Fluid bolus recommendation

Abstract

A method for managing fluid administration to a patient includes accessing a plurality of features associated with administration of a fluid bolus to the patient, receiving a first fluid bolus recommendation indicative of a predicted change in a physiological parameter of the patient in response to the fluid bolus based on the plurality of features, a second fluid bolus recommendation is determined based on the first fluid bolus recommendation and a subset of the plurality of features, and the second fluid bolus recommendation is provided.

Description

[0001] Related applications This application claims priority to U.S. Provisional Application No. 63 / 579,793, filed on August 30, 2023, entitled “FLUID BOLUS RECOMMENDATION,” the disclosure of which is incorporated herein by reference in its entirety. Background Art

[0002] The present disclosure relates generally to the field of fluid administration, including hemodynamic management devices and methods that can facilitate fluid administration, blood product transfusion, and blood pressure support medication administration. Summary of the Invention

[0003] Described herein are devices, methods, and systems related to the management of fluid bolus administration to a patient.

[0004] For purposes of summarizing the present disclosure, certain aspects, advantages, and novel features have been described. It should be understood that not all of these advantages are necessarily achieved by any particular example. Thus, the disclosed examples may be implemented in a manner that achieves or optimizes one advantage or group of advantages as taught herein without necessarily achieving other advantages as may be taught or suggested herein.

[0005] Any example methods and structures disclosed herein for treating a patient also include similar methods and structures performed or placed on a simulated patient, for example, for training; for demonstration; for program and / or device development; and the like. The simulated patient can be physical, virtual, or a combination of physical and virtual. The simulation can include a simulation of all or a portion of a patient (e.g., the entire body, a portion of the body (e.g., the chest), a system (e.g., the cardiovascular system), an organ (e.g., the heart), or any combination thereof). The physical elements can be natural, including human or animal cadavers or parts thereof; synthetic; or any combination of natural and synthetic. The virtual elements can be entirely in silica, or overlaid on one or more physical components. The virtual elements can be presented on any combination of a screen, a head-mounted device, a hologram, a projection, a speaker, headphones, a pressure transducer, a temperature transducer, or using any combination of suitable technologies.

[0006] Any of the various systems, devices, apparatuses, etc. disclosed herein can be sterilized (e.g., with heat, radiation, ethylene oxide, hydrogen peroxide, etc.) to ensure that they are safe for use with patients, and the methods herein can include sterilizing the associated systems, devices, apparatuses, etc. (e.g., with heat, radiation, ethylene oxide, hydrogen peroxide, etc.). BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Various examples are depicted in the drawings for purposes of illustration only and are not intended to limit the scope of the present invention in any way. Furthermore, the various features of the different disclosed examples can be combined to form additional examples, which are a part of this disclosure. Throughout the drawings, reference numbers can be re-used to indicate correspondence between referenced elements.

[0008] Figure 1 An example system for administering IV fluids and / or medications to a patient is shown in accordance with one or more examples.

[0009] Figure 2 A system for delivering one or more fluids to a patient using a control device, such as the control device of Figure 1 , in accordance with one or more examples is shown.

[0010] Figure 3 A data flow for inputting data to a bolus recommendation engine in accordance with one or more examples is shown.

[0011] Figure 4 A flowchart showing a process for providing a fluid bolus recommendation in accordance with one or more examples is provided.

[0012] Figure 5 A flowchart showing analysis and / or selection of features from a set of candidate features in accordance with one or more examples is shown.

[0013] Figure 6 A flowchart showing a process for providing a fluid bolus recommendation in accordance with one or more examples is provided.

[0014] Figure 7 An example structure of a responsiveness estimation engine in accordance with one or more examples is shown.

[0015] Figure 8 A responsiveness framework for predicting a patient's response to a fluid bolus in accordance with one or more examples is shown. DETAILED DESCRIPTION

[0016] The headings provided herein are merely for convenience and do not necessarily affect the scope of the claimed invention or that of the corresponding disclosure.

[0017] While certain preferred examples are disclosed below, it is to be understood that the subject matter of this application is not limited to the specifically disclosed examples, but extends to other alternative examples and / or uses thereof and modifications and equivalents thereof. Accordingly, the scope of the claims that can be presented hereafter is not to be understood as being limited to the specific examples described below. For example, in any method or process disclosed herein, the acts or operations can be performed in any suitable order and are not necessarily limited to any particular suggested order. Various operations can be described as being performed in one chronological order; however, the described order does not necessarily indicate a strict chronological order. Moreover, the described order is not meant to limit the scope of the examples described herein. For instance, certain aspects of the examples can be performed in any order that is practical based on the context. Furthermore, the structures, systems and / or devices described herein can be embodied as integrated components or as separate components. For the purposes of comparison of various examples, certain aspects and advantages of the examples are described. Not necessarily all such aspects or advantages are realized by any particular example. Thus, for example, various examples can be practiced that implement or optimize one advantage or a group of advantages as taught herein without necessarily implementing other aspects or advantages as can also be taught or suggested herein.

[0018] For ease of identification, certain reference numbers are reused in different ones of the figures of the disclosure to refer to devices, components, systems, features, and / or modules that can have similar features in one or more respects. However, reuse of a common reference number in the figures does not necessarily indicate that the referenced features, devices, components, or modules are the same or similar with respect to any particular example disclosed herein. Rather, one of ordinary skill in the art can learn from the context in which a particular reference number is used in a particular figure the degree to which reuse of the common reference number can imply similarity between the referenced subject matter. Use of a particular reference number in the context of a description of a particular figure can be understood to relate to the device, component, aspect, feature, module, or system identified in that particular figure and not necessarily to any device, component, aspect, feature, module, or system identified by the same reference number in another figure. Moreover, aspects of separate figures identified with a common reference number can be interpreted to share a characteristic or to be completely independent of one another.

[0019] Where an alphanumeric reference identifier comprising a numeric portion and an alphabetic portion is used (e.g., "10a," where "10" is the numeric portion and "a" is the alphabetic portion), reference in the written description to only the numeric portion (e.g., "10") may refer to any feature identified in a drawing using such a numeric portion (e.g., "10a," "10b," "10c," etc.), even where such features are identified by a reference identifier concatenating the numeric portion with one or more alphabetic characters (e.g., "a," "b," "c," etc.). That is, as an example, reference in this written description to feature "10" may be understood to refer to identifying feature "10a" in a particular drawing of the present disclosure, or to identifiers "10" or "10b" in the same drawing or another drawing.

[0020] Certain standard anatomical positional terms are used herein to refer to animal (i.e., human) anatomical structures relative to various exemplary embodiments. Although certain spatially relative terms (such as "external," "internal," "upper," "lower," "below," "above," "vertical," "horizontal," "top," "bottom," and similar terms) are used herein to describe the spatial relationship of one device / element or anatomical structure to another, it should be understood that these terms are used herein to describe the positional relationship between elements / structures, as illustrated in the accompanying figures, for ease of description. It should be understood that these spatially relative terms are intended to encompass different orientations of elements / structures in use or operation, in addition to the orientations depicted in the accompanying figures. For example, an element / structure described as being "above" another element / structure may refer to a position below or beside such other element / structure relative to the subject patient or an alternative orientation of the element / structure, and vice versa. It should be understood that spatially relative terms (including those listed above) are to be understood relative to the respective illustrated orientations of the referenced figures.

[0021] Fluid Management System Excessive and / or insufficient fluid administration may lead to complications. Goal-directed fluid management can help optimize the amount and / or timing of fluid administration. Some examples described herein relate to guiding effective intravenous fluid administration during surgery and / or otherwise.

[0022] Maintaining adequate oxygen delivery during surgery can prevent damage to vital organs and the resulting complications. Maintaining adequate intraoperative cardiac output is crucial for oxygen delivery. Hemodynamically guided fluid management (also known as goal-directed therapy) can help optimize cardiac output and / or improve outcomes in high-risk surgical patients.

[0023] Goal-directed therapy can require clinicians to follow a standardized system (such as a computer system) that determines when fluid should be administered. A common feature of some systems is an effort to maintain a pre-defined stroke volume (SV) and limit SV variation (e.g., less than 12%). The complexity and variety of goal-directed therapy systems can make their implementation challenging. As a result, compliance with these systems is often poor.

[0024] Using invasive (such as arterial) pressure information, example systems can recommend fluid administration when a patient can respond to a pre-defined increase in fluid bolus in SV. Automatically assessing hemodynamic status and prompting specific fluid recommendations can facilitate intraoperative fluid management during surgery.

[0025] Some example systems utilize an open-loop fluid management workflow. The system can automatically perform and / or the clinician can be guided by the system while maintaining full control over fluid administration. The fluid management system can perform various functions, including (1) integrating monitored hemodynamic variables and / or continuously analyzing a patient's fluid responsiveness; (2) analyzing response to fluid bolus; and / or (3) predicting a patient's current fluid responsiveness and prompting the clinician to consider a fluid bolus when appropriate.

[0026] Some example systems and / or methods provided involve hemodynamic monitoring, which includes dynamic parameters of fluid responsiveness (e.g., Fluid Predictors) derived from various physical data, which can include arterial pressure waveforms, etc.

[0027] In addition to other physiological data, the systems described herein can utilize dynamic predictors of fluid responsiveness and / or other dynamic data. Dynamic predictors can include pulse pressure variation (PPV), stroke volume variation (SVV), pleth variability, and / or electrocardiogram (EKG) waveform characteristics, which description will simply refer to this group as "Fluid Predictors" or FP. This term should be considered to mean any described predictor of fluid responsiveness.

[0028] Other terms and abbreviations used herein include: CO - cardiac output; APCO - arterial pressure cardiac output; SWI - stroke work index; CI - cardiac index; dP / dt - maximum rate of change of arterial blood pressure waveform; patient or subject - "patient" or "subject" is the organism that the system monitors and manages (in one example, the patient can be a human; however, the patient can include an animal); Vital Signs or Vitals - any statistical measure of a physiological process occurring within a patient - include waveforms derived from physiological processes. Vital signs can include, for example: Heart Rate (HR) - the number of ventricular contractions per minute; Stroke Volume (SV) - the volume of blood ejected by the left ventricle during contraction, in milliliters; Systolic Blood Pressure (SBP) - the highest blood pressure felt in the systemic arterial vascular tree during a heart cycle; Diastolic Blood Pressure (DBP) - the lowest blood pressure felt in the systemic arterial vascular tree during a heart cycle; Mean Arterial Pressure (MAP) - the average blood pressure of the systemic arterial system over one or more heart cycles, typically calculated as ((SBP+DBP+DBP) / 3); Systemic Vascular Resistance (SVR) - the amount of force exerted by the body's vasculature on circulating blood; Cardiac Output - the total volume of blood ejected by the left ventricle in one minute; Dynamic Prediction Index (DP) - one or more measures of preload dependency derived from arterial pressure waveforms, plethysmogram waveforms, EKG waveforms, chest ultrasound, bioimpedance, bioresponse, and including operational maneuvers such as passive leg raising and remote expiratory pause; Intravenous Fluid (IV Fluid) - any fluid intended for intravenous administration to a monitored subject for the purpose of vascular volume expansion or replacement, or to act as a carrier for intravenous medications. IV fluids will thus include, but are not limited to: crystalloids solutions such as Lactated Ringer's Solution, normal saline, dextrose solutions, Plasmalyte, and generally balanced salt solutions and sugar solutions; colloid solutions such as albumin, starches, and the like; and blood products and blood analogs such as whole blood, platelets, fresh frozen plasma, cryoprecipitate, packed red blood cells, recovered cell solutions, or any substitute intended to mimic or replace these products; Fluid bolus - the administration of a specific volume of IV fluid over a discrete time span; "Efficiency" of a fluid bolus - the degree to which intravascular administration of the fluid increases cardiac output; or, for example, the degree to which intravascular administration of the fluid improves oxygen delivery to tissues; Prediction - the calculated percentage by which a fluid bolus is expected to result in an increase in patient cardiac output; Vasopressor - a medication controlled by the system that can include those intended to manipulate blood pressure and cardiac output, such as, for example, ephedrine, phenylephrine, norepinephrine, epinephrine (adrenalin), dopamine dobutamine, milrinone, dopexamine, nitroglycerin, nitroprusside, and other vasopressors, inotropic drugs, and vasodilators.

[0029] Figure 1An example system 100 for administering IV fluid 102 and / or medication 104 to a patient 101 is shown in accordance with one or more examples. The system 100 can include a control device 106 (e.g., control circuitry) configured to manage the hemodynamic delivery of the IV fluid 102 (e.g., IV fluid bag) and / or medication 104 to the patient 101 via a first pump 108 (e.g., fluid pump and / or infusion pump), a second pump 110 (e.g., fluid pump and / or infusion pump), and / or various additional pumps. The control device 106 can be electronically coupled / coupled to the first pump 108 and / or the second pump 110 via various circuitry and / or electronic interfaces.

[0030] The control device 106 can include one or more sensors and / or display devices. For example, the display device can include a display screen and / or user interface device (e.g., keyboard, mouse, etc.). The first pump 108 can be coupled to the IV fluid 102 and / or the second pump 110 can be coupled to the medication 104, which can include various containers containing fluid medication. The IV fluid 102 and / or medication 104 can be fluidly coupled to the patient 101 via tubing and / or the administration of the IV fluid 102 and / or medication 104 can be administered to the patient 101 and / or dynamically controlled by the control device 106.

[0031] The control device 106 can include one or more processing units and / or one or more computer-readable storage media. The processing units can be implemented within one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital information processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, micro-controllers, microprocessors, other electronic units designed to perform the functions described above, and / or combinations thereof. The computer-readable storage media can include one or more memories for storing data, including read only memory (ROM), random access memory (RAM), magnetic RAM, core memory, magnetic disk storage mediums, optical storage mediums, flash memory devices, and / or other machine readable mediums for storing information. The computer-readable storage media can be embodied in one or more of the following: one or more portable or fixed storage devices, optical storage devices, wireless channels, and / or various other storage devices and mediums as will occur to those skilled in the art.

[0032] The control device 106 can be coupled to the fluid pumps via one or more interface ports. The interface ports can include one or more standard USB ports and standard serial ports. The USB connectors can also be used to import patient data in real-time from other sources, such as a patient monitor. Optionally, external components can be connected to the control device 106 to allow for direct monitoring of patient vital signs by the control device 106. One or more interfaces can allow for data transfer to and from the control device 106 to share data with other networked devices.

[0033] The first pump 108, the second pump 110, and / or the control device 106 can include a single device and / or can be integrated in a single housing. Alternatively, the first pump 108 and / or the second pump 110 can be external to the control device 106. The first pump 108 can be configured to regulate and / or drive the flow of IV fluid 102 to the patient 101 and / or to select which fluid to use from one or more fluid containers. The IV fluid can include one or more of a crystalloid, a colloid, or a blood product, among other fluids. The second pump 110 can be integrated with or external to the control device 106 and / or can include a syringe pump system for use with multiple drug vials and / or containers.

[0034] After flushing and / or before use, various tubing can be coupled together with the control device 106. In some examples, multiple IV tubes can be connected to the patient 101. One end of the IV tubing can include bag taps for use with IV solutions, colloids, and blood products. The opposite end of the IV tubing can be a male luer lock for connection to an IV tubing set and / or claves. The IV tubing can also have a side port distal to the main fluid pump to which a drug syringe from a drug vial can be attached.

[0035] Figure 2 A system 200 for delivering one or more fluids to a patient 201 using a control device 206 (e.g., the control device 106 of FIG. 1) is shown in accordance with one or more examples. Figure 1

[0036] ​In some examples, the patient 201 can be monitored through the use of one or more clinical sensors 212 and / or monitors. The various sensors 212 can be coupled to the control device 206 via various circuitry. The sensors 212 can be integrated with and / or can be separate from (e.g., detached and / or remotely connected from) the control device 206. The sensors 212 can be configured to measure vital signs and / or various physiological data from the patient. Such measured data can be stored at a sensor data storage 214 of the control device 206. In some examples, the measured data can be noise and / or validity filtered prior to being stored at the data storage 214. The sensor data storage 214 can be configured to store additional data, including population data and / or various data collected from patients other than the patient 201.

[0037] The sensor data analysis processor 216 can access data from the sensor data storage 214. The processor 216 can be configured to make predictions regarding the likely efficacy of a fluid bolus. In some examples, the processor 216 can be configured to analyze sensor data from the sensors 212 in response to a fluid bolus and / or compare sensor data from the sensors 212 to data obtained from a prior population of patients (e.g., patients with similar vital signs).

[0038] Data from the processor 216 can be transmitted to a bolus log data storage 218 and / or a bolus recommendation engine 220 automatically and / or upon request. In some examples, data can be sent to the log data storage 218 when a fluid bolus is initiated and / or terminated. The log data storage 218 can also receive input from a pump controller 222 (e.g., when a fluid bolus is initiated or terminated), which includes certain details of the administered bolus.

[0039] The bolus recommendation engine 220 can be configured to analyze historical data of the patient 201 and / or predict the current efficacy of a fluid bolus. Such predictions can involve analysis and / or sub-analysis of sensor data from the clinical sensors 212.

[0040] Engine 220 can be configured to determine a combined prediction of the patient's 201 current state's cardiac output based on log data 218, sensor data 214, and / or various population data. The combined prediction can optionally result in a command being transmitted from engine 220 to pump controller 222 to allow pump controller 222 to automatically cause activation and / or deactivation of one or more fluid pumps 224 configured to pump one or more fluids to patient 201. However, engine 220 can additionally or alternatively present the prediction and / or recommendation to display 226 (e.g., a monitor) to allow a clinician to view the results and / or manually pump and / or command pumping of fluid pumps 224.

[0041] Fluid pumps 224 can be externally controlled fluid pumps that contain independent command interfaces, alarm systems, and / or configurations. Control of fluid pumps 224 can be achieved through serial, network, wireless, Bluetooth, and / or other electronic protocols. Fluid pumps 224 can include one, two, or more than two physical fluid pumps 224.

[0042] In alternative examples, fluid pumps 224 can be integrated components of control device 206. Fluid pumps 224 can or can not include alarm and / or control configurations. If fluid pumps 224 do not include alarm and control devices, the alarm and control devices of pumps 224 can be included in control device 206. User interfaces (e.g., display 226 and / or associated devices) of control device 206 can be used to affect certain aspects of fluid pumps 224.

[0043] Pumps 224 can be configured to deliver IV fluids and / or medications to patient 201 at determined rates and / or times and / or according to clinician control. Display 226 can include a touchscreen interface for monitoring patient vital signs and / or for entering patient data and user preferences into control device 206.

[0044] Control device 206 can be configured to monitor and / or store various hemodynamic data including variables such as heart rate, mean arterial pressure, and advanced variables estimated using pulse contour analysis including SV, SV variation, and systemic vascular resistance. Data can be recorded at the start and / or end time of each bolus as well as fluid type and volume. Engine 220 can be configured to utilize fluid delivery data and / or hemodynamic data to estimate the percent change in SV (e.g., variation in SV) caused by a fluid bolus. Engine 220 can be configured to calculate the expected change in SV by superimposing the start and stop times of a fluid bolus on SV measurements and / or displaying these data at display 226 each time a fluid gives within a specified range.

[0045] The prediction of the patient 201 current fluid responsiveness at the engine 220 can combine predictions from the population model and the bolus log model. The population model describes the relationship between SV change and the predicted change in SV. The bolus log model uses the hemodynamic response to past fluid boluses to determine whether the patient is responsive to fluid. The engine 220 can generate a predicted change in SV by identifying boluses given at similar hemodynamic states and aggregating the responses. In some examples, the engine 220 can be configured to compare the measured change in SV to the predicted change in SV for those boluses used in the bolus log model and be able to correct for systematic bias in the prediction model (e.g., that the model is overestimating or underestimating the patient’s response to fluid). In some examples, the average of the population model prediction and the bolus log prediction can be weighted at the engine 220 by the quality of information in the bolus log model to produce a final prediction. The engine 220 can determine whether a fluid bolus recommendation should be generated. If the predicted change in SV is greater than a threshold setting, the output of the engine 220 can be a fluid recommendation prompt displayed on the display 226.

[0046] Figure 3 A data flow 300 for inputting data to a bolus recommendation engine 320 is shown in accordance with one or more examples. In some examples, various dynamic physiological data 326 can be collected from various clinical sensors and / or monitors. For example, the physiological data 326 can include sensor data 328 related to a monitored patient and / or collected from one or more sensors and / or monitors. The physiological data 326 can also include a blood pressure waveform 330 for the patient. In some examples, the sensor data 328 can be used to generate and / or derive the blood pressure waveform 330. For example, the blood pressure waveform 330 can represent measured sensor data 328 tracked over a period of time.

[0047] The dynamic physiological data 326 can be input to an advanced hemodynamic feature processor 332. The processor 332 can be configured to perform analysis and / or sub-analysis on the physiological data 326 to generate additional features related to the patient (e.g., SV, SV change rate, APCO, etc.). Example features generated by the processor 332 are described in greater detail herein. The features generated by the processor 332 can be transmitted to the bolus recommendation engine 320 to generate a prediction of the patient’s response to a bolus and / or provide a recommendation of a bolus to the patient. The processor 332 can additionally or alternatively receive demographic data and / or data related to previous patients and / or patients with similar traits to the current patient. The demographic data can include sensor data and / or blood pressure waveform data received from previous patients.

[0048] In some examples, the engine 320 can additionally receive bolus log data 318 related to boluses previously provided to the patient and / or similar patients. The data 318 can include bolus start times, bolus stop times, and / or bolus volumes previously provided to the patient and / or other patients. The data 318 can be indicative of the engine 320 regarding the patient’s response to past boluses to help generate recommendations for future boluses. In some examples, the data 318 can include intervention data and / or any data that can be used to determine intervention actions, which can include bolus log data.

[0049] The predictions and / or recommendations generated by the engine 320 can be presented in a display device for review by a clinician. In some examples, the display device can provide an interface for initiating an action of one or more fluid pumps and / or can present options for selection by the clinician. For example, in the case of a recommended fluid bolus, the display device can indicate the recommended bolus and / or can present a “start bolus” and / or “down” option, each of which can be selected by the clinician.

[0050] Figure 4 A flowchart is provided showing a process 400 for providing fluid bolus recommendations in accordance with one or more examples. The process 400 can involve automatically providing fluid boluses in response to recommendations and / or can involve displaying output for review and / or response by a clinician.

[0051] At step 402, the process 400 involves receiving feature data from a processor. The processor can be configured to generate the feature data using various dynamic physiological data specific to an individual patient and / or multiple patients. Example feature data can include basic features (e.g., cardiac output, beat-to-beat output, and / or amplitude, area, slope, duration, and / or display parameters related to cardiac output, beat-to-beat output, and / or other features), interactive features (e.g., baroreflex sensitivity), complexity features (e.g., sample entropy, approximate entropy, variability), and / or spectral features (e.g., spectral power of different frequency bands). The features can be specific to points along a waveform. For example, each feature can be measured at 20 second intervals. In some examples, the features can include a rate of change value for one or more features measured over a period of time. For example, the features can include a change in beat-to-beat output over a 20 second period.

[0052] The feature data can be transmitted by the processor and / or can be configured to be received at a bolus recommendation engine and / or similar device.

[0053] At step 404, process 400 involves receiving bolus log data. The bolus log data can be accessed from a log data store of a control device. In some examples, the bolus log data can be provided to the data store from a pump controller and / or a sensor data analysis processor. The log data can include data related to a particular patient and / or data related to multiple patients.

[0054] At step 406, process 400 involves estimating a patient response to a potential fluid bolus (e.g., an IV fluid and / or medication) for a particular patient. Estimating the patient response can involve evaluating and / or comparing the characteristic data and / or the bolus log data.

[0055] At decision block 408, the engine can determine whether to recommend a fluid based on the estimated patient response. The determined recommendation and / or recommendation can be presented to a clinician for consideration. In the case of recommending a fluid, a "recommend" response can be generated at step 410 and / or a fluid bolus can be provided. In the case of not recommending a fluid, a negative response can be generated at step 414 and / or a fluid bolus can not be provided. In the case of neither explicitly recommending nor explicitly not recommending a fluid, a "test" response can be generated at step 412. In response to the "test response," a limited bolus can be provided and / or recommended.

[0056] Figure 5A flow diagram illustrating analysis and / or selection of features from a candidate feature set is shown, in accordance with one or more examples. In some examples, a feature dataset 502 can include a set of various features that can be used to generate a prediction of a patient’s response to a fluid bolus and / or a recommendation to administer a fluid bolus to a patient. Example features of the feature dataset 502 can include basic features (e.g., cardiac output, beat-to-beat output and / or amplitude, area, slope, duration, and / or display parameters related to cardiac output, beat-to-beat output, and / or other features), interaction features (e.g., baroreflex sensitivity), complexity features (e.g., sample entropy, approximate entropy, variability), and / or spectral features (e.g., spectral power of different frequency bands). The features can be specific to a point along a waveform. For example, each feature can be measured at 20 second intervals. In some examples, the features can include a rate of change value for one or more features measured over a period of time. For example, a feature can include a beat-to-beat output change over a 20 second period. Additional features can include a stroke volume index (SVI), a rate of change of SVI, an approximate entropy of diastolic pressure, an area from the start of a heartbeat to the maximum of the systolic (minus diastolic pressure), a pressure at the dicrotic notch minus diastolic pressure early (e.g., 60 seconds ago), a rate of change of beat-to-beat output, a rate of change of approximate entropy of mean systolic pressure, a pressure at the dicrotic notch minus diastolic pressure 0 seconds ago, a skew approximate entropy rate of change, a SWI rate of change, a CI rate of change, a dP / dt approximate entropy, a dP / dt variability rate of change, a pulse pressure variation (PPV) rate of change, a systemic vascular resistance (SVR), a systemic vascular resistance index (SVRI), and / or a heart rate approximate entropy.

[0057] The feature dataset 502 can include multiple iterations of a single measurement type measured at different time points and / or over different time periods. For example, the feature dataset 502 can include a cardiac output rate of change measurement at a 0 second point, a 20 second point, a 40 second point, and / or a 60 second point. Similarly, the feature dataset 502 can include a beat-to-beat output measurement at a 0 second point, a 20 second point, a 40 second point, and / or a 60 second point.

[0058] The feature dataset 502 can be input to a performance / correlation analysis engine 504 to identify one or more features to include in a first feature subset 506 and / or a second feature subset 508. For example, the engine 504 can be configured to analyze each feature to determine a correlation of the feature to determine a prediction and / or recommendation of administering a bolus. For example, some features can have a relatively strong correlation to a patient’s response to a fluid bolus and thus can be deemed more relevant than other features for use by a recommendation engine.

[0059] In some examples, relevant features can be classified into the first subset 506 and / or the second subset 508. Features that are not classified into the first subset 506 and / or the second subset 508 can be ignored and / or classified into an irrelevant subset 510. The first subset 506 can be associated with a “recommend” response, and / or the second subset 508 can be associated with a “test” response. For example, in the case that the process 400 outputs a “recommend” response, the features of the first subset 506 can be applied to a deeper analysis to determine whether to administer a fluid bolus to the patient. In the case that the process 400 outputs a “test” response, the features of the second subset 508 can be applied. Figure 4 Figure 4 In the case that the process 400 outputs a “recommend” response, the features of the first subset 506 can be applied to a deeper analysis to determine whether to administer a fluid bolus to the patient. In the case that the process 400 outputs a “test” response, the features of the second subset 508 can be applied. In some examples, different features can be relatively more relevant and / or useful in the deeper analysis of the “recommend” case than in the “test” case, and vice versa. Thus, features determined to be most useful for the “recommend” case can be applied to such a case, and / or features determined to be most useful for the “test” case can be applied to such a case. In some examples, a feature can appear in both the first subset 506 and the second subset 508 and / or in neither.

[0060] Figure 6 A flowchart illustrating a process 600 for providing a fluid bolus recommendation according to one or more examples is provided. The process 600 can involve automatically providing a fluid bolus in response to a recommendation and / or can involve displaying an output for review and / or response by a clinician.

[0061] At step 602, the process 600 involves receiving feature data from a processor and / or determining (e.g., generating) feature data. The feature data can be associated with administering a fluid bolus to a patient. The processor can be configured to generate the feature data using various dynamic physiological data specific to an individual patient and / or multiple patients. Example feature data can include basic features (e.g., cardiac output, beat-to-beat output and / or amplitude, area, slope, duration, and / or display parameters related to cardiac output, beat-to-beat output, and / or other features), interactive features (e.g., baroreflex sensitivity), complexity features (e.g., sample entropy, approximate entropy, variability), and / or spectral features (e.g., spectral power of different frequency bands). The features can be specific to points along a waveform. For example, each feature can be measured at 20 second intervals. In some examples, the features can include a rate of change value for one or more features measured over a period of time. For example, the features can include a beat-to-beat output change over a 20 second period.

[0062] ​The characteristic data may be transmitted by the processor and / or may be configured to be received at a bolus recommendation engine and / or similar device. In some examples, the characteristic data may be collected using one or more sensors that may be attached to the patient and / or otherwise communicate with the patient. For example, the sensors may include wrist and / or finger cuffs.

[0063] In some examples, the characteristic data can be derived from a physiological signal associated with administering a fluid bolus to a patient. For example, the physiological signal can be received prior to the characteristic data set. In some examples, the physiological signal is arterial blood pressure, a signal proportional to arterial blood pressure, or otherwise related to arterial blood pressure.

[0064] At step 604, process 600 involves receiving bolus log data. The bolus log data can be accessed from a log data storage device of a control device. In some examples, the bolus log data can be provided to the data storage device from a pump controller and / or a sensor data analysis processor. The log data can include data associated with a specific patient and / or data associated with multiple patients.

[0065] At step 606, process 600 involves estimating a particular patient's response to a potential fluid bolus (e.g., IV fluid and / or medication). Estimating the patient's response may involve evaluating and / or comparing characteristic data and / or bolus log data. In some examples, estimating the patient's response may be performed by a bolus recommendation engine as described herein. Estimating the patient's response may involve receiving and / or generating a recommendation indicating a predicted change in one or more physiological parameters of the patient in response to one or more fluid boluses.

[0066] At decision block 608, the engine may determine whether to recommend a fluid based on the estimated patient response. The determined suggestion and / or recommendation may be presented to the clinician for consideration. However, in some examples, the suggestion determined at decision block 608 may not be presented to the clinician. Instead, subsequent suggestions determined at decision blocks 616 and / or 624 may be presented to the clinician. In the event that a fluid is recommended, at step 610, a first subset of features associated with the patient (e.g., Figure 5 The first subset 506 of the data may be analyzed and / or applied to the responsiveness estimation engine 630. For example, the first subset may include a set of measurement types that include various features described herein. Features may include a sub-analysis of data tracked by one or more physiological sensors attached to and / or in communication with the patient. In some examples, the first subset may be a subset of the feature dataset received in step 602. The responsiveness estimation engine 630 may be configured to evaluate and / or re-evaluate the recommendations made by the bolus recommendation engine 620.

[0067] In the case of no fluid being recommended, a negative response can be generated at step 614 and / or no fluid bolus can be provided. In the case of a test response being recommended (e.g., neither fluid is explicitly recommended nor is fluid explicitly not recommended), at step 612, a second subset of features (e.g., the second subset 508 of features) related to the patient can be analyzed and / or applied to a responsiveness estimation engine 630. For example, the second subset can include a set of measurement types that include various features described herein. The features can include a sub-analysis of data tracked by one or more physiological sensors attached to and / or in communication with the patient. In some examples, the second subset can be a subset of the feature dataset received at step 602. Figure 5

[0068] The responsiveness estimation engine 630 can be configured to apply deep learning and / or machine learning to analyze the first subset of features and / or the second subset of features to determine whether a fluid bolus (e.g., a normal or full bolus) is recommended at decision block 616 and / or a test bolus (e.g., a test and / or partial bolus) is recommended at decision block 624. In the case of a fluid bolus (e.g., a full bolus) being recommended, a “recommend” response can be generated and / or output at step 618 and / or step 626. In the case of a test bolus (e.g., a test and / or partial bolus) being recommended, a “test” response can be generated and / or output at step 622 and / or step 628. The analysis performed at decision blocks 616 and / or 624 can be the same and / or similar, however, different subsets of features can be applied by the engine 630 at blocks 616 and 624.

[0069] The output of steps 618, 622, 626, and / or 628 can be provided to a display device (e.g., a monitor) and / or to a computer and / or robotic system configured to automatically dispense fluid in response to the output. For example, in response to the output “recommend” response, a “recommend” message and / or option can be presented to the display device and / or a fluid pump can be automatically activated to provide a bolus to the patient. In response to the output “test” response, a “test” message and / or option can be presented to the display device and / or a fluid pump can be automatically activated to provide a test bolus to the patient.

[0070] The engine 630 can be configured to provide a double check and / or a second assessment of the responsiveness and / or expected responsiveness of the patient to a fluid bolus. In some examples, the engine 630 can confirm the results determined by the engine 620 and / or can modify the results determined by the engine 620 based on a more in-depth analysis and / or more focused analysis of particular features of the feature dataset received at step 602.

[0071] ​While the responsiveness estimation engine 630 is shown as receiving recommendations from the bolus recommendation engine 620, in some examples, the responsiveness estimation 630 can be configured to determine recommendations without receiving prior recommendations. For example, the engine 630 can be configured to generate fluid bolus recommendations using only feature data and / or bolus log data and / or without using suggested fluid recommendations from engine 620. Engine 630 can be configured to apply a subset of feature data or a full set of feature data.

[0072] Based at least in part on the recommendations generated in process 600, one or more fluid boluses can be delivered and / or administered to a patient. For example, a control device can be configured to generate signals to one or more fluid pumps to cause the one or more fluid pumps to administer one or more fluid boluses to a patient in response to one or more generated recommendations.

[0073] Figure 7 An example structure of a responsiveness estimation engine 730 is shown in accordance with one or more examples. In some examples, engine 730 can be configured to apply machine learning and / or deep learning to feature data received at engine 730. Engine 730 can include a plurality of models and / or data structures configured to analyze different groups and / or types of feature data. For example, engine 730 can include a first deep learning model 732 and / or a second deep learning model 734. Engine 730 can use first model 732 for a first subset of features and / or types of features, and / or can use second model 734 for a second subset of features and / or types of features. First model 732 and / or second model 734 can be configured to receive input of one or more features and / or sets of feature data and / or output a“recommendation” and / or a“test” response indicating a predicted responsiveness of a patient to a fluid bolus.

[0074] First model 732 and / or second model 734 can include a neural network and / or one or more layers and / or functions configured to analyze feature data. In some cases, the one or more layers and / or functions can be configured to remove unnecessary data and / or apply weights and / or biases to one or more portions of data.

[0075] First model 732 and / or second model 734 can include an input layer 736 configured to receive and / or process received feature data. In some examples, input layer 736 can be configured to format feature data into a format suitable for subsequent layers and / or functions of first model 732 and / or second model 734.

[0076] In some examples, the first model 732 and / or the second model 734 can include one or more dropout layers 738 configured to drop connections and / or prevent overfitting of the feature data. The dropout layers 738 can be employed after the input layer 736 and / or the dense layer 740. For example, the dropout layers 738 can remove some data added by the dense layer 740.

[0077] The dense layer 740 can include a neuron architecture configured to apply weights to the feature data. An example dense layer 740 can connect neurons between layers of the first model 732 and / or the second model 734. The dense layer 740 can detect complex patterns in the feature data and / or perform a non-linear transformation on the feature data (e.g., add weights and / or biases). The first model 732 and / or the second model 734 can include multiple dense layers 740. In some examples, the second model 734 can include more dense layers 740 and / or more dropout layers 738 than the first model 732. For example, the second model 734 can be configured to analyze feature data where the “test” response is initially generated by the engine and / or the feature data input to the second model 734 can have a higher degree of ambiguity and / or can include fewer connections than the feature data input to the first model 732 (the first model can be configured to analyze feature data where the “recommendation” response is initially generated by the engine). Accordingly, more dense layers 740 can be included in the second model 734 to focus the model on establishing connections between points of the feature data. The second model 734 can be relatively more complex and / or can include more layers relative to the first model 732.

[0078] The first model 732 can include an LI regularization layer 742 configured to shrink (e.g., apply reduced weights) to one or more features determined by the LI regularization layer 742 to be relatively less important. In some cases, the layer 742 can be configured to enhance sparsity and / or reduce some weights to zero to facilitate feature selection and / or simplify the first model 732.

[0079] In some examples, the first model 732 and / or the second model 734 can include a sigmoid layer 744 configured to map input and / or analyzed data to values within a given range (e.g., between 0 and 1). The sigmoid layer 744 can be configured to create non-linearity in the feature data.

[0080] The engine 730 can include a neural network that includes coefficients derived from feature data input to the engine 730 from previous patients. Additionally, the engine 730 can be configured to dynamically generate additional coefficients from a current patient.

[0081] Figure 8 A responsiveness framework 800 for predicting patient responsiveness to fluid boluses is shown in accordance with one or more examples. The framework 800 can be embodied in certain control circuitry, including one or more processors, data storage devices, connectivity features, substrates, passive and / or active hardware circuit devices, chips / dies, etc. For example, the framework 800 can be embodied in a responsiveness estimation engine 820 described in Figure 6 and Figure 7 The framework 800 can be configured to employ machine learning functionality to perform automatic object detection on features and / or bolus log data collected from one or more patients, for example.

[0082] In some examples, the engine 820 can be trained in accordance with known features and / or bolus log data 840 and / or known patient responsiveness data 842. For example, prior to receiving a fluid bolus, the engine 820 can be provided with feature data and / or log data previously collected from a patient. The engine 820 can also be provided with data relating to the patient’s responsiveness following administration of a fluid bolus. For example, the engine 820 can be provided with matching between input data of a prior patient and a known “recommended” (e.g., responder) or “test” (e.g., non-responder or partial responder) to train the engine 820 to match input data to a “recommended” or “test” outcome. The engine 820 can include any suitable or desirable transformation and / or classification architecture, such as any suitable or desirable artificial neural network architecture.

[0083] The engine 820 can be configured to adaptively adjust one or more parameters or weights associated with training input / output data to correlate known input and output data. For example, the engine 820 (e.g., a multi-layer perceptron and / or convolutional neural network) can be trained using a labeled data set and / or machine learning. In some implementations, the machine learning framework can be configured to perform learning / training in any suitable or desirable manner.

[0084] Known responsiveness data 842 can be generated at least in part by manually labeling patient features and / or log data with “recommended” or “test” responsiveness. For example, a relevant medical expert can determine and / or apply a manual label to indicate where a patient can benefit from a fluid bolus. The known input / output pairs can indicate parameters of the engine 820, which can be dynamically updatable in some embodiments.

[0085] The framework 800 can also be configured to generate real-time patient responder labels 852 associated with real-time features and / or log data 850 using a trained version of the engine 820. For example, during a medical procedure, real-time features and / or bolus log data 850 associated with a particular patient and / or one or more previous patients can be processed using the engine 820 to generate a real-time responder label 852 (e.g., “recommend” or “test”) that estimates the patient’s response to an additional fluid bolus. In some examples, the engine 820 can be configured to weight and / or bias the features and / or log data provided to the engine 820 to associate patient-specific data with training input data and / or other input data previously analyzed at the engine 820.

[0086] The engine 820 can utilize any of a variety of machine learning methods, which can include logistic regression (LR), support vector machines (SVM), extreme gradient boosting (XGBoost), SVM+LR, and / or XGBoost+LR.

[0087] Additional Examples According to examples, certain actions, events, or functions of any of the processes or algorithms described herein can be performed in a different order, can be added, merged, or entirely omitted. As a result, in certain examples, not all described actions or events are necessary for practice of these processes.

[0088] Example 1 : A method for managing fluid administration to a patient, the method comprising: receiving a physiological signal; determining a plurality of features associated with administering a fluid bolus to a patient derived from the physiological signal; determining a first fluid bolus recommendation based on the plurality of features, the first fluid bolus recommendation indicating a predicted change in a physiological parameter of the patient in response to the fluid bolus; determining a second fluid bolus recommendation based on the first fluid bolus recommendation and a subset of the plurality of features; and providing the second fluid bolus recommendation.

[0089] Example 2: The method of any example herein, particularly example 1, wherein the physiological signal is an arterial blood pressure or a signal proportional to an arterial blood pressure.

[0090] Example 3: The method of any example herein, particularly example 1, further comprising generating the first fluid bolus recommendation based on the predicted change in the physiological parameter.

[0091] Example 4: The method of any example herein, particularly example 1, wherein the first fluid bolus recommendation comprises a recommendation for a normal bolus or a test bolus.

[0092] Example 5: The method of any example herein, particularly example 1, wherein the second fluid bolus recommendation is based on a first subset of the plurality of features.

[0093] Example 6: The method of any example herein, particularly example 5, wherein the second fluid bolus recommendation is based on a second subset of the plurality of features.

[0094] Example 7: The method of any example herein, particularly example 1, wherein the plurality of features comprises one or more of stroke volume, cardiac output, stroke volume variation, stroke volume index, and cardiac index of the patient.

[0095] Example 8: The method of any example herein, particularly example 1, wherein the plurality of features comprises one or more of rate of change of stroke volume, rate of change of cardiac output, rate of change of stroke volume variation, rate of change of stroke volume index, and rate of change of cardiac index.

[0096] Example 9: The method of any example herein, particularly example 1, wherein the second fluid bolus recommendation comprises a recommendation of a normal bolus or a test bolus.

[0097] Example 10: The method of any example herein, particularly example 1, wherein analyzing the subset involves adjusting one or more weights of the subset.

[0098] Example 11: The method of any example herein, particularly example 1, wherein analyzing the subset involves adjusting one or more biases of the subset.

[0099] Example 12: The method of any example herein, particularly example 1, further comprising accessing the plurality of features from one or more sensors attached to the patient.

[0100] Example 13: The method of any example herein, particularly example 1, further comprising providing the second fluid bolus recommendation to a display device.

[0101] Example 14: The method of any example herein, particularly example 1, further comprising providing the second fluid bolus recommendation to a pump system configured to automatically dispense the fluid bolus.

[0102] Example 15: A system comprising: one or more clinical sensors; control circuitry configured to: receive a physiological signal from the one or more clinical sensors; determine a plurality of features associated with administering a bolus of fluid to a patient derived from the physiological signal; determine a first fluid bolus recommendation based on the plurality of features, the first fluid bolus recommendation indicating a predicted change in a physiological parameter of the patient in response to the fluid bolus; determine a second fluid bolus recommendation based on the first fluid bolus recommendation and a subset of the plurality of features; and provide the second fluid bolus recommendation.

[0103] Example 16: The system of any example, including example 15, wherein the physiological signal is an arterial blood pressure or a signal proportional to an arterial blood pressure.

[0104] Example 17: The system of any example, including example 15, wherein the control circuitry is further configured to generate the first fluid bolus recommendation based on the predicted change in the physiological parameter.

[0105] Example 18: The system of any example, including example 15, wherein the second fluid bolus recommendation is based on a first subset of the plurality of features.

[0106] Example 19: The system of any example, including example 18, wherein the second fluid bolus recommendation is based on a second subset of the plurality of features.

[0107] Example 20: The system of any example, including example 15, further comprising a display device, wherein the control circuitry is further configured to provide the second fluid bolus recommendation to the display device.

[0108] Example 21 : A method for administering fluid to a patient, the method comprising: receiving a physiological signal; determining a plurality of features associated with administering a first fluid bolus to a patient derived from the physiological signal; determining a first fluid bolus recommendation based on the plurality of features, the first fluid bolus recommendation indicating a predicted change in a physiological parameter of the patient in response to the first fluid bolus; determining a second fluid bolus recommendation based on the first fluid bolus recommendation and a subset of the plurality of features; and delivering a second fluid bolus to the patient via one or more fluid pumps based at least in part on the second fluid bolus recommendation.

[0109] Example 22: The method of any example, including example 21, wherein the physiological signal is an arterial blood pressure or a signal proportional to an arterial blood pressure.

[0110] Example 23: The method of any example herein, particularly example 21, wherein the first fluid bolus recommendation comprises a recommendation of a normal bolus or a test bolus.

[0111] Example 24: The method of any example herein, particularly example 21, wherein the second fluid bolus recommendation is based on a first subset of the plurality of features.

[0112] Example 25: The method of any example herein, particularly example 24, wherein the second fluid bolus recommendation is based on a second subset of the plurality of features.

[0113] Example 26: The method of any example herein, particularly example 21, wherein the plurality of features comprises one or more of stroke volume, cardiac output, stroke volume variation, stroke volume index, the patient’s cardiac index, rate of change of stroke volume, rate of change of cardiac output, rate of change of stroke volume variation, rate of change of stroke volume index, and rate of change of cardiac index.

[0114] Example 27: The method of any example herein, particularly example 21, wherein the second fluid bolus recommendation comprises a recommendation of a normal bolus or a test bolus.

[0115] Example 28: The method of any example herein, particularly example 21, wherein analyzing the subset involves adjusting one or more weights of the subset.

[0116] Example 29: The method of any example herein, particularly example 21, wherein analyzing the subset involves adjusting one or more biases of the subset.

[0117] Example 30: The method of any example herein, particularly example 21, further comprising accessing the plurality of features from one or more sensors attached to the patient.

[0118] Example 31: A system comprising: one or more clinical sensors; and control circuitry configured to: receive, from the one or more clinical sensors, a physiological signal; determine, from the physiological signal, a plurality of features associated with administration of a fluid bolus to a patient; determine, based on the plurality of features, a first fluid bolus recommendation, the first fluid bolus recommendation indicating a predicted change in a physiological parameter of the patient in response to the fluid bolus; determine, based on the first fluid bolus recommendation and a subset of the plurality of features, a second fluid bolus recommendation; and deliver, via one or more fluid pumps, a second fluid bolus to the patient based at least in part on the second fluid bolus recommendation.

[0119] Example 32: The system of any example herein, particularly example 31, wherein the physiological signal is an arterial blood pressure or a signal proportional to an arterial blood pressure.

[0120] Example 33: The system of any example herein, particularly example 31, wherein the control circuitry is further configured to generate the first fluid bolus recommendation based on the predicted change in the physiological parameter.

[0121] Example 34: The system of any example herein, particularly example 31, wherein the second fluid bolus recommendation is based on a first subset of the plurality of features.

[0122] Example 35: The system of any example herein, particularly example 34, wherein the second fluid bolus recommendation is based on a second subset of the plurality of features.

[0123] Example 36: The system of any example herein, particularly example 31, wherein the plurality of features comprises one or more of stroke volume, cardiac output, stroke volume variation, stroke volume index, cardiac index, rate of change of stroke volume, rate of change of cardiac output, rate of change of stroke volume variation, rate of change of stroke volume index, and rate of change of cardiac index of the patient.

[0124] Example 37: The system of any example herein, particularly example 31, wherein the second fluid bolus recommendation comprises a recommendation of a normal bolus or a test bolus.

[0125] Example 38: The system of any example herein, particularly example 31, wherein determining the second fluid bolus recommendation based on the first fluid bolus recommendation and the subset of the plurality of features involves adjusting one or more weights of the subset of the plurality of features.

[0126] Example 39: The system of any example herein, particularly example 31, wherein determining the second fluid bolus recommendation based on the first fluid bolus recommendation and the subset of the plurality of features involves adjusting one or more biases of the subset of the plurality of features.

[0127] Example 40: The system of any example herein, particularly example 31, wherein the control circuitry is further configured to access the plurality of features from one or more sensors attached to the patient.

[0128] Conditional language, such as “can,” “could,” “might,” “may,” “e.g.,” “for instance,” etc., unless specifically stated otherwise, are understood with the ordinary meaning by a person of ordinary skill in the art, which typically means that a selection is possible among other alternatives. Such conditional language is also intended to convey that a feature, element, or step can or can not be included in the present example, and that the feature, element, or step can or can not be included in other examples. The terms “comprising,” “including,” “having” and the like are synonymous and are used in the sense of the term “including” in the sense of open-endedness, that is, the terms “comprising,” “including,” “having” and the like mean “including, but not limited to,” and are used in the sense of open- endedness, that is, the terms “comprising,” “including,” “having” and the like mean “including, but not limited to.” Also, the term “or” is used in its inclusive sense (and not in its exclusive sense) so that when used, for example, in a list of two or more items, the term “or” means at least one of the items. Conjunctive language such as the phrase “at least one of X, Y, and Z,” unless specifically stated otherwise, is understood to present that an item, term, element, etc. can be either X, Y, or Z. Thus, such conjunctive language is not generally intended to be limiting as to a particular aspect of an example to X, Y, or Z individually.

[0129] It is to be understood that the foregoing example description is intended to be illustrative only and not limiting of the true scope of the present disclosure. For example, while the above description illustrates various examples, it is expressly not intended that any particular example described herein be limited as to its scope unless such description expressly limits or defines it. Although the aspects of the disclosure have been fully described by reference to the drawings, it is to be understood that other changes can be made and equivalents employed, and also that not all of the features and / or aspects of the aspects of the disclosure necessarily have to be used in every implementation. Therefore, the true scope of the disclosure is not to be limited to the examples described herein but is to be accorded the full scope that resides in the art and equivalents thereof. Furthermore, the aspects of the disclosure have been fully described as related to the present examples, it is to be understood that numerous other modifications and changes can be devised by those skilled in the art that will transform the present examples into other examples that fall within the ambit of the aspects of the disclosure as set forth in the claims. More generally, it is intended that all such alterations and changes be considered as equivalents of the aspects of the disclosure described herein, meeting the objectives of the aspects of the disclosure. The disclosure managers expect to meet the objectives of the aspects of the disclosure through the system and method aspects described herein and equivalents thereof. It is therefore intended that the disclosure not be limited to the described examples for carrying out the aspects of the disclosure, but that the full scope of the aspects of the disclosure be the combinations of features that fall within the ambit of the aspects of the disclosure as set forth in the claims and their equivalents, including full use of equivalents with respect to operational steps outlined in the claims.

[0130] It should be understood that certain ordinal terms (e.g., “first” or “second”) can be used to easily refer to elements of a particular example, and do not necessarily indicate physical or chronological order. Thus, as used herein, ordinal terms (e.g., “first,” “second,” “third,” etc.) used to modify an element, such as a structure, a component, an operation, etc., do not necessarily indicate priority or order of the element with respect to another element with a similar or identical name but for the ordinal term. Further, as used herein, the indefinite articles “a” and “an” can indicate “one or more” rather than “one.” Further, an operation performed “based on” a condition or event can also be performed based on one or more other conditions or events that are not explicitly recited.

[0131] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the example belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

[0132] For ease of description, spatially relative terms “external,” “internal,” “upper,” “lower,” “below,” “above,” “vertical,” “horizontal,” and similar terms, can be used herein for the purpose of describing the orientation of one element or component relative to another element or component as shown in the figures. It will be understood that the spatially relative terms are intended to encompass different orientations of the device in use or operation, in addition to the orientations depicted in the figures. For example, in the case of an inversion of the device shown in the figures, the device that is positioned “below” or “under” another device can be placed “above” the other device. Accordingly, the illustrative term “below” can include both lower and upper positions. The device can also be oriented in other directions, and the spatially relative terms can be interpreted differently depending on the orientation.

[0133] Unless explicitly stated otherwise, comparative and / or quantitative terms (such as “less,” “more,” “larger,” etc.) are intended to encompass the concept of equality. For example, “less” can mean “less than” in the strictest mathematical sense, but can also mean “less than or equal to.”

Claims

1. A method for managing the administration of a fluid to a patient, the method comprising: Receive physiological signals; determining a plurality of features derived from the physiological signal associated with administering a fluid bolus to the patient; determining a first fluid bolus recommendation based on the plurality of characteristics, the first fluid bolus recommendation indicative of a predicted change in a physiological parameter of the patient in response to the fluid bolus; determining a second fluid bolus recommendation based on the first fluid bolus recommendation and a subset of the plurality of features; and Providing the second fluid bolus is recommended. 2 . The method according to claim 1 , wherein the physiological signal is arterial blood pressure or a signal proportional to arterial blood pressure. 3 . The method of claim 1 , further comprising generating the first fluid bolus recommendation based on the predicted change in the physiological parameter.

4. The method of claim 1 or claim 2, wherein the first fluid bolus recommendation comprises a recommendation for a normal bolus or a test bolus.

5. The method of claim 1 or claim 2, wherein the second fluid bolus recommendation is based on a first subset of the plurality of features. The method of claim 5 , wherein the second fluid bolus recommendation is based on a second subset of the plurality of features.

7. The method of claim 1 or claim 2, wherein the plurality of characteristics comprises one or more of: the patient's stroke volume, cardiac output, stroke volume variability, stroke volume index, and cardiac index.

8. The method according to claim 1 or claim 2, wherein: The plurality of features includes one or more of: a rate of change of stroke volume, a rate of change of cardiac output, a rate of change of stroke volume variation, a rate of change of stroke volume index, and a rate of change of cardiac index.

9. The method of claim 1 or claim 2, wherein the second fluid bolus recommendation comprises a recommendation for a normal bolus or a test bolus.

10. The method of claim 1 or claim 2, wherein determining a second fluid bolus recommendation based on the first fluid bolus recommendation and a subset of the plurality of features involves adjusting one or more weights of the subset of the plurality of features.

11. The method of claim 1 or claim 2, wherein determining a second fluid bolus recommendation based on the first fluid bolus recommendation and a subset of the plurality of features involves adjusting one or more biases of the subset of the plurality of features.

12. The method of claim 1 or claim 2, further comprising accessing the plurality of features from one or more sensors attached to the patient.

13. The method of claim 1 or claim 2, further comprising providing the second fluid bolus recommendation to a display device.

14. The method of claim 1 or claim 2, further comprising providing the second fluid bolus recommendation to a pump system configured to automatically dispense the fluid bolus.

15. A system comprising: one or more clinical sensors; and A control circuit system configured to: receiving physiological signals from the one or more clinical sensors; determining a plurality of features derived from the physiological signal associated with administering a fluid bolus to the patient; determining a first fluid bolus recommendation based on the plurality of characteristics, the first fluid bolus recommendation indicative of a predicted change in a physiological parameter of the patient in response to the fluid bolus; determining a second fluid bolus recommendation based on the first fluid bolus recommendation and a subset of the plurality of features; and Providing the second fluid bolus is recommended.

16. The system of claim 15, wherein the physiological signal is arterial blood pressure or a signal proportional to arterial blood pressure.

17. The system of claim 15 or claim 16, wherein the control circuitry is further configured to generate the first fluid bolus recommendation based on the predicted change in the physiological parameter.

18. The system of claim 15 or claim 16, wherein the second fluid bolus recommendation is based on a first subset of the plurality of features.

19. The system of claim 18, wherein the second fluid bolus recommendation is based on a second subset of the plurality of features.

20. The system of claim 15 or claim 16, further comprising a display device, wherein the control circuitry is further configured to provide the second fluid bolus recommendation to the display device.

21. A method for administering a fluid to a patient, the method comprising: Receive physiological signals; determining a plurality of characteristics derived from the physiological signal associated with administering a first fluid bolus to the patient; determining a first fluid bolus recommendation based on the plurality of characteristics, the first fluid bolus recommendation indicative of a predicted change in a physiological parameter of the patient in response to the first fluid bolus; determining a second fluid bolus recommendation based on the first fluid bolus recommendation and a subset of the plurality of features; and Based at least in part on the second fluid bolus recommendation, a second fluid bolus is delivered to the patient via one or more fluid pumps.

22. The method of claim 21, wherein the physiological signal is arterial blood pressure or a signal proportional to arterial blood pressure.

23. The method of claim 21 or claim 22, wherein the first fluid bolus recommendation comprises a recommendation for a normal bolus or a test bolus.

24. The method of claim 21 or claim 22, wherein the second fluid bolus recommendation is based on a first subset of the plurality of features.

25. The method of claim 24, wherein the second fluid bolus recommendation is based on a second subset of the plurality of features.

26. The method of claim 21 or claim 22, wherein: The multiple characteristics include one or more of the following: the patient's stroke volume, cardiac output, stroke volume change, stroke volume index, cardiac index, rate of change of stroke volume, rate of change of cardiac output, rate of change of stroke volume change, rate of change of stroke volume index and rate of change of cardiac index.

27. The method of claim 21 or claim 22, wherein the second fluid bolus recommendation comprises a recommendation for a normal bolus or a test bolus.

28. The method of claim 21 or claim 22, wherein determining a second fluid bolus recommendation based on the first fluid bolus recommendation and a subset of the plurality of features involves adjusting one or more weights of the subset of the plurality of features.

29. The method of claim 21 or claim 22, wherein determining a second fluid bolus recommendation based on the first fluid bolus recommendation and a subset of the plurality of features involves adjusting one or more biases of the subset of the plurality of features.

30. The method of claim 21 or claim 22, further comprising accessing the plurality of features from one or more sensors attached to the patient.

31. A system comprising: one or more clinical sensors; and A control circuit system configured to: receiving physiological signals from the one or more clinical sensors; determining a plurality of features derived from the physiological signal associated with administering a fluid bolus to the patient; determining a first fluid bolus recommendation based on the plurality of characteristics, the first fluid bolus recommendation indicative of a predicted change in a physiological parameter of the patient in response to the fluid bolus; determining a second fluid bolus recommendation based on the first fluid bolus recommendation and a subset of the plurality of features; and Based at least in part on the second fluid bolus recommendation, a second fluid bolus is delivered to the patient via one or more fluid pumps.

32. The system of claim 31, wherein the physiological signal is arterial blood pressure or a signal proportional to arterial blood pressure.

33. The system of claim 31 or claim 32, wherein the control circuitry is further configured to generate the first fluid bolus recommendation based on the predicted change in the physiological parameter.

34. The system of claim 31 or claim 32, wherein the second fluid bolus recommendation is based on a first subset of the plurality of features.

35. The system of claim 34, wherein the second fluid bolus recommendation is based on a second subset of the plurality of features.

36. A system according to claim 31 or claim 32, wherein the plurality of characteristics comprises one or more of the following: the patient's stroke volume, cardiac output, stroke volume change, stroke volume index, cardiac index, rate of change of stroke volume, rate of change of cardiac output, rate of change of stroke volume change, rate of change of stroke volume index and rate of change of cardiac index.

37. The system of claim 31 or claim 32, wherein the second fluid bolus recommendation comprises a recommendation for a normal bolus or a test bolus.

38. The system of claim 31 or claim 32, wherein determining a second fluid bolus recommendation based on the first fluid bolus recommendation and a subset of the plurality of features involves adjusting one or more weights of the subset of the plurality of features.

39. The system of claim 31 or claim 32, wherein determining a second fluid bolus recommendation based on the first fluid bolus recommendation and a subset of the plurality of features involves adjusting one or more biases of the subset of the plurality of features.

40. The system of claim 31 or claim 32, wherein the control circuitry is further configured to access the plurality of features from one or more sensors attached to the patient.

Images