Heart failure diagnostic tools and methods using signal trace analysis

Abstract

The disclosed systems and methods assess patient status by automatically extracting features from vascular area traces and generating cardiac health indication parameters based on the extracted features to provide a normalized hyperemia index, thereby providing improved therapeutic effects for heart failure patients. Examples include a diagnostic engine (102) receiving an area trace from a trace generator (104) and communicating patient status information over a user interface (116). The diagnostic engine may include a trace feature detector (106), an index generator (108), a boundary generator (110), an index generator (112), and decision logic (114). The disclosed system may be executed using one or more processing devices or in a networked distributed processing system.

Description

[0001] Related application data

[0002] This application claims the benefit of priority to U.S. Provisional Patent Application Serial No. 63 / 318,216, filed on March 9, 2022, entitled “Heart Failure Diagnostic Tools and Methods Using Signal Trace Analysis,” which is incorporated herein by reference in its entirety.

[0003] field

[0004] The present disclosure relates to heart failure diagnosis, and more particularly, to tools and methods that facilitate the use of data generated by heart failure related sensors to improve patient outcomes.

[0005] background

[0006] Hemodynamic congestion is now commonly measured using catheter-based filling pressures of right atrial pressure (RAP) and pulmonary capillary wedge pressure (PCWP). However, due to the small range of these pressures relative to potential measurement errors, ranging from 2 to 6 mmHg (RAP) and 4 to 12 mmHg (PCWP), it is challenging to capture the relevant pressure changes using common pressure sensing detectors such as Swan-Ganz catheters or other implantable pressure sensors (which may typically have accuracy limitations within about ±5 mmHg).

[0007] The applicant has previously developed and disclosed a number of different sensors for determining a patient's fluid state based on direct measurements of vessel dimensions, which refer to the geometry, i.e., cross-sectional area, and dilation or collapse of the vessel. Such measurements of the vessels, particularly the inferior vena cava (IVC), can be more directly related to the patient's circulating blood volume and state of congestion. Thus, these measurements using these sensors can potentially be used to estimate a patient's circulating blood volume and state of congestion. In particular, these sensors can potentially be used to determine whether the circulating blood volume is too high or too low, whether the circulating blood volume is increasing or decreasing, and potentially determine what treatment should be prescribed, such as a diuretic or a vasodilator.

[0008] The new device developed and disclosed by the applicant includes an external ultrasound device and an implantable sensor that can be placed for a long time and is suitable for monitoring patients with chronic diseases. Examples of such implantable wireless sensors and external ultrasound devices are disclosed in, for example, the following documents: U.S. Patent Application No. 15 / 549,042 (U.S. Patent No. 10,905,393, issued on February 2, 2021) and U.S. Patent Application No. 16 / 177,183 (U.S. Patent No. 10,806,352, issued on October 20, 2020) and entitled "Wireless Vascular Monitoring Implants" filed on October 31, 2018, each of which is incorporated herein in its entirety. In other clinical situations, such as short-term acute condition monitoring and in-hospital treatment, vascular size sensors for direct fluid state determination and monitoring can be catheter-based. For example, an example of such a catheter-based sensor is disclosed in U.S. patent application Ser. No. 15 / 750,100, filed on February 2, 2018, and entitled “Devices and Methods for Measurement of Vena Cava Dimensions, Pressure and Oxygen Saturation” (U.S. Pat. No. 11,039,813, issued on June 22, 2021), which is incorporated herein in its entirety.

[0009] The present applicant has also developed and disclosed novel diagnostic and treatment systems and methods based on the use of the above-mentioned sensor devices, some of which are disclosed in the above-mentioned patent applications, and some of which are disclosed in the following documents: U.S. Patent No. 11,564,596, entitled “Systems and Methods for Patient Fluid Management,” issued on January 31, 2023, and U.S. Patent No. US2021 / 0244381 A1, entitled “Patient Fluid Management Systems and Methods Employing Integrated Fluid Status Sensing,” published on August 12, 2021, each of which is incorporated herein by reference in its entirety.

[0010] Despite the above-mentioned advances, challenges remain regarding the clinical interpretation of the signals generated by various vessel area sensing technologies and, in some cases, the new datasets generated thereby.

[0011] One of the challenges is to calibrate measurements of vessel volume-based parameters that arise from intra- and inter-individual variability due to heterogeneity in vessel shape to generate comparable quantitative output metrics indicative of right atrial filling pressure or congestion status. For example, the area of ​​the IVC depends on both the location along the IVC and the individual being observed, making it difficult to plot comparisons of absolute area measurements between patients as a group, which can inhibit the clinical use of the generated data. While commonly available data normalization algorithms can be applied to address this limitation, when any feature is arbitrarily normalized, the normal range will likely be surrounded by thresholds of abnormal and non-intuitive values, i.e., the normalized variable will be calculated in units (which may be dimensionless) with associated thresholds that are unfamiliar to clinicians and are not part of standard practice and guidelines. This abnormal numbering is a challenge for physicians and patients, especially as methods evolve and diagnostic algorithms improve with ever-greater degrees of sophistication, and patients increasingly become relevant consumers of the data.

[0012] There may also be challenges with general clinical acceptance of new data types. Long-standing and well-known physiological parameters have well-known ranges (e.g., blood pressure, temperature, respiratory rate, etc.). These develop over time into domain-specific knowledge and are incorporated into clinical guidelines, etc. Therefore, to help promote acceptance and encourage use of new enabling systems and datasets, correlations of new signals and datasets to known physiological parameters need to be developed to facilitate use by clinicians.

[0013] With multiple discrete inputs to diagnostic and therapeutic algorithms, it is challenging to set actionable thresholds without the instructions becoming overly complex. Therefore, there is a need to reduce the data review burden on users, both patients and healthcare providers, when new data sets and parameters are introduced. Challenges for users include definition of action thresholds and determining which medication from their inventory to deploy and how many doses to deploy, as they often do not have reliable inputs to inform medication titration decisions, good methods to monitor patient medication adherence, or good measurements of individual responses to different doses of medication.

[0014] Various embodiments disclosed herein address these challenges and present solutions designed to increase the speed and accuracy of heart failure diagnosis and treatment decisions and thereby improve patient outcomes and reduce hospitalization rates and costs.

[0015] Overview

[0016] In one embodiment, the present disclosure relates to an automatic heart failure diagnosis device. The device includes: a trace feature detector, which is used to identify selected features and at least one of the amplitude and timing of the identified features for received periodic vascular area traces representing changes in the patient's fluid state over time, wherein the selected features include one or more of the following items: the interval time of each respiratory cycle, the area amplitude of respiratory adjustment, the interval time of each cardiac cycle, the area amplitude of cardiac adjustment, the main cardiac peak, the secondary cardiac peak, the respiratory-related area reduction, the action type, and the maximum area and minimum area associated with the identified action; an indicator generator, which is used to generate heart function-related parameters for each area trace based on the identified features, amplitude and timing, and the heart function-related parameters include one or more of the following items: maximum vascular area (A 最大 ), minimum vascular area (A 最小 ), mean vascular area (A 平均 ), heart rate (HR), respiratory rate (RR), collapsibility index (CI), collapse (C) and cardiac output (CO); a boundary generator, which is used to generate at least a patient's lower vascular area boundary (LB) or a patient's upper vascular area boundary (UB) using the generated cardiac function-related parameters; and an index generator, which is used to set a patient congestion index based on one or more of the generated cardiac function-related parameters and at least one of the patient's lower vascular area boundary (LB) or upper vascular area boundary (UB), the congestion index indicating the patient's fluid state on a normalized scale for each trace cycle.

[0017] In another embodiment, the present disclosure relates to a system for automatically determining a patient's fluid status using periodic vascular area traces. The system includes: a trace feature detector for identifying selected features of the vascular area trace and at least one of the amplitude and timing of the identified features of the vascular area trace; an indicator generator for generating cardiac function-related parameters for each area trace based on the identified amplitude and timing, the cardiac function-related parameters including at least the maximum vascular area (A 最大 ) and minimum vascular area (A 最小 ); and an index generator for determining a maximum vessel area (A 最大 ), minimum vascular area (A 最小 ) and at least one of the patient's lower vascular area boundary (LB) or upper vascular area boundary (UB) to generate a patient congestion index that indicates the patient's fluid state on a normalized scale for each trace cycle.

[0018] In yet another embodiment, the present disclosure is directed to a computer-based method. The method includes: receiving a patient's calm respiratory vascular area trace in at least one processing device; receiving patient-specific information including at least patient weight and patient age in the processing device; filtering the calm respiratory vascular area trace at the processing device to identify component signals including at least a respiratory trace, a cardiac trace and an average trace; extracting amplitude features and timing features corresponding to at least one or more of area maximum, area minimum, collapse, respiratory collapse, cardiac collapse, heart rate and respiratory rate from the trace at the processing device; generating cardiac function-related parameters at the processing device using executable program instructions, the executable program instructions defining cardiac function-related parameters based on the extracted amplitude features and timing features, the cardiac function-related parameters including one or more of the following: respiratory rate, cardiac output, heart rate, collapsibility index, collapse, maximum vascular area, minimum vascular area and average vascular area; generating a patient congestion index based on one or more of the amplitude features and the cardiac function-related parameters; and using a machine learning-based model, applying weights to the congestion index, the cardiac function-related parameters and the patient-specific information to return a hospitalization probability score for the patient.

[0019] In another embodiment, the present disclosure relates to a method for automatically determining a patient's fluid state using a periodic vascular area trace. The method includes: receiving a vascular area trace; for the vascular area trace, identifying a selected feature and at least one of the amplitude and timing of the identified feature; based on the identified amplitude and timing, generating a cardiac function-related parameter for each area trace, the cardiac function-related parameter including a measured vascular area and a vascular area boundary; generating a patient congestion index, the patient congestion index representing a relationship between the measured vascular area determined for the area trace and the patient's vascular area boundary, the congestion index indicating the patient's fluid state on a normalized scale for each trace period. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] For the purpose of illustrating the present disclosure, the accompanying drawings show several aspects of one or more embodiments of the present disclosure. However, it should be understood that the present disclosure is not limited to the precise arrangements and instrumentalities shown in the accompanying drawings, wherein:

[0022] Figure 1 is a high-level schematic diagram of a system encompassed by the present disclosure.

[0023] Figure 2 is an example of an IVC area trace, including a detailed view of a portion of the trace with some key trace features identified.

[0024] Figure 3 is a block diagram of an embodiment of an overall system according to the present disclosure.

[0025] Figure 4 is a block diagram of an embodiment of a trace generator according to the present disclosure.

[0026] Figure 5 is a block diagram of an embodiment of a feature detector according to the present disclosure.

[0027] Figure 6 is a series of area trace plots showing the area response to different patient motions.

[0028] Fig. 6A Yes Figure 6 An enlarged detailed view of the area trace of the supine breath-hold maneuver is shown.

[0029] Figure 7 is a block diagram of an embodiment of a metric generator according to the present disclosure.

[0030] Fig. 7A is a block diagram illustrating an embodiment of components of an indicator generator for extracting breathing frequency from an area trace.

[0031] Figure 7B An example of a spectrogram generated from an area trace is included, which shows that in an alternative embodiment another component of the indicator generator extracts heart rate and breathing rate from the area trace.

[0032] Figure 7C is an area trace illustrating that another component of the indicator generator in an alternative embodiment determines an area trace feature.

[0033] Figure 7 D is an area trace showing the components of the indicator generator determining the collapsibility index in an alternative embodiment.

[0034] Figure 8 is a block diagram of an embodiment of a boundary generator according to the present disclosure.

[0035] Fig. 9 Area trace plots for normal breathing and an inspiratory breath-hold maneuver are shown, and an example of using the maneuver to identify the upper boundary is shown.

[0036] Fig.10 is a block diagram of an embodiment of an index generator according to the present disclosure.

[0037] Fig.11 is a block diagram of an embodiment of decision logic according to the present disclosure.

[0038] Fig.12 is a block diagram showing an embodiment of an interface device according to the present disclosure.

[0039] Fig.13 is a flow chart illustrating an embodiment of an overall system flow according to the present disclosure.

[0040] Fig.14 is a block diagram illustrating components of an exemplary computing device.

[0041] Fig.15 An example of deriving the IVC axis based on area and collapse parameters according to embodiments disclosed herein is shown.

[0042] Fig.16 An example of estimating venous return or cardiac output based on changes in blood vessel area according to another embodiment disclosed herein is shown.

[0043] Fig.17 An example of signal traces used in determining cardiac output according to an embodiment disclosed herein is shown, wherein trace (a) is a measured area trace, trace (b) is a filtered respiratory component trace, and trace (c) is a filtered cardiac component trace.

[0044] Fig.18 is a flow chart illustrating another alternative embodiment for cardiac output estimation based on sensor data.

[0045] Fig.19 Another example of signal traces used in determining cardiac output according to an embodiment disclosed herein is shown, wherein trace (a) is a measured area trace, trace (b) is a simulated respiratory component trace, and trace (c) is a simulated cardiac component trace.

[0046] FIG. 20A to FIG. 20D An example of the use of systemic vascular resistance based on sensor data and blood pressure data according to an embodiment disclosed herein is shown, which shows the trend of heart failure patients experiencing symptoms over days, wherein: Fig. 20A The SVR is shown. Fig. 20B The mean IVC area is shown. Fig. 20C shows the respiratory rate, and Fig.20D Daily diuretic administration is shown.

[0047] Detailed Description

[0048] Figure 1 At a high level, a system 10 for receiving and analyzing hemodynamic sensor traces to help provide more accurate diagnosis and improved treatment of heart-related conditions, particularly conditions in heart failure patients, is depicted. In general, a system according to the embodiments described herein will include a sensor subsystem that generates a signal representative of the patient's hemodynamic function, a data analysis subsystem that receives the signal data from the sensor subsystem, processes the received data to automatically generate a diagnosis and treatment recommendation, and an interface subsystem that allows the patient and medical care providers to interact with the system. Figure 1The system 10 shown illustrates one such embodiment, and the system 10 includes an implantable sensor 12 and a processing device 14 worn by the patient, which receives the raw sensor signal and provides initial signal processing. The processing device 14 communicates with another processing platform 16 via a wireless communication link 18. Between the processing device 14 and the processing platform 16, the system 10 reads and interprets the cardiac health status information contained in the sensor signal to provide diagnosis and treatment recommendations based on the extracted information. Data communication can optionally be facilitated by a patient personal device 20 such as a phone or tablet, which can also be used as an input and output device for the patient as further described herein. An optional medical care provider device 22 can also be provided to facilitate interaction between the medical care provider and the patient and the system as further described herein.

[0049] The sensor 12 may include an external sensor system or an implantable sensor system. Examples of the sensor 12 include a vascular size sensor (such as an IVC area or diameter sensor) and a vascular pressure sensor. With respect to the vascular size sensor, a variety of different sensor types may be used to generate the area trace signal, including, for example, an implantable variable inductance coil and an implantable or external ultrasound device. In one specific embodiment, the sensor 12 is an implantable wireless resonant circuit sensor, and the processing device 14 includes a strip antenna, for example, as described in the incorporated USP 10,806,352. As described in the aforementioned incorporated patents and published applications, other sensor types may be employed, such as implantable or external ultrasound sensors and implantable resistance-based sensors.

[0050] Based on the specific configuration of the system according to the present disclosure, the communication link 18 can be wired, wireless, or a combination of wired and wireless. Those skilled in the art can configure the communication link 18 with an appropriate data transmission protocol selected from many available standards. The communication link 18 is preferably a two-way communication link. For example, a personal area network (PAN) connection such as Bluetooth can connect the sensor processing device 14 to the patient's personal device 20. In some embodiments, the personal device 20 can include one or more software applications to perform signal processing functions for sensor signals. In other embodiments, the personal device 20 can act only as an edge device in the system to facilitate communication with a processing platform (16) configured as a cloud platform, wherein the communication occurs via a cellular data link. The communication link 18 between different system platforms and components can also include an Internet connection to provide, for example, data transfer between a cloud platform including the processing platform 16 and a medical care provider device 22. In some embodiments, all or part of the functions of the processing device 14 can also be performed as a cloud-based computing device.

[0051] Figure 2An example of an IVC area signal trace generated by system 10 is shown in FIG. Generally, as used herein, an area trace refers to a signal that represents the area of ​​a vessel over time for a specific discrete time period corresponding to the trace readings and contains data representing cardiac and respiratory function and other characteristics. Processing platform 16 (such as diagnostic engine 102 described below) extracts this data by interpreting features of the area trace. Some features of the area trace are described in Figure 2 As shown therein, feature (a) represents the interval time (t i,resp ). Feature (b) represents the area amplitude of respiratory regulation. Feature (c) represents the interval time of each cardiac cycle (t i,card ). Feature (d) indicates the area amplitude of cardiac regulation in the IVC. Feature (e) indicates the main cardiac peak. Feature (f) indicates the second cardiac peak of the same cardiac cycle as the previous peak (e). Feature (g) indicates the respiratory-related area reduction of the IVC.

[0052] Based on the teachings of this disclosure, ordinary technicians in this field can derive many different ways to configure and implement Figure 1 A system 10. An example embodiment is Figure 3 100 is illustrated in FIG. 100. As shown therein, the sensor subsystem of the system 100 includes: an area trace generator 104; a data analysis subsystem including a diagnostic engine 102, the diagnostic engine 102 including modules such as a feature detector 106, an indicator generator 108, a boundary generator 110, an index generator 112, and a decision logic 114; and an interface subsystem including one or more interface devices 116 to facilitate user interaction with the system, including uploading patient-specific data and receiving reports and notifications about the patient's cardiac health status. The diagnostic engine 102 can also communicate with one or more databases 118 and can receive additional patient-related information from other patient sensors 120 as external input.

[0053] The trace generator 104 is a combined hardware and processing device including, for example, Figure 4 The data acquisition device 200 and signal processing device 202 are depicted. The data acquisition device 200 may include one or more of the aforementioned sensors 12. The signal processing device 202 is a signal amplifier / processor configured to receive the raw data signal from the data acquisition device 200 and generate a readable area trace signal suitable for communication and further processing within the diagnostic engine 102. Figure 4An example of a signal processing device is shown in . In other examples, the signal processing device 202 may include a software application executed on a patient's personal device or a cloud platform. Examples of area traces generated in the signal processing device 202 for communicating with the feature detector 106 include a conventional breathing trace 204 and a motion trace 206.

[0054] Feature detector 106 is a device, such as circuitry, software, or a combination thereof, that extracts relevant data from the area trace signal. The extracted data will typically include the entire reporting period (e.g. Figure 2 The amplitude and timing of selected features identified in the area trace in FIG. 1 (shown in the 60 s trace in FIG. 1 ); that is, multiples of: the interval time of each respiratory cycle, the area amplitude adjusted by respiration, the interval time of each cardiac cycle, the area amplitude adjusted by cardiac movement, the primary cardiac peak, the secondary cardiac peak, and the respiration-related area reduction. Figure 5 As shown, functional components commonly understood and configurable by those skilled in the art for this purpose may include input signal integrity checking, envelope detection 302, region of interest (ROI) detection 304, peak detection 306, and frequency analyzer 308. In some embodiments, feature detector 106 may also be an artificial intelligence driven pattern recognizer 310. Pattern recognizer 310 employs machine learning and pattern analysis techniques for identifying specific patterns in area traces that may be relevant in normalization and in diagnosis and treatment determination, as discussed further below.

[0055] In one embodiment, pattern recognizer 310 compares the incoming area trace signal to known area trace patterns to determine whether the incoming area trace reflects a feature such as a signal response to a patient action. As used herein, "action" refers to a physical action taken by the patient, either actively or in response to a command, that stimulates an identifiable perturbation of the IVC area. Figure 6 Some examples of area traces for different patient actions are shown in Figure 1, including Supine: Quiet Breathing, Sniffing, Supine: PLR (Passive Leg Raise), Supine: Breath Holding, Sitting, and Sitting to Standing. Fig. 6A An enlarged view of the area trace pattern for supine breath holding is shown in FIG. The area trace patterns representing different actions may be stored in a database such as database 118 ( Figure 3) and accessed by the pattern recognizer 310. In one embodiment, this is achieved by converting the trace signal into a bag of words and applying the SFA Bag of Symbols (BOSS) model. The model is then trained to detect motion-related features from such a bag of words in order to classify action / no action with better than 80% accuracy. As further described below, the boundary generator 110 can exploit knowledge of the types of trace patterns. In another embodiment, the pattern recognizer 310 can be used to identify features within the area trace signal that have been shown to predict clinical events (e.g., such as atrial fibrillation or tricuspid regurgitation).

[0056] The feature detector 106 may also perform a data integrity check. Features of the area trace signal may be used to confirm that the system has been used correctly. For example, the quality check may be trained based on supervised data, e.g., prescribed action types. Each area trace may then be quality checked using a model that predicts action types or artifacts to avoid deriving error-inducing information, such as excessive movement during recording, insufficient number or quality of actions performed, etc. Auxiliary sensors (e.g., accelerometers, blood pressure, weight, activity monitors), patient inputs may also be used to facilitate data integrity checks.

[0057] The data extracted from the area trace signal by the feature detector 106 is provided to the indicator generator 108. The indicator generator 108 is a software-based machine configured with a number of different modules, such as Figure 7 As shown, these modules generate specific cardiac-related indicators, on which cardiac-related diagnostic and therapeutic determinations can be based. Included in the frequency-derived indicator group 311 are at least a heart rate module 312 and a respiratory rate module 314. Based on the information received from the feature detector 106, the heart rate module 312 determines the heart rate (HR) based on the following relationship:

[0058] [1]HR=1 / (t i,card )

[0059] where t i,card is the cardiac cycle interval, which can be used Figure 2 The mean or median of the feature (c).

[0060] Likewise, based on information from feature detector 106, respiratory rate module 314 calculates respiratory rate (RR) based on the following relationship:

[0061] [2]RR=1 / (t i,resp )

[0062] where t i,resp is the breathing cycle interval, which can be used Figure 2 The mean or median of the feature (a).

[0063] Fig. 7A An alternative embodiment for determining breathing rate by extraction from the area trace is shown in FIG. 1 and discussed in more detail below.

[0064] Included within the area derived indicator group 318 are at least three area determination modules and a collapsibility index module 326. Additional optional modules may include modules for determining collapsibility 328 and cardiac output 329. Maximum IVC area (A 最大 ) module 320 can be based on the maximum main cardiac peak value ( Figure 2 The value of feature (e) in A is determined by 最大 . Minimum IVC area (A 最小 ) module 322 may reduce the valley ( Figure 2 The value of feature (g) in A is determined by 最小 Alternatively, by finding the global maximum and minimum values, excluding trace segments that are anomalous in terms of the overall trace / excluding area segments of the trace that may have been erroneous due to artifacts, the maximum and minimum area values ​​are extracted from the trace. An example of this technique is shown in Figure 7C The mean IVC area (A 平均 ) module 324 determines A based on the maximum area and the minimum area 平均 (A 最大 and A 最小 half of the sum of Figure 7 As shown in D.

[0065] The collapsibility index (CI) module 326 of the indicator generator 108 uses the determined area parameter to determine the collapsibility index of the IVC based on the following relationship:

[0066] [3]CI=(A 最大 -A 最小 ) / A 最大 *100%

[0067] In another alternative embodiment, collapse is included separately via the collapse module 328 (in Figure 7 The area trace in D is marked), where the collapse is determined based on the following relationship:

[0068] [4] Collapse = A 最大 -A 最小

[0069] Therefore, the collapsibility index can also be expressed as:

[0070] [5]CI=collapse / A 最大 *100%

[0071] For example, respiratory collapse can be determined from the area trace. This has the potential advantage that it is another somewhat independent signal that can be used to predict volume or congestion state or pressure - low collapse at high volume / pressure, high collapse at normal blood volume / normal pressure, and low collapse at low blood volume / low pressure, giving an "n" shaped collapse relative to area curve. Trending of features of the original trace or action trace may also prove useful, i.e., an accelerated increase in area may require more urgent or more drastic action than a gradual increase. In another example, other frequency-based signals may also be extracted and used as input to the calculation. As another example, it is known that the IVC area signal changes with each breath of the patient, and therefore the low frequency oscillations of the IVC can be used to extract the respiratory rate. The respiratory rate has been shown to be a prediction of heart failure status, and therefore will be a powerful input for overall patient state estimation. In addition, an indicator of the combined frequency and area input 316 can be determined.

[0072] Inputs from wearable devices such as activity trackers can also predict the effectiveness of treatment for heart patients, and reduced activity can also be integrated into the calculation as a predictor of worsening status. In addition, weight is a factor that can be used to predict decompensation of heart failure, for example, a weight gain of more than 2 kg in 2 days can be considered significant. The patient's daily weight can also be integrated into the calculation. Sleep, activity, heart rate variability, blood pressure and other wearable outputs can also be integrated into the congestion index. In addition, data from implantable sensors and their entire system 100 can be integrated into the system of wearable devices as a source of additional data for predicting and / or monitoring health status.

[0073] Included within the other input derived metrics group 332 are metrics derived from other external sensors (meaning external sensors relative to the system 100, which may include in vivo sensors and off-body sensors), such as pulse oximetry, body temperature, blood pressure, urine output, cardiac output, and catheter pressure, among others. Patient-specific information 334 typically includes information about patient physiological parameters (such as height, age, weight, gender), and may also include current activity information entered by the patient or care provider through a user interface. Another alternative external sensor input is an accelerometer reading from a patient-worn component of the sensing system (such as a patient-worn processing device 14), which may be embodied as an antenna strap as described in the incorporated patent disclosure. Such accelerometer readings may be used to determine patient position and activity / motion during the trace period, and thereby increase the accuracy of the metrics derived from the trace.

[0074] like Figure 3 and Fig.11As shown, the indicators from the indicator generator 108 can also be directly provided to the decision logic 114 for application according to a specific diagnostic or treatment algorithm. The indicators can also be provided to the interface device 116 for display and user monitoring. The area-based indicators are also provided to the boundary generator 110 for boundary determination.

[0075] To overcome the calibration challenges created by intra- and inter-individual variations caused by heterogeneity in vessel shape as mentioned above, embodiments of the present disclosure generate specific reference points to which periodic IVC area readings can be compared to assess current patient fluid status and associated cardiac health. In one example, Figure 8 The boundary generator 110 shown uses IVC area data extracted or predicted from the area trace to determine upper and lower boundaries for this purpose. Boundaries can be considered static or dynamic. Static boundaries are maximum / absolute values ​​that are not expected to change significantly over a long period of time. The absolute maximum / minimum area values ​​representing the upper and lower edges of a single IVC area range can be considered static boundaries. The absolute maximum area can be related to the total circulating blood volume. It can be assumed that these boundaries are anatomical constants and that the IVC of a particular patient can always only reach a specific size. In this way, the boundaries can be considered static over time. This allows a single action to be used to understand the patient's IVC range and use these boundaries over time. This essentially means performing a single calibration of the feature. Static boundaries can be used, for example, in association with a specific set of conditions to be diagnosed.

[0076] Dynamic boundaries, on the other hand, may change over shorter time periods. Dynamic boundaries do not necessarily indicate absolute anatomical limits, but rather represent current limits resulting from changing physiological parameters such as venous tone and / or intra-abdominal pressure. Thus, dynamic boundaries may represent new and clinically relevant limits with each quiet breathing reading, or occasionally when an estimate is available by motion or other means, to use information that is closest in time to the reading for volume status assessment. In other words, based on the overall patient fluid state in terms of total fluid distribution between vascular and extravascular fluid, the dynamic boundaries as defined herein may represent a more clinically relevant basis for assessing the patient's fluid state at a particular reading. In addition, the dynamic maximum area boundary may be related to right atrial pressure or vascular tone.

[0077] like Figure 8As shown, the boundary generator 110 includes two components for generating boundaries, a lower boundary determination module 336 and an upper boundary determination module 340. In various embodiments, the lower boundary determination module 336 and the upper boundary determination module 340 can be alternatively configured to determine the lower boundary and the upper boundary based on motion, or to predict the boundary based on regular breathing. In another alternative, the lower boundary determination module 336 and the upper boundary determination module 340 can include motion determination submodules 338, 350 and prediction submodules 342a, 342b. Fig. 9 The operation of the action determination submodules 338, 350 is shown, showing the area trace of the inspiratory breath-hold action overlaid on the conventional respiratory area trace. The lower boundary determination 336 takes into account any physiological or device-related limitations that prevent complete collapse of the IVC, such as sensor radial force, sensor positioning across venous branches, or other physical limiting features of the anatomical structure or implantable device. The submodule 350 can receive information about the type of action in a variety of ways. For embodiments that employ a feature detector 106 including a trajectory pattern recognizer 310, the system can directly provide the action type information. Alternatively, when instructing the patient to perform an action, a user such as a care provider will also input instructions to perform the action via the interface device 116. The trace pattern recognizer 310 (where present) can also be used to confirm the patient's compliance with the action instruction by comparing the received area trace pattern with the stored pattern for the indicated action. Regardless of the source, when informed of the performance of the action, the action determination submodule 350 will be the maximum area ( Fig. 9 Point 349 in is identified as the upper boundary.

[0078] In some cases, it may not be practical or desirable to require the patient to perform an action. The prediction submodule 342a is configured to predict the upper boundary (UB) based on the normal breathing area trace based on the following relationship developed by the applicant:

[0079] [6]UB=(-1 / slope)·CI+A 平均

[0080] In an alternative embodiment, the “slope” used in equation [6] is −0.18% / mm as identified in the literature: 2 Constant reference slope of: Huguet et al., Three-Dimensional Inferior Vena Cava for Assessing Central Venous Pressure in Patients withCardiogenicShock, J Am Soc Echocardiogr. 2018; 31: 1034-43 (https: / / doi.org / 10.1016 / j.echo.2018.04.003), which is incorporated herein by reference in its entirety. The slope used can be retrieved from the data storage 346. In other alternative embodiments, the slope can be personalized based on the area and collapse data of each patient, or can be trained based on a collection of area and collapse or other data over a period of time. Additional alternative techniques 344 for upper boundary generation can also be derived based on the teachings of the present disclosure.

[0081] In some embodiments, the lower boundary (LB) determination module 336 also employs the action determination submodule 338 to identify the lower boundary as the minimum area point in the action area trace in substantially the same manner as the upper boundary determination submodule 350 identifies the upper boundary. Fig. 9 351 in). For example, the lower boundary determination can be performed using the above method via a sniffing action or any other action that reduces the size of the IVC. Alternatively, the lower boundary prediction 342b uses individual patient or population area and collapse data to predict the lower boundary using a method similar to the upper boundary prediction 342a using equation [7]:

[0082] [7]LB = (-1 / slope)·C+ / A 平均

[0083] A further alternative submodule 339 may also be provided to derive a lower boundary value on a population basis as an average minimum sensor size measured in the IVC, or sensor information regarding the area of ​​the IVC at a given equivalent pressure, for example by comparing a sensor-derived area / pressure curve to a standard area / pressure curve (determined based on radial or plate force information) and which is a programmed constant for all patients.

[0084] The upper and lower boundaries generated by boundary generator 110 are provided to index generator 112, such as Fig.10As shown. These boundary values ​​enable the volume measurement of the vessel to be normalized to return a number ranging from 0% to 100% or 0 to 10, where 0% is close to 0 mmHg internal filling pressure or low volume / hypocongestion / hypovolemia, and 100% is close to the maximum possible filling pressure in the vessel (~20 mmHg) or high volume / hypercongestion / hypervolemia. The index generator 112 assigns the boundaries to a relative scale, where the lower boundary is set to the scale minimum and the upper boundary is set to the scale maximum. In some embodiments, the relative scale is established as a 0 to 100% scale. Once the relative scale is established, the current area reading is scored on the relative scale, for example, an area reading that falls halfway between the upper and lower boundaries will be scored as 50%. Because each patient can display a range of IVC areas based on multiple factors including, but not limited to, anatomical size, current fluid status, body position, vascular tone status, and cardiac and abdominal pressures, normalization as described herein takes these parameters into account to produce a more standardized congestion index that is comparable over time for patients and between patients to help facilitate more accurate diagnosis and treatment of heart failure-related conditions. As described above, the new IVC area-based metrics disclosed herein may not be well understood initially, and correlation with well-known physiological measurements may aid clinical understanding by early adopters and ultimately aid clinical quality. In one embodiment, the congestion index may be determined based on the relationship of the average area to the upper and lower boundaries as in Equation [8]:

[0085] [8] Congestion index = 100*((A 平均 -LB) / (UB-LB))

[0086] In another example embodiment, the index generator 112 uses the lower boundary determined from the boundary generator 110 with the collapsibility index 326 to generate a vascular congestion index using the following equation:

[0087] [9] Congestion index = (A 最大 -A 最小 ) / (A 最大 -LB)*100%

[0088] An alternative to this equation could be to subtract from 100 in order to reverse the direction of the exponent, resulting in the following equation:

[0089]

[10] Congestion index = 100-(A 最大 -A 最小 ) / (A 最大 -LB)*100%

[0090] In yet another alternative, the hyperemia index may be determined using only the upper boundary as the ratio of the mean area to the upper boundary, as follows:

[0091]

[11] Congestion index = 100*(A 平均 / UB).

[0092] In another alternative embodiment, the index generator 112 optionally receives the index from the index generator 108 and the boundary generator 110 and factors it into a normalized IVC congestion index. In such an embodiment, the index generator 112 can compress multiple features into a single index. For example, the index generator 112 can be configured to output a single number within a standardized range for each patient. This single digital output can be an indicator of a measurable variable, such as RAP, worsening heart failure, impending hospitalization, probability of an impending event, a reminder to change medication, etc. This output will be common to all patients and contain or summarize information from all inputs to provide an actionable output for the clinician. In some ways, this is comparable to a scale such as a temperature of 37 degrees Celsius as an acceptable normal level and acceptable high and low thresholds. In this configuration, the index generator 112 can output a congestion index ranging from 0 to 100%, which indicates the risk of impending heart failure hospitalization within a selected time period (e.g., the next thirty days).

[0093] The IVC congestion index generated by the index generator 112 may be reported directly by the user interface device 116, but may also be a valuable input for the decision logic 114, such as Fig.11 As shown. The decision logic 114 includes one or more instruction sets for determining the patient's cardiac health status and treatment recommendations based on the cardiac function parameters and congestion index generated by the indicator generator 108. The decision logic 114 can be embodied in one or more circuits, executed by one or more data processing devices, or distributed on a combination of such platforms. In general, the congestion index, optionally together with other indicators from the indicator generator 108, is an optional input to various diagnostic and treatment algorithms 358, 360 executed in the decision logic 114. In one example, the threshold value can be set as follows:

[0094] 0 to 30%: Below the normal range, which may indicate minimal risk of hospitalization for heart failure due to volume overload,

[0095] However, there is a risk of hypovolemia, and patient data require attention and review, potentially reducing diuresis,

[0096] 30% to 70%: Normal range, potentially optimizing guideline-directed medical therapy,

[0097] 70% to 100%: Above the normal range, this may indicate an elevated risk for impending heart failure hospitalization and the patient's data needs attention and review, potentially increasing diuresis.

[0098] In this way, any one feature of the daily congestion index with the selected cut-off thresholds (ie, lower and upper limits) is now scaled to the final congestion index as follows:

[0099] (1) The low range from 0 to the lower threshold will be scaled from 0 to 30%

[0100] (2) The normal range from the lower threshold to the upper threshold will be scaled from 30% to 70%

[0101] (3) The high range from the upper threshold to 100% will be scaled from 70% to 100%.

[0102] As an example, utilizing the methods outlined above, a priority list algorithm may be created for moving patients to a higher priority state after crossing a threshold for a defined number of days, and for removing patients when within a normal range for a defined number of consecutive days. In addition, the primarily area-based congestion index described above may be enhanced by factoring in additional cardiac function-related parameters as further described below to provide a potentially more comprehensive picture of the patient's cardiac health status. Other examples of such algorithms are disclosed in the aforementioned and incorporated USP 11,564,596, and alternative embodiments may inform the up- or down-titration of specific medications (such as diuretics, vasodilators, or other medications) based on a pre-specified patient-specific prescription. Additional examples of conditions for which treatment recommendations may be diagnosed and generated via algorithms 358 and 360 include conditions such as tricuspid regurgitation, atrial / ventricular tachycardia / bradycardia.

[0103] In other embodiments, motion or similar changes may be utilized to move the patient between two or more different states. The signal differences between these states may be indicators of different physiological conditions. An example of this is in the case of a HFpEF patient, where a deterioration in the condition parameters is only evident upon movement. In this case, the no-motion state and the motion state may be used to provide signals to diagnose the presence of HFpEF. Prone breathing is another example, where motion may result in shortness of breath, and this may also be evidenced by signals indicating changes in respiratory rate / amplitude. These motions may be detected based on signals in the feature detector 106, or may be inputs from the interface device 116.

[0104] The decision logic 114 can utilize any number of indicators as inputs. In some cases, only a single indicator or a combination of related indicators, such as an area indicator, may be used. In an alternative embodiment, the indicator weighting submodule 354 applies a weighting factor to one or more of the single indicator inputs. The weighting factor can be determined via analysis of clinical and preclinical data and is pre-specified, or can be continuously calculated and updated based on user-specific inputs. In this case, the notification can include instructions to take therapeutic actions, instructions to take additional readings, or instructions to alert the care provider of possible clinical problem situations.

[0105] like Fig.12 As shown, it is expected that the interface device 116 will include at least a user interface 402 that can be accessed by a care provider, for example via a care provider device 400, for data input 404, such as patient-specific information, prescription details, user settings, action instructions, algorithm updates and other system hygiene. The user interface 402 can also provide notifications generated by the decision logic 114 or other components of the diagnostic engine 102, and can be a web-based or mobile application. In some embodiments, the interface device 116 may include one or more processors 406 and associated processing capabilities in the form of hardware / software. Diagnostic or treatment reminders 408 can also be delivered via the care provider device 400, for example as described in the aforementioned and incorporated USP 11,564,596. The interface device 116 can also include a patient personal device 410, which can be used for various functions in the system in addition to patient notifications, for example, Figure 1 shown.

[0106] It is also possible to use a continuous process flow diagram (such as Fig.13 ) to describe embodiments of the present disclosure, or alternative embodiments of the system may have versions of this form. As shown therein, different data types call different processes. For example, data 504 representing a sensor time series of quiet breathing calls process 506 for filtering (respiration trace, heart trace, average trace, noise trace), and calls process 508 for feature extraction (A 最大 , A 最小, collapse, CI, respiratory collapse, cardiac collapse, heart rate, respiratory rate). Similarly, data 512 representing sensor time series of motion calls process 514 to filter and transform using BOSS (Symbolic Fourier Approximation (SFA) - Bag of Symbols), and calls process 516 to evaluate the motion trace using features from BOSS and features from other common descriptive statistics to evaluate trace quality and extract the maximum area achieved. External data 518 such as blood pressure, weight, blood oxygen, etc. calls process 520 to incorporate such data representing other externally sensed features from other sensors or direct input. The output of these processes can be stored in database 510.

[0107] Features output from database 510 (or directly from an upstream process) invoke process 522 to generate indices as described herein, including, for example, congestion index 524, right atrial pressure (RAP) 526, respiratory rate (RR) 528, cardiac output (CO) 530, systemic vascular resistance (SVR) 532, IVC tone 534, heart rate (HR) 536, patient weight 538, patient activity 540, and other indices described herein. The generated indices are then processed 542 to produce an IVC congestion index score 544 based on or enhanced from all of the indices, which is presented to the user or used as an input to decision logic 114 as a basis for further diagnostic or treatment decisions.

[0108] In some embodiments, the enhanced index process 542 is configured as an AI or machine learning based model that is trained based on all of the described inputs and determines the optimal weights for each indicator in order to predict a specific output, such as "worsening of heart failure," "increased congestion," "likelihood of hospitalization," etc. One such example would be a multivariate logistic regression model that uses a training data set to develop a model that returns the probability of hospitalization based on inputs 524-540 or a subset of inputs 524-540. Another embodiment would be a decision tree model that segments the data from each input to be classified as, for example, "high likelihood of hospitalization" or "unlikely to be hospitalized" based on the training set data, repeating the process for each input parameter until a model is created to accurately predict the likelihood of hospitalization.

[0109] Reference below FIG. 7A to FIG. 7B and Figures 15 to 20A to Fig.20D Additional alternative embodiments are described for generating various indicators within the components of the indicator generator 108. For example, additional alternative embodiments of the respiratory rate module 314 of the indicator generator 108 provide for generating various indicators from the components of the indicator generator 108. Fig. 7AThe area trace shown extracts the respiratory rate. As shown therein, the respiratory rate module 314A receives the area trace as input 375. Then, a Butterworth bandpass filter module 377 is used to remove baseline drift and higher frequency components from the input signal. An example range for this filter is 0.08 Hz to 0.75 Hz. The signal is then segmented into a number of windows with a specific lag module 379, an example would be five hundred and twelve (512) windows with a four (4) second lag. The mean is then subtracted from each window module 381. These windows from 381 are then used to estimate the power spectral density. This is done using the Welch method to account for possible non-stationarity in the signal, i.e., variations in the respiratory rate module 383. For example, this can use a Hann window with a sample size of two hundred and fifty-six (256). The power spectral density from module 383 is then filtered to examine only possible respiratory rate modules 385, such as eight (8) to forty (40) times per minute. Then, in module 387, all peaks in the filtered spectrum are found, for example, using peak prominence. In module 389, the maximum detected peak is extracted. Then, the signal quality of the peak is evaluated by comparing the signal quality of the peak with the signal quality outside the range excluding the potential harmonics of the peak. If the signal quality is greater than the threshold in module 391, the number of breaths per minute estimated by the maximum peak is accepted for the window, otherwise the respiratory frequency for the window is rejected in module 393. Modules 383, 385, 387, 389, 391 and 393 are repeated on all windows given by 381. Module 395 takes all these respiratory frequencies in breaths per minute as input. If the number of respiratory frequencies is greater than the threshold, the average of the accepted respiratory frequencies is returned as the global respiratory frequency of input 375. The final output 397 of the respiratory frequency module 314A returns a separate value for each of the respiratory frequency (RR), signal quality (SQ) and respiratory collapsibility index derived for each area trace input.

[0110] As yet another indicator example, each beat of the heart also perturbs the IVC, and this higher frequency signal can also be extracted from the IVC area trace by spectrally decomposing the area trace using a Fast Fourier Transform (FFT), as shown in Figure 7B As shown. It is well known that heart rate is also a useful indicator in patient management, and this can be included in the generated indicators. An increased or decreased rate or increased rate variability can be a predictor of worsening treatment effect. In addition, pacing related indicators can be determined from the area trace.

[0111] In another example, if Fig.15As shown, assuming an idealized "n"-shaped curve of IVC area relative to IVC collapse, the ellipticity of the IVC can be estimated (thus providing an approximation of the IVC diameter if existing diameter-based clinical guidelines are used for thresholds). Furthermore, it can be assumed that the IVC shape varies from a flat ellipse to a perfect circle along this "n" curve, with the axial ratio correspondingly varying from low (~0.1) to equal (1) along the curve. Therefore, knowledge of the collapse and area allows the position on the "n" shape to be approximately determined, which in turn allows the axial ratio of the IVC to be estimated. The approximate axial ratio can in turn be inserted into the area equation for the ellipse, reducing the number of unknown variables to 1, thereby allowing an exact solution for the major and minor diameters of the IVC (for comparison with existing clinical guidelines).

[0112] In another embodiment, the area changes observed in the IVC are used to estimate venous return (VR) / cardiac output (CO) by measuring the area change per unit time as the integral under the curve (AUC) and a simple initial calibration to clinical cardiac output measurements. This measurement is based on the volume that crosses the IVC causing the IVC to temporarily expand relative to its minimum size. Fig.16 This distortion is shown in an idealized cross-sectional depiction of an IVC with (b) and (d), which correspond to the area trace ( Figure 2 ), identifying the circle and the area shift. In this example, cardiac output can be determined using the following equation

[12] :

[0113]

[12] CO=VR=corfacresp(Rcol*RR)+corfaccard(Hcol*HR)

[0114] The correction factors (corfacresp and corfaccard) are derived from the reference cardiac output measured by the calibration step (e.g., ultrasound for estimating cardiac output). The correction factors can be split into respiratory and cardiac contributing elements, since the distance of the device to the heart can affect the detected cardiac component relative to respiratory regulation. For example, two-thirds of the blood flow into the right atrium originates from the IVC. As blood flows through the IVC, it dilates into a compliant vasculature. Similar to stroke volume estimation based on left atrial images, changes in the cross-sectional area of ​​the IVC are related to cardiac output, which has been reported in the following literature: Barberi et al., 2004 [Intensive Care Med (2004) 30: 1740-1746, DOI 10.1007 / s00134-004-2259-8] (which is incorporated herein by reference in its entirety). The IVC is regulated by respiratory effort and cardiac activity. The concept of relating IVC area changes to cardiac output relies on four main features - area changes with respiration and cardiac activity and with heart rate and respiratory rate. Due to the complexity of venous return / cardiac output - which is a time integral - it is necessary to calibrate the m 2 The area measured in units of 1 / min is integrated over time so that it can be related to the cardiac output measured in liters / minute. To do this, at least one simultaneous cardiac output measurement combined with the recording of the free-breathing IVC sensor area is required. Assuming 0 m 2 and 0 l / min is the second calibration point. A simple linear calibration is possible, ultimately leading to a value in l / min / m 2 The conversion factor f is a unit, so that:

[0115]

[13] CO = fcorr*area-time-integral(sensor).

[0116] In different embodiments, multiple calibration points are created by phlebotomy / blood draw and reinfusion or by injection of a known bolus of saline. When the respiratory regulation of area is related to the cardiac regulation of area in a weighted manner, cardiac output accuracy is expected to improve. In addition, tricuspid regurgitation information can be incorporated to correct for regurgitation related area regulation. Correction factor weighting can be included to adjust for cardiac and respiratory contributions to the observed collapse. This approach requires more than a single reference point for calibration. Measurement of cardiac output during an exciting maneuver (such as an inspiratory or expiratory breath hold) can allow the respiratory component to be separated from its cardiac component - since the cardiac component is only visible during the breath hold.

[0117] like Fig.17As further shown in , the measured area trace and the filtered respiratory and cardiac component traces can be used to calculate the area under the curve (AUC) as an indicator of cardiac output. The change in area in the blood vessels is related to the volume delivered to the right atrium - this is generally referred to as venous return. Venous return is accepted as equal to cardiac output. Venous return consists of blood flow from the superior vena cava and the inferior vena cava. Since the sensor is expected to be located in one of these locations, an initial calibration step is required. Common clinical measurement methods such as ultrasound or thermodilution can provide a reference cardiac output level. In some embodiments, a single cardiac output measurement can be used, or alternatively, a second cardiac output and sensor reading can be obtained by using motion or other methods. Using such a reference and the calculated AUC, a correction factor is established to link the two. The correction factor can be further refined by adding information about the contribution of respiration and cardiac movement to blood flow in the corresponding sensor location, for example, by splitting a single correction factor according to a predetermined ratio applied to cardiac movement and respiration.

[0118] In another embodiment, if Fig.18As shown, cardiac output estimation process 700 can be used to estimate cardiac output directly from sensor data. Input 703 of process 700 includes extracted features related to area under the curve, such as cardiac collapse (Hcol) and respiratory collapse (Rcol), as well as heart rate (HR) and respiratory rate (RR). Using these inputs, the cardiac component of the output is set to the product of heart rate, cardiac collapse and sensor length at 706. Similarly, the respiratory component of the output is set to the product of heart rate, respiratory collapse and sensor length at 709. The sensor length in this process can refer to the longitudinal length of the sensor coil portion of the wireless resonant circuit sensor implant, which is configured and sized to be implanted in a patient's blood vessel in contact with the vessel wall. An example of such a sensor implant includes an expandable and contractible variable inductance coil having a plurality of adjacent strands formed around an open center to allow blood to flow through the open center in a substantially unimpeded manner. In some such embodiments, the coil is constructed and dimensioned to: (i) extend around the inner periphery of the blood vessel when implanted in the blood vessel, and (ii) move with the blood vessel wall in response to expansion and collapse of the blood vessel. The sensor implant also includes a capacitor that forms a variable inductance resonant circuit with the variable inductance coil, the variable inductance resonant circuit having a variable characteristic frequency related to the diameter or area of ​​the expandable and contractible variable inductance coil. In another alternative, the expandable and contractible variable inductance coil includes a plurality of strands, and the inductance varies in response to changes in the expansion and contraction of the coil based on a cross-sectional area or diameter across the center of the coil opening. For example, such a sensor implant is described in more detail in USP 10,806,352, which is incorporated by reference (e.g., RC-WVM implant 12 and the height (A) corresponding to the sensor length described therein). It should be noted that steps 706 and 709 can occur in any order or simultaneously. Then, at 712, the total cardiac output is set to the sum of the two component outputs multiplied by the correction factor. One of ordinary skill in the art can use the examples of measured cardiac output and the associated respiratory and cardiac components to derive correction factors based on the teachings of the present application.

[0119] like Fig.19 As further shown in (a) to (c) of FIG. 1 , the parameters cardiac collapse (Hcol), respiratory collapse (Rcol), heart rate (HR) and respiratory rate (RR) can be used to synthetically calculate the respiratory trace and the cardiac trace (simulating area adjustment) to again use the AUC method as another alternative means for assessing cardiac output. In this embodiment, the collapse component is used to inform the amplitude of the oscillation, while the rate informs the oscillation frequency. Therefore, Fig.19 The breathing trace in (b) can be simulated as:

[0120]

[14] Respiration trace = Rcol*sin(2*pi*RR*time sample vector),

[0121] where a time sample is a linear range of values ​​starting from a starting value to a final timestamp in discrete steps defined by the sampling frequency. Similarly, the cardiac trace (c) can be modeled as:

[0122]

[15] Heart trace = Hcol*sin(2*pi*HR*timesamplevector).

[0123] In a further alternative embodiment, right atrial pressure (RAP) can be estimated by correlation with area trace data. For example, paired data of IVC area trace and RAP measured by a pressure catheter can be used to create a regression model using individual or population level training data, which can then be used to predict RAP based on input of an area metric generated from the sensor area trace (e.g., area mean). Vascular tone, which refers to the degree of contraction experienced by a blood vessel relative to its maximal dilation state, can also be estimated based on area trace data. For example, the variation of extracted features over time (such as A achieved using motion varying between days) can be used to predict RAP. 最大 ) can indicate changes in IVC tension. Therefore, using the correlation of the trained model with the area trace data can provide an IVC tension estimate.

[0124] In another embodiment, mean arterial pressure (MAP) can be determined using an external blood pressure cuff and can be combined with the aforementioned cardiac output and RAP (directly measured or estimated as described above) to estimate systemic vascular resistance (SVR) using the following equation

[16] :

[0125]

[16] SVR=(MAP-RAP) / CO.

[0126] exist FIG. 20A to FIG. 20D An example of such a calculation based on human sensor data is provided in , which shows trend data over days for heart failure patients experiencing symptoms (dyspnea). Traces of SVR (systemic vascular resistance), area average from the sensor, respiratory rate from the sensor, and daily dose of diuretic (furosemide) are shown. As symptoms worsen (dyspnea / shortness of breath) in mid-March, the area increases logarithmically with increasing respiratory rate. Towards the end of dyspnea, SVR increases dramatically in an exponential manner. As the dose of diuretics increases, the area decreases dramatically over a few days, and SVR initially recovers, followed by a brief increase for a few days until the drug is ideally titrated to promote a reduction in the patient's respiratory rate to normal values.

[0127] The systems and methods described herein may be implemented using one or more computing devices and, unless otherwise indicated, may include standard computing components of a processor, memory, storage, and communication bus, and may include high-resolution graphics rendering hardware and software when necessary. Such components may be configured and programmed by a person of ordinary skill based on the teachings of the present disclosure. Software programming that implements the systems and methods described herein may reside on a non-transitory computer-readable medium as a computer program product. Computing devices typically also include cloud-based computing embodiments and computing systems with elements distributed across a network.

[0128] In various alternative embodiments, by way of non-limiting example, signal processing, data extraction, diagnostic, and treatment-related functions (such as those in trace generator 104, feature detector 106, indicator generator 108, boundary generator 110, index generator 112, decision logic 114, or interface device 116 (including processor 406)) may be performed as one or more computing devices, or may be performed collectively in a single or multiple computing devices. Fig.14 An example of such a computing device is shown, where computing device 600 includes one or more processors 602, memory 604, storage device 606, a high-speed interface 608 connected to memory 604 and a high-speed expansion port 610, and a low-speed interface 612 connected to a low-speed bus 614 and the storage device 606. Each of components 602, 604, 606, 608, 610, and 612 are interconnected using various buses or other suitable connections, such as Figure 8 602 can process instructions for execution within computing device 600, including instructions stored in memory 604 or on storage device 606 to display graphical information via GUI 618 or on an external user interface device coupled to high-speed interface 608 using display 620 coupled to high-speed interface 608. As will be appreciated by those of ordinary skill, some of the described functions may not require an independent GUI or display. In other implementations, multiple processors and / or more buses and multiple memories and memory types may be used appropriately. In addition, multiple computing devices 600 may be connected, each of which provides a portion of the necessary operations (e.g., as a server group, a blade server group, or a multi-processor system).

[0129] The memory 604 stores information within the computing device 600. In one embodiment, the memory 604 is a computer-readable medium. In one implementation, the memory 604 is one or more volatile memory units. In another implementation, the memory 604 is one or more non-volatile memory units.

[0130] The storage device device 606 can provide mass storage for the computing device 600 and can contain information such as timing control, time slice size and / or static color chromaticity and timing, as described above. In one embodiment, the storage device device 606 is a computer readable medium. In various different embodiments, the storage device 606 can be a floppy disk device, a hard disk device, an optical disk device or a tape device, a flash memory or other similar solid-state storage device, or a device array, including a device or other configuration in a storage area network. In one implementation, a computer program product can be specifically embodied in an information carrier. The computer program product includes instructions that, when executed, execute one or more methods such as those described above. The information carrier is a computer or machine readable medium, such as a memory 604, a storage device 606, or a memory on a processor 602.

[0131] The high-speed interface 608 manages bandwidth-intensive operations of the computing device 600, while the low-speed interface 612 manages less bandwidth-intensive operations. This allocation of responsibilities is exemplary only. In one implementation, the high-speed interface 608 is coupled to the memory 604, the display 620 (e.g., through a graphics processor or accelerator), and to a high-speed expansion port 610 that can accept various expansion cards (not shown). In this implementation, the low-speed interface 612 is coupled to the storage device 606 and the low-speed expansion port 614. The low-speed expansion port, which can include various communication ports (e.g., USB, Bluetooth, Ethernet, wireless Ethernet), can be coupled to one or more input / output devices as part of the GUI 618 or as another external user interface, such as a keyboard, a pointing device, a scanner, or a networking device such as a switch or a router, for example, through a network adapter.

[0132] Various implementations of the systems and techniques described herein can be realized in digital electronic circuitry, integrated circuitry, specially designed ASICs (application specific integrated circuits), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include implementations in one or more computer programs executable and / or interpretable on a programmable system that includes at least one programmable processor, which may be special or general purpose, coupled to receive data and instructions from and transmit data and instructions to a storage system, at least one input device, and at least one output device.

[0133] These computer programs (also referred to as programs, software, software applications or code) include machine instructions for a programmable processor and may be implemented in high-level procedural and / or object-oriented programming languages ​​and / or in assembly / machine languages. As used herein, the terms "machine-readable medium", "computer-readable medium" refer to any computer program product, apparatus and / or device (e.g., disk, optical disk, memory, programmable logic device (PLD)) for providing machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal for providing machine instructions and / or data to a programmable processor.

[0134] To provide interaction with a user, the systems and techniques described herein may be implemented on a computer having a display device separate from the video display 620. LED displays are now the most common, however older display technologies (e.g., CRT (cathode ray tube) or LCD (liquid crystal display) monitors) may be used. Other interface devices may include a keyboard and pointing device (e.g., a mouse or trackball) through which a user may provide input to the computer. Other types of devices may also be used to provide interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic, voice, or tactile input).

[0135] The systems and techniques described herein can be implemented in a computing system that includes a back-end component (e.g., as a data server), or includes a middleware component (e.g., an application server), or includes a front-end component (e.g., a client computer with a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described herein), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of wired or wireless digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), and the Internet.

[0136] A computing system may include a client and a server. Generally speaking, the client and the server are remote from each other and typically interact through a communication network. The relationship between the client and the server is generated by computer programs running on respective computers and having a client-server relationship with each other. The processing power and capabilities described herein may also be implemented as cloud-based or other network-based processing modules, and may also be implemented using a software as a service ("SaaS") model.

[0137] Another alternative embodiment of the present disclosure includes a method for creating a patient-specific heart failure diagnostic tool. Such a method may include monitoring a patient physiological parameter associated with a patient fluid volume, the physiological parameter having a measurable value; instructing the patient to perform an action; in response to the patient performing an action, recording a change in the value of the monitored physiological parameter; when the patient performs the action, setting the value of the physiological parameter to the patient's maximum value; determining a minimum value; and setting the maximum value to the upper limit of a standardized scale associated with the risk of adverse heart failure events, and setting the minimum value to the lower limit of the standardized scale to establish a customized heart failure risk assessment tool for the patient. Using this method, the measured value of the monitored physiological parameter is comparable to a standardized scale for determining the level or type of medical intervention.

[0138] In another alternative embodiment, a heart failure diagnostic method includes the following steps: monitoring a patient physiological parameter related to the patient's fluid volume, the physiological parameter having a measurable value; instructing the patient to perform an action; recording a change in the value of the monitored physiological parameter in response to the patient performing the action; setting the value of the physiological parameter to a maximum value for the patient when the patient performs the action; determining a minimum value; setting the maximum value to an upper limit of a standardized scale associated with the risk of adverse heart failure events, and setting the minimum value to a lower limit of the standardized scale; comparing the measured value of the monitored physiological parameter to the standardized scale; and determining a level or type of medical intervention for the patient based on the comparison with the standardized scale.

[0139] The foregoing method may also include collecting IVC area trace data for multiple patients in multiple fluid volume states; storing the collected data in a database; analyzing the data to determine a curve between a maximum and a minimum among many different maximum values; and creating a standardized scale based on the determined curve. In other embodiments, the patient physiological parameter related to the patient fluid volume includes the IVC size; and monitoring includes measuring the IVC size. Monitoring may also include measuring the IVC size with an implantable wireless sensor. In another alternative, monitoring may include measuring the IVC size and an external ultrasound imaging device. Compared to traditional heart failure management programs, the systems and methods described herein provide unique data and novel data interpretation tools to help clinicians manage heart failure, reduce hospitalizations, better manage patient treatment, and thereby improve patient outcomes.

[0140] The illustrative embodiments of the present disclosure have been described in detail above. Note that in this specification and the claims appended hereto, unless otherwise expressly stated or indicated, conjunction language such as used in the phrases "at least one of X, Y, and Z" and "one or more of X, Y, and Z" should be understood to mean that each item in the conjunction list may be expressed as any number excluding all other items in the list or as any number in conjunction with any or all other items in the conjunction list, and each item may also be expressed as any number. Applying this general rule, in the aforementioned examples where the conjunction list consists of X, Y, and Z, the conjunction phrases should each include: one or more of X; one or more of Y; one or more of Z; one or more of X and one or more of Y; one or more of Y and one or more of Z; one or more of X and one or more of Z; and one or more of X, one or more of Y, and one or more of Z.

[0141] Various modifications and additions may be made without departing from the spirit and scope of the present disclosure. The features of each of the various embodiments described above may be appropriately combined with the features of the other described embodiments to provide a variety of feature combinations in associated new embodiments. In addition, although the foregoing describes a plurality of separate embodiments, what is described herein is merely an illustration of the application of the principles of the present disclosure. In addition, although the specific methods herein may be illustrated and / or described as being performed in a particular order, it is within the technical scope of the art that the ordering is highly variable to implement aspects of the present disclosure. Therefore, this description is intended to be by way of example only and not to otherwise limit the scope of the present disclosure or the invention as set forth in the appended claims.

Claims

1. An automatic heart failure diagnostic device comprising: a trace feature detector for identifying, for a received periodic vascular area trace representing changes in a patient's fluid state over time, selected features and at least one of amplitude and timing of the identified features, wherein the selected features include one or more of the following: interval time per respiratory cycle, respiratory-adjusted area amplitude, interval time per cardiac cycle, cardiac-adjusted area amplitude, primary cardiac peak, secondary cardiac peak, respiratory-related area reduction, motion type, and maximum and minimum areas associated with the identified motion; An indicator generator for generating cardiac function-related parameters for each area trace based on the identified features, amplitudes and timings, the cardiac function-related parameters comprising one or more of the following: maximum vessel area (A 最大 ), minimum vascular area (A 最小 ), mean vascular area (A 平均 ), heart rate (HR), respiratory rate (RR), collapsibility index (CI), collapse (C), and cardiac output (CO); a boundary generator configured to generate at least a lower blood vessel area boundary (LB) of the patient or an upper blood vessel area boundary (UB) of the patient using the generated cardiac function-related parameters; and An index generator for setting a patient congestion index based on: one or more of the generated cardiac function-related parameters; and at least one of the patient's lower vascular area boundary (LB) or upper vascular area boundary (UB), the congestion index indicating the patient's fluid state on a normalized scale for each of the trace cycles. 2 . The apparatus of claim 1 , further comprising a trace generator configured to generate the periodic vessel area trace.

3. The apparatus of claim 1 or claim 2, further comprising decision logic configured to determine at least one of a diagnostic status and a treatment regimen for the patient based at least on the congestion index from the index generator.

4. The apparatus of claim 1, claim 2 or claim 3, wherein: The boundary generator sets the lower area boundary (LB) according to one or more of the following relationships: the average minimum area measured in the IVC using a plurality of measurements from the patient or population information from a plurality of patients; =(-1 / S)·C+A 平均 , where (S) is the received reference slope; and The minimum vessel area (A) of the area trace associated with the detected motion 最小 ).

5. The device according to any one of claims 1 to 4, wherein: The boundary generator sets the upper area boundary (UB) according to one or more of the following relationships: an average maximum area measured in the IVC using at least one of a plurality of measurements from the patient or population information from a plurality of patients; =(-1 / S)·CI+A 平均 , where (S) is the received reference slope; and The maximum vessel area (A) of the area trace associated with the detected motion 最大 ).

6. Apparatus according to any one of the preceding claims, wherein: The index generator sets the hyperemia index according to one or more of the following relationships: Equal to (A 最大 -A 最小 ) / (A 最大 -LB) * 100%; Equal to 100*(A 平均 / UB); Equal to 100*((A 平均 -LB) / (UB-LB)); and Equal to 100 - (A 最大 -A 最小 ) / (A 最大 -LB) * 100%.

7. Apparatus according to any one of the preceding claims, wherein: The indicator generator is further configured to generate one or more of right atrial pressure (RAP), IVC tone, or systemic vascular resistance (SVR), wherein: predicting RAP based on the correlation between the received area trace and the measured right atrial pressure; using the trained model to relate IVC tension to the maximum area of ​​the area trace over time; and SVR was set equal to the measured mean atrial pressure minus RAP divided by collapse (CO).

8. The apparatus of any of the preceding claims, further comprising a machine learning-based model that applies weights to the congestion index, generated cardiac function-related parameters, and input patient-specific information to return a hospitalization probability score for the patient.

9. The apparatus according to any one of claims 2 to 8, wherein: The trace generator includes at least one of the following: External ultrasonic transducer; Implantable ultrasound sensor; An implantable resonant circuit coil sensor; or An implantable sensor comprising sensing electrodes.

10. A system for automatically determining a patient's fluid status using periodic vessel area traces, the system comprising: a trace feature detector for identifying selected features of the vessel area trace and at least one of an amplitude and a timing of the identified features with respect to the vessel area trace; An indicator generator for generating heart function related parameters for each area trace based on the identified amplitude and timing, the heart function related parameters including at least a maximum vessel area (A 最大 ) and minimum vascular area (A 最小 );as well as An index generator for determining an index based on at least one of a lower vascular area boundary (LB) or an upper vascular area boundary (UB) of the patient, the maximum vascular area (A 最大 ) and the minimum vessel area (A 最小 ) to generate a patient congestion index that indicates the patient fluid status on a normalized scale for each of the trace cycles.

11. The system according to claim 10, wherein: The selected area trace features also include one or more of the following: interval time of each respiratory cycle, area amplitude adjusted by respiration, interval time of each cardiac cycle, area amplitude adjusted by cardiac movement, main cardiac peak, secondary cardiac peak, respiration-related area reduction, action type, and maximum and minimum areas associated with the identified action.

12. A system according to claim 10 or claim 11, wherein: The indicator generator generates cardiac function related parameters, wherein the cardiac function related parameters further include one or more of the following: mean vascular area (A 平均 ), heart rate (HR), respiratory rate (RR), collapsibility index (CI), collapse (C), and cardiac output (CO).

13. A system according to any one of claims 10 to 12, wherein: The indicator generator is configured to receive patient-specific information and also generate the cardiac function-related parameter based on the received patient-specific information.

14. The system according to any one of claims 10 to 13, further comprising a boundary generator, the boundary generator being configured to generate at least the lower area boundary (LB) or the upper area boundary (UB) using the generated cardiac function-related parameters.

15. The system of claim 14, wherein: The boundary generator: At least receive motion detection and minimum vessel area (A 最小 );and The lower area boundary (LB) is set to the minimum vessel area (A) of the area trace associated with the motion detection. 最小 ).

16. The system of claim 14, wherein: The boundary generator: The collapse value (C) and average area (A) of the receiving area trace 平均 ) and a reference slope (S); and Set the lower area boundary (LB) equal to (-1 / S)·C+A 平均 .

17. The system of claim 14, wherein: The boundary generator: Receives the minimum area (A) of a selected number of area traces 最小 );and The lower area boundary (LB) is set to the average minimum area measured in the IVC using multiple measurements from a single patient or population information from multiple patients.

18. A system according to any one of claims 14 to 17, wherein: The boundary generator: receiving at least motion detection and maximum vessel area; and The upper area boundary (UB) is set to the maximum vessel area of ​​the area trace associated with the motion detection.

19. A system according to any one of claims 14 to 17, wherein: The boundary generator: The collapsibility index (CI) and the average area (A) of the receiving area trace were 平均 ) and reference slope (S); and Set the upper boundary (UB) equal to (-1 / S)·CI+A 平均 .

20. The system according to any one of claims 14 to 17, wherein: The boundary generator: Receives the maximum area (A) of the selected number of area traces 最大 );and The upper area boundary (UB) is set to an average maximum area measured in the IVC using at least one of a plurality of measurements from the patient or population information from a plurality of patients.

21. The system of any of the preceding claims, further comprising decision logic for determining at least one of a diagnostic status and a treatment regimen for the patient based at least on the congestion index from the index generator.

22. The system of claim 21, wherein: The decision logic: In response to the congestion index being between about 0 and 30, generating instructions to issue a reminder to bring to the user's attention and review the risk of hypovolemia; responsive to the congestion index being between about 30 and 70, generating instructions to issue a user notification that the patient's fluid status is within a normal range; and In response to the congestion index being between about 70 and 100, instructions are generated to issue a reminder to bring to the user's attention and review the risk of hypervolemia.

23. A system according to claim 21 or 22, wherein: The decision logic additionally receives a plurality of cardiac function related parameters from the indicator generator; and The decision logic includes a model trained by a data set, and the model trained by the data set returns the probability of hospitalization of the patient based on the congestion index and input cardiac function-related parameters.

24. The system of any preceding claim, further comprising one or more interface devices in communication with the indicator generator, index generator, and decision logic.

25. The system of claim 24, wherein: The one or more interface devices include a patient-personal device configured to wirelessly communicate with at least the indicator generator and the index generator; The patient interface device is configured for inputting the patient-specific information; and The patient device is configured to receive and display the patient congestion index.

26. The system of claim 25, wherein: The one or more interface devices include a medical care provider device configured to communicate with the indicator generator, index generator, and decision logic; The medical care provider device is configured for inputting patient-specific information; The medical care provider device is configured to input treatment and diagnostic algorithm changes in the decision logic; and The medical care provider device is configured to receive and display generated cardiac function related parameters, patient treatment or diagnostic status reminders, and the patient congestion index.

27. A system according to any one of the preceding claims, further comprising a trace generator, wherein: The trace generator comprises: at least one transducer configured to monitor changes in blood vessel area and to generate a sensor signal representative of the monitored changes; and A processing system is configured to receive the sensor signal and generate the periodic area trace based on the sensor signal.

28. The system of claim 27, wherein: The at least one transducer includes an implantable coil configured to generate a variable frequency signal related to changes in area of ​​the blood vessel when positioned within the blood vessel.

29. The system of claim 28, wherein: The processing system comprises: an antenna that the patient can wear; an excitation circuit in communication with the antenna; and Signal receiving circuitry is in communication with the antenna.

30. The system of any one of claims 27, 28 or 29, wherein: The sensor comprises: an expandable and contractible variable inductance coil comprising a plurality of adjacent strands formed about an open center to permit blood to flow through the open center in a substantially unimpeded manner, the coil being configured and dimensioned to move with the vessel wall in response to expansion and contraction of the vessel; and A capacitor, the capacitor together with the variable inductance coil forms a variable inductance resonant circuit, the variable inductance resonant circuit having a variable characteristic frequency related to a diameter or an area of ​​the expandable and contractible variable inductance coil.

31. A system according to any one of claims 12 to 30, wherein: The indicator generator: The interval time between receiving cardiac cycles (ti card );and The heart rate (HR) is set equal to 1 / (ti card ).

32. A system according to any one of claims 12 to 31, wherein: The indicator generator: The interval time between receiving breathing cycles (ti resp );and The respiratory rate (RR) is set equal to 1 / (ti resp ).

33. A system according to any one of claims 12 to 32, wherein: The indicator generator: The maximum receiving vessel area (A 最大 ) and the minimum blood vessel area (A 最小 );and Set Collapse (C) equal to A 最大 -A 最小 .

34. A system according to any one of claims 12 to 33, wherein: The indicator generator: receiving the maximum values ​​of collapse (C) and vessel area; and Set the Collapsibility Index (CI) equal to C / A 最大 *100%.

35. A system according to any one of claims 12 to 34, wherein: The indicator generator: Receive the collapsed cardiac component (H col ), collapsed respiratory component (R col ), heart rate (HR), and respiratory rate (RR); and Set cardiac output (CO) to H col The product of HR and R col The sum of the products of RR.

36. A system according to any one of claims 12 to 34, wherein: The indicator generator: receiving area trace; and Returns the cardiac output (CO) as the product of the area-time integral of the area trace and the correction factor.

37. A system according to any one of claims 10 to 36, wherein: The index generator: The maximum receiving vessel area (A 最大 ), minimum blood vessel area (A 最小 ) and the lower area boundary (LB); and The hyperemia index is set equal to (A 最大 -A 最小 ) / (A 最大 -LB)*100%.

38. A system according to any one of claims 10 to 36, wherein: The index generator: receiving the mean vessel area and upper boundary; and The hyperemia index is set equal to 100*(A 平均 / UB).

39. A system according to any one of claims 10 to 36, wherein: The index generator: The maximum receiving vessel area (A 最大 ), minimum blood vessel area (A 最小 ), upper area boundary (UB) and lower area boundary (LB); and The hyperemia index is set equal to 100*((A 平均 -LB) / (UB-LB)).

40. The system of claim 36, wherein: The exponent generator is also set equal to 100-(A 最大 -A 最小 ) / (A 最大 -LB)*100%.

41. A computer-based method comprising: receiving, within at least one processing device, a quiet respiratory vascular area trace of the patient; receiving within the processing device patient-specific information including at least patient weight and patient age; filtering the quiet respiratory vascular area trace at the processing device to identify component signals including at least a respiratory trace, a cardiac trace, and an average trace; extracting, at the processing device, from the traces amplitude features and timing features corresponding to at least one or more of area maximum, area minimum, collapse, respiratory collapse, cardiac collapse, heart rate, and respiratory rate; Generate cardiac function related parameters at the processing device using executable program instructions, the executable program instructions defining the cardiac function related parameters based on the extracted amplitude features and timing features, the cardiac function related parameters including one or more of the following: respiratory rate, cardiac output, heart rate, collapsibility index, collapse, maximum vascular area, minimum vascular area, and mean vascular area; generating a patient congestion index based on one or more of the amplitude characteristic and the cardiac function-related parameter; and Using a machine learning based model, weights are applied to the congestion index, the cardiac function related parameters, and the patient specific information to return a hospitalization probability score for the patient.

42. The method of claim 41 further comprising generating a reminder with instructions directing the user to modify or maintain treatment in response to the returned probability of hospitalization.

43. The method of claim 41 or 42, further comprising receiving external sensor input comprising at least patient-worn accelerometer data indicative of the patient's activity level, wherein: The external sensor inputs are incorporated into the machine learning model weights.

44. The method of claim 41, 42 or 43, further comprising generating and weighting additional cardiac function related parameters, the additional cardiac function related parameters comprising one or more of right atrial pressure (RAP), IVC tone or systemic vascular resistance (SVR), wherein: predicting RAP based on the correlation between the received area trace and the measured right atrial pressure; The trained model was used to relate IVC tension to the maximum area of ​​the area trace over time; and SVR was set equal to the measured mean atrial pressure minus RAP divided by collapse (CO).

45. A computer-based system comprising one or more processing and memory devices configured and programmed to perform the method of any one of claims 41, 42, 43 or 44.

46. ​​A method for automatically determining a patient's fluid status using periodic vessel area traces, comprising: receiving vessel area trace; identifying, for the vessel area trace, selected features and at least one of amplitude and timing of the identified features; generating cardiac function-related parameters for each area trace based on the identified amplitude and timing, the cardiac function-related parameters including measured vessel area and vessel area boundaries; A patient congestion index is generated, the patient congestion index representing a relationship between a measured vessel area determined for the area trace and a vessel area boundary of the patient, the congestion index indicating a patient fluid state on a normalized scale for each cycle of the trace.

47. The method of claim 46, further comprising generating at least one of a lower area boundary (LB) or an upper area boundary (UB) using the generated cardiac function-related parameters, and wherein, The generating a patient congestion index is based on at least one of the lower area boundary or the upper area boundary.

48. The method of claim 47, wherein: The generating at least one of a lower area boundary (LB) or an upper area boundary (UB) comprises: detecting motion in the received area trace; detecting a minimum blood vessel area corresponding to the detected motion; and The lower area boundary (LB) is set to the detected minimum vessel area.

49. The method of claim 47, wherein: The generating at least one of a lower area boundary (LB) or an upper area boundary (UB) comprises: The collapsibility index (CI) and the average area (A 平均 ); receiving a reference slope; and Set the lower area boundary (LB) equal to (-1 / S)·C+A 平均 .

50. The method of claim 47, wherein: The generating at least one of a lower area boundary (LB) or an upper area boundary (UB) comprises: Receives the minimum area (A) of a selected number of area traces 最小 );and The lower area boundary (LB) is set to the average minimum sensor size measured in the IVC using multiple measurements from an individual patient or population information from multiple patients.

51. The method according to any one of claims 47 to 50, wherein: The generating at least one of a lower area boundary (LB) or an upper area boundary (UB) comprises: detecting motion in the received area trace; detecting a maximum blood vessel area corresponding to the detected motion; and The upper area boundary (UB) is set to the detected maximum vessel area.

52. The method according to any one of claims 47 to 50, wherein: The generating at least one of a lower area boundary (LB) or an upper area boundary (UB) comprises: The collapsibility index (CI) and the average area (A) of the receiving area trace were 平均 ); receiving a reference slope (S); and Set the upper boundary (UB) equal to (-1 / S)·CI+A 平均 .

53. The method according to any one of claims 47 to 50, wherein: The generating at least one of a lower area boundary (LB) or an upper area boundary (UB) comprises: Receives the maximum area (A) of the selected number of area traces 最大 );and The upper area boundary (UB) is set to an average maximum sensor size measured in the IVC using at least one of a plurality of measurements from the patient or population information from a plurality of patients.

54. The method of any one of claims 47 to 53, further comprising determining a diagnostic status for the patient based at least on the congestion index.

55. The method of any one of claims 47 to 54, further comprising determining a treatment regimen for the patient based at least on the congestion index.

56. The method of claim 54 or claim 55, further comprising: In response to the congestion index being between about 0 and 30, generating a reminder for user attention and review of a risk of hypovolemia; responsive to the congestion index being between about 30 and 70, generating a user notification that the patient's fluid status is within a normal range; and In response to the congestion index being between about 70 and 100, an alert is generated for user attention and review of risk of hypervolemia.

57. The method of any one of claims 54, 55 or 56, further comprising determining a probability of hospitalization of the patient based on the congestion index and weighting of a plurality of additional generated cardiac function-related parameters using a model trained using a data set.

58. The method of any of the preceding claims, further comprising generating a periodic vessel area trace for the patient.

59. The method of claim 58, wherein: The generating of periodic blood vessel area traces comprises: monitoring changes in the area of ​​the blood vessel using a transducer device to produce a sensor signal representative of the monitored changes; and The sensor signal is processed to generate the periodic area trace.

60. The method of claim 59, wherein: The monitoring includes: The sensor signal is received from a sensor implanted in the IVC, the sensor configured to generate a variable frequency signal related to changes in vessel area.

61. The method according to any one of claims 46 to 60, wherein: The generating of cardiac function related parameters comprises: The interval time between receiving cardiac cycles (ti card );and Set the heart rate (HR) equal to 1 / (ti card ).

62. The method according to any one of claims 46 to 61, wherein: The generating of cardiac function related parameters comprises: The interval time between receiving breathing cycles (ti resp );and Set the respiratory rate (RR) equal to 1 / (ti resp ).

63. The method according to any one of claims 46 to 62, wherein: The generating of cardiac function related parameters comprises: The maximum receiving vessel area (A 最大 ) and the minimum vascular area (A 最小 );and Set Collapse (C) equal to A 最大 -A 最小 .

64. The method according to any one of claims 46 to 63, wherein: The generating of cardiac function related parameters comprises: receiving the maximum values ​​of collapse (C) and vessel area; and Set the Collapsibility Index (CI) equal to C / A 最大 *100%.

65. The method according to any one of claims 46 to 64, wherein: The generating of cardiac function related parameters comprises: Receive the collapsed cardiac component (H col ), collapsed respiratory component (R col ), heart rate (HR), and respiratory rate (RR); and Set cardiac output (CO) to H col The product of HR and R col The sum of the products of RR.

66. The method according to any one of claims 46 to 65, wherein: Generating a patient congestion index comprises: The maximum receiving vessel area (A 最大 ), minimum blood vessel area (A 最小 ) and the lower area boundary (LB); and The hyperemia index is set equal to (A 最大 -A 最小 ) / (A 最大 -LB)*100%.

67. The method according to any one of claims 46 to 65, wherein: Generating a patient congestion index comprises: receiving the mean vessel area and upper boundary; and The hyperemia index is set equal to 100*(A 平均 / UB).

68. The method according to any one of claims 46 to 67, wherein: Generating a patient congestion index comprises: The maximum receiving vessel area (A 最大 ), minimum blood vessel area (A 最小 ), upper area boundary (UB) and lower area boundary (LB); and The hyperemia index is set equal to 100*((A 平均 -LB) / (UB-LB)).

69. The method according to any one of claims 46 to 68, wherein: The selected area trace features also include one or more of the following: interval time of each respiratory cycle, area amplitude regulated by respiration, interval time of each cardiac cycle, area amplitude regulated by heart, main cardiac peak, secondary cardiac peak, respiration-related area reduction, action type, and maximum and minimum areas associated with the identified action.

70. The method according to any one of claims 46 to 69, wherein: Generating the cardiac function related parameters includes generating one or more of heart rate (HR), respiratory rate (RR), collapsibility index (CI), collapse (C) and cardiac output (CO).

71. The method of any one of claims 46 to 70, wherein: The method is a computer-based method, wherein the method has steps performed in one or more processing devices.

Images