feature sequence. Based on the distance between the vascular feature sequences of different analyzed patients, all analyzed patients were clustered to obtain several clusters; the arteriovenous fistula care plans of the other test patients in the cluster to be tested were used as the reference arteriovenous fistula care plans for the test patients.

9. The intelligent nursing system for vascular health assessment and arteriovenous fistula in hemodialysis patients according to claim 5, characterized in that, The weight of the degree of obstruction of venous return is greater than the weight of the degree of obstruction of arterial supply.

10. The intelligent nursing system for vascular health assessment and arteriovenous fistula in hemodialysis patients according to claim 3, characterized in that, The intensity of the upward trend is positively correlated with both the initial and final venous pressure and the initial degree of obstruction of venous return.

Images