Method and apparatus for predicting decoupling during non-cardiac medical procedure

Abstract

Methods and apparatus are provided for predicting decoupling during a non-cardiac medical procedure. The method includes receiving one or more patient characteristics associated with a patient scheduled to perform a non-cardiac medical procedure; providing the one or more patient characteristics as inputs to a machine learning model trained to output a decoupling prediction; processing, using at least one computer processor, the one or more patient characteristics with a machine learning model to output a decoupled prediction for the patient, where the decoupled prediction is associated with a non-cardiac medical procedure; and displaying, on a user interface, an indication of the decoupling prediction for the patient output from the machine learning model.

Description

Technical Field

[0001] This disclosure relates to techniques for predicting cardiac decoupling during non-cardiac medical procedures. Background Technology

[0002] Cardiovascular disease is a leading cause of morbidity, mortality, and the global burden of healthcare. A wide variety of treatments for heart health have been developed, ranging from medications to mechanical devices and transplants. Temporary cardiac support devices, such as cardiac pump systems (also known as “intracardiac blood pumps”), provide hemodynamic support and facilitate cardiac recovery. Intracardiac blood pumps have traditionally been used to temporarily assist a patient’s heart pumping function during emergency cardiac procedures such as stent placement performed after a heart attack, cardiac arrest, and / or cardiogenic shock. Intracardiac blood pumps can also be used to relieve the workload on a patient’s heart, allowing the heart to recover from such cardiac procedures or from heart attacks, cardiac arrest, cardiogenic shock, or cardiac injury (e.g., caused by viral infections). In this regard, intracardiac blood pumps can be surgically or percutaneously introduced into the heart and used to deliver blood from one location in the heart or circulatory system to another. For example, when deployed in the left side of the heart, an intracardiac blood pump can pump blood from the left ventricle of the heart into the aorta. Similarly, when deployed in the right heart, an intracardiac pump can pump blood from the inferior vena cava into the pulmonary artery. The intracardiac pump can be powered by an external motor via a thin drive shaft (or drive cable) or by an onboard motor located inside the patient. Examples of such devices include the Impella® series (Abiomed, Inc., Danvers, Massachusetts). Summary of the Invention

[0003] Intracardiac pumps can be used to provide cardiac support to patients who would otherwise be at high risk (e.g., due to their age, medical history, comorbidities, etc.). For example, in some cases, patients who require a medical procedure (who may have been refused the procedure based on the risk of adverse consequences (e.g., the procedure itself may cause hemodynamic instability and / or death during and / or after the procedure)) may become eligible for such a procedure if the risk is reduced by using an intracardiac pump before, during, or after the procedure.

[0004] When using an intracardiac pump, decoupling occurs when a patient's heart relies on the pump rather than its natural cardiac function for support. Some embodiments of this disclosure relate to techniques for analyzing pressure signals within a patient's heart to detect decoupling during intracardiac pump use. Some embodiments of this disclosure relate to using patient characteristics (e.g., age, sex, cardiac function, etc.) of patients who experience decoupling during non-cardiac surgery to inform and / or train a predictive algorithm or model about the likelihood that other patients will experience decoupling during non-cardiac surgery. Such predictive algorithms or models can be used, among other things, particularly as clinical decision support tools to provide healthcare providers with recommendations regarding the efficacy of using an intracardiac pump during non-cardiac medical procedures (e.g., non-cardiac surgery). In some embodiments, the output of the predictive algorithm / model can be used as a predictor of a patient's chances of survival without intracardiac pump support.

[0005] In some embodiments, a computer-implemented method is provided. The computer-implemented method includes: receiving one or more patient characteristics associated with a patient scheduled for a non-cardiac medical procedure; providing the one or more patient characteristics as input to a machine learning model trained to output a decoupling prediction; using at least one computer processor, utilizing the machine learning model to process the one or more patient characteristics to output a decoupling prediction for the patient, wherein the decoupling prediction is associated with the non-cardiac medical procedure; and displaying an indication of the decoupling prediction for the patient output from the machine learning model on a user interface.

[0006] In one aspect, the one or more patient characteristics include one or more of the following: physical characteristics, medical history information, medication information, or physiological indicators associated with the patient. In another aspect, receiving one or more patient characteristics includes receiving the one or more patient characteristics from an electronic medical record associated with the patient. In another aspect, the machine learning model is trained using training data from multiple patients who have undergone different types of non-cardiac medical procedures. In another aspect, displaying indications of decoupling predictions for the patient includes displaying a predicted amount of decoupling that the patient may experience during the non-cardiac medical procedure.

[0007] In another aspect, the non-cardiac medical procedure has an expected procedure duration, and the displayed predicted decoupling amount includes displaying the predicted decoupling amount as a percentage of the expected procedure duration. In another aspect, the computer-implemented method further includes determining that the patient would benefit from using an intracardiac blood pump associated with the non-cardiac medical procedure when the predicted decoupling amount exceeds a threshold, and displaying a recommendation associated with using the intracardiac blood pump on the user interface. In another aspect, the recommendation includes a time period during which using the intracardiac blood pump would be beneficial to the patient. In another aspect, the time period includes one or more of the following: before, during, or after the non-cardiac medical procedure.

[0008] In some embodiments, a controller for a heart pump system is provided. The controller includes at least one hardware processor configured to: receive one or more patient characteristics associated with a patient scheduled for a non-cardiac medical procedure; provide the one or more patient characteristics as input to a machine learning model trained to output a decoupling prediction for the patient, wherein the decoupling prediction is associated with the non-cardiac medical procedure; and display an indication of the decoupling prediction for the patient output from the machine learning model on a user interface associated with the heart pump system.

[0009] In some embodiments, a method of treating a patient is provided. The method includes: receiving a decoupled prediction for a patient scheduled for a non-cardiac medical procedure, wherein the decoupled prediction is output from a trained machine learning model in response to providing one or more patient characteristics associated with the patient as input; determining, based on the decoupled prediction, a time period during which the patient would benefit from using an intracardiac blood pump associated with the non-cardiac medical procedure; inserting the intracardiac blood pump into the patient's heart during the determined time period; and performing the non-cardiac medical procedure to treat the patient.

[0010] In one aspect, the one or more patient characteristics include one or more of the following: physical characteristics associated with the patient, medical history information, medication information, or physiological indicators. In another aspect, the one or more patient characteristics are provided as input from electronic medical records associated with the patient to the trained machine learning model. In another aspect, the trained machine learning model is trained using training data from multiple patients who have undergone different types of non-cardiac medical procedures. In another aspect, the time period includes one or more of the following: before, during, or after the non-cardiac medical procedure. In another aspect, the method further includes: tracking the amount of decoupling experienced by the patient during the non-cardiac medical procedure; and retraining the trained machine learning model based at least in part on the amount of decoupling experienced by the patient and the one or more patient characteristics.

[0011] In some embodiments, a computer-implemented method is provided for training a machine learning model to predict decoupling during non-cardiac medical procedures. The computer-implemented method includes: receiving patient data from a plurality of patients, each of whom has undergone a non-cardiac medical procedure associated with an intracardiac blood pump; analyzing the patient data by at least one hardware processor to detect one or more decoupling events during the non-cardiac medical procedure associated with each of the plurality of patients; training the machine learning model based at least in part on the one or more decoupling events detected during the non-cardiac medical procedure and one or more patient characteristics associated with the patient associated with the non-cardiac medical procedure; and outputting the trained machine learning model for predicting decoupling in one or more patients not associated with the patient data.

[0012] In one aspect, the patient data includes one or more patient characteristics associated with the patient, and pressure information sensed and / or derived therefrom by one or more pressure sensors of a mechanical circulatory support device including the intracardiac pump. In another aspect, the pressure information includes left ventricular pressure signals and aortic pressure signals. In another aspect, analyzing the patient data to detect one or more decoupling events includes detecting a decoupling event when the peak value of the aortic pressure signal is greater than the peak value of the left ventricular pressure signal during a cardiac cycle. In another aspect, the method further includes determining the amount of decoupling of the patient during the non-cardiac medical procedure based on the detected one or more decoupling events, and training the machine learning model at least in part based on the one or more decoupling events includes training the machine learning model at least in part based on the amount of decoupling of the patient during the non-cardiac medical procedure. In another aspect, receiving patient data from multiple patients includes receiving the patient data from corresponding electronic medical records associated with the multiple patients. Attached Figure Description

[0013] Figure 1A The illustration shows a pump system according to some embodiments of the present disclosure.

[0014] Figure 1B yes Figure 1A A cross-sectional view of a portion of the pump system.

[0015] Figure 2 The illustration shows a pump system according to some embodiments of the present disclosure.

[0016] Figure 3 The illustration shows a pump system according to some embodiments of the present disclosure.

[0017] Figure 4 This is a flowchart of a process for updating an algorithm or model according to some embodiments of the present disclosure, the algorithm or model being used to predict decoupling during non-cardiac procedures.

[0018] Figure 5 The illustration shows example pressure signals that can be used to detect decoupled stress signals during medical procedures, according to some embodiments of the present disclosure.

[0019] Figure 6 The illustration depicts a process for detecting decoupling during a medical procedure using a comparison of pressure signals, according to some embodiments of the present disclosure.

[0020] Figure 7 This is a flowchart of a process for predicting the amount of decoupling during non-cardiac medical procedures using updated decoupling algorithms / models, according to some embodiments of this disclosure.

[0021] Figure 8This is a flowchart of the process for using an intracardiac blood pump to reduce cardiac risks during non-cardiac medical procedures. Detailed Implementation

[0022] Using an intracardiac blood pump to provide cardiac support to a patient during non-cardiac medical procedures (e.g., non-cardiac surgeries such as cyst removal, knee replacement, weight-loss surgery, etc.) can reduce the risk of the patient experiencing cardiac-related adverse events during the non-cardiac medical procedure. By reducing such risk, the patient may become eligible for procedures that, without the use of a blood pump, might be considered too risky by a healthcare provider. Determining whether the cardiac benefits of using an intracardiac blood pump during a non-cardiac medical procedure will be sufficient to outweigh the risks involved in placing the device in the patient's heart before, during, and / or after the non-cardiac medical procedure may depend on several factors. Decoupling is a state in which the heart's inherent function of ejecting blood through the heart valves is significantly reduced, such that blood circulation through the patient's heart is maintained primarily or entirely by a continuous flow of blood pumped by an intracardiac blood pump placed in the patient's heart. Some embodiments of this disclosure relate to predicting whether and / or to what extent decoupling may occur during a particular patient's non-cardiac medical procedure, based at least in part on one or more patient characteristics. When deciding whether the use of an intracardiac blood pump would be beneficial before, during, and / or after a non-cardiac medical procedure, physicians or other healthcare providers can be better informed by estimating the likelihood and / or extent of decoupling during a possible non-cardiac surgery.

[0023] To provide a comprehensive understanding of the systems, methods, and apparatus described herein, certain illustrative examples will be described. While the various examples may describe specific medical procedures and / or the use of intracardiac blood pumps, it will be understood that the techniques described herein can be employed in any suitable context.

[0024] exist Figure 1A and 1BA pump system including a pump 100 is illustrated for use in some embodiments of this disclosure. As shown, the pump 100 may be coupled to a control unit 170. The pump 100 may include a distal abrasion-free tip 102, a pump housing 104 surrounding a rotor 108, an outlet tube 106, a distal bearing 110, a proximal bearing 112, an inlet 116, an outlet 118, a conduit 120, a handle 130, a cable 140, and a motor 150. As will be understood, although an abrasion-free tip is shown, in some embodiments, the pump may not include such a tip. The pump housing 104 may be configured as a frame structure formed by a mesh with openings, which may be at least partially covered by an elastic material. As will be understood, although shown as a frame structure, in some embodiments, the pump housing may be solid. The proximal portion of the pump housing 104 may extend into and be mounted within the hollow interior of the outlet tube 106, and the distal portion of the pump housing 104 may extend distally beyond the distal end of the outlet tube 106. An exposed opening in the pump housing 104 extending distally beyond the outlet tube 106 can form an inlet 116 for the pump 100. The proximal end of the outlet tube 106 may include multiple openings forming an outlet 118 for the pump 100. A rotor 108 is rotatably mounted between a distal bearing 110 and a proximal bearing 112 and can be coupled to the distal end of a drive shaft 114. The drive shaft 114 can be flexible and can extend through a conduit 120, through the hollow interior of the outlet tube 106, into a handle 130, and be coupled to a motor 150 housed within the handle 130. The proximal end of the handle 130 can be coupled to a control unit 170 via a cable 140. Fluid can circulate through the conduit 120 in the vicinity of the drive shaft 114 and in the space around the distal bearing 110 and the proximal bearing 112 to lubricate those components and reduce friction during operation of the pump 100.

[0025] Control unit 170 may include one or more memories 172, one or more processors 174, a user interface 176, and one or more sensors, such as a current sensor 178. The processors 174 may include one or more microcontrollers, one or more microprocessors, one or more application-specific integrated circuits (ASICs), one or more digital signal processors, program memory, or other computing units. The processors 174 may be communicatively coupled to other components of control unit 170 (e.g., memory 172, user interface 176, one or more current sensors 178) and may be configured to control one or more operations of pump 100. As will be understood, although control unit 170 is shown as connected to pump 100, the control unit may be connected to a combination of Figure 2 and Figure 3The pump 200 or pump 300 is described. As a non-limiting example, the control unit 170 can be implemented as an automatic Impella controller from ABIOMED Ltd., Danvers, Massachusetts. TM In some respects, memory 172 is included as part of processor(s) 174, rather than being provided as a separate component.

[0026] During operation, one or more processors 174 may be configured to control the power delivered to motor 150 via a power supply line (not shown) in cable 140 (e.g., by controlling a power supply (not shown)), thereby controlling the speed of motor 150. One or more current sensors 178 may be configured to sense motor current associated with the operating state of motor 150, and one or more processors 174 may be configured to receive the output of one or more current sensors 408 as a motor current signal. One or more processors 174 may be further configured to determine the flow rate through pump 100 based at least in part on the motor current signal and motor speed, as described in more detail herein. One or more current sensors 178 may be included in control unit 170, or may be located along any portion of the power supply line in cable 140. Alternatively or additionally, one or more current sensors 178 may be included in motor 150, and one or more processors 174 may be configured to receive the motor current signal via a data line (not shown) in cable 140 coupled to one or more processors 174 and motor 150.

[0027] The memory 172 can be configured to store computer-readable instructions and other information for various functions of the components of the control unit 170. In one aspect, the memory 172 includes volatile and / or non-volatile memory, such as electrically erasable programmable read-only memory (EEPROM).

[0028] User interface 176 can be configured to receive user input via one or more buttons, switches, knobs, etc. Additionally, user interface 176 may include a display configured to show information, and one or more indicators (such as light indicators, audio indicators, etc.) for conveying information and / or providing alarms regarding the operation of pump 100.

[0029] Pump 100 can be designed to be inserted into a patient's body, such as via an infusion system, for example, into the left ventricle of the heart. Although some of the systems and / or methods disclosed herein are described as being for regulating the pump rate of a pump inserted into the left ventricle of the heart, it should be understood that the systems and / or methods described herein can also be applied to other types of ventricular support systems, such as ventricular support systems inserted into the right ventricle of the heart. In one aspect, housing 104, rotor 108, and outflow tube 106 can be radially compressible so that pump 100 can achieve a relatively small outer diameter, for example, 9 Fr (3 mm), during insertion. When pump 100 is inserted into a patient's body, for example, into the left ventricle, handle 130 and motor 150 can remain outside the patient's body. As will be understood, in other embodiments, the motor of the pump system can be inside the patient's body during insertion. During operation, motor 150 is controlled by one or more processors 404 to drive drive shaft 114 and rotor 108 to rotate to deliver blood from inlet 116 to outlet 118. To be understood, rotor 108 can be rotated in the opposite direction by motor 150 to deliver blood in the opposite direction (in which case, opening 118 forms an inlet and opening 116 forms an outlet). In one aspect, pump 100 can be configured to be used for weeks to months to years to support cardiac function in patients with chronic heart failure; however, it should be understood that the techniques described herein are not limited to any particular type of procedure and / or duration of use.

[0030] Figure 2 The diagram illustrates a blood pump system used in some embodiments. Figure 2 An exemplary intracardiac blood pump assembly 200 suitable for left ventricular support is depicted according to aspects of this disclosure. Figure 2 As shown, an intracardiac blood pump assembly suitable for left ventricular support may include an elongated catheter 202, a motor 204, a cannula 210, a blood inflow cage 214 disposed at or near the distal end 212 of the cannula 210, a blood outflow cage 206 disposed at or near the proximal end 208 of the cannula 210, and an optional non-invasive extension 216 disposed at the distal end of the blood inflow cage 214.

[0031] In some aspects of this technology, motor 204 can be configured to rotatably drive an impeller (not shown) to generate sufficient suction to draw blood into cannula 210 through blood inflow cage 214 and expel blood from cannula 210 through blood outflow cage 206. In this respect, the impeller can be positioned distal to the blood outflow cage 206, for example, within the proximal end 208 of cannula 210 or within a pump housing 207 coupled to the proximal end 208 of cannula 210. In some aspects of this technology, instead of being driven by an onboard motor 204, the impeller can be coupled to an elongated drive shaft (or drive cable) driven by a motor located outside the patient's body.

[0032] The conduit 202 may house the wires that couple the motor 204 to one or more electrical controllers and / or sensors. Alternatively, in the case where the impeller is driven by an external motor, an elongated drive shaft may pass through the conduit 202. The conduit 202 may also include a cleaning fluid conduit, an inner cavity configured to accommodate a guide wire, etc.

[0033] The blood inflow cage 214 may include one or more orifices or openings configured to allow blood to be drawn into the cannula 210 when the motor 204 is in operation. Similarly, the blood outflow cage 206 may include one or more orifices or openings configured to allow blood to flow out of the cannula 210 from the intracardiac pump assembly 200. The blood inflow cage 214 and the blood outflow cage 206 may be made of one or more suitable biocompatible materials. For example, the blood inflow cage 214 and / or the blood outflow cage 206 may be formed from biocompatible metals such as stainless steel or titanium, or biocompatible polymers such as polyurethane. Furthermore, the surfaces of the blood inflow cage 214 and / or the blood outflow cage 206 may be treated in various ways, including but not limited to etching, texturing, or coating or plating with another material. For example, the surfaces of the blood inflow cage 214 and / or the blood outflow cage 206 may be laser-textured.

[0034] The sleeve 210 may include a flexible hose portion. For example, the sleeve 210 may be at least partially composed of a polyurethane material. Furthermore, the sleeve 210 may include a shape memory material. For example, the sleeve 210 may include a combination of a polyurethane material and one or more strands or turns of a shape memory material (such as nitinol). The sleeve 210 may be formed such that it includes one or more bends or curves in its relaxed state, or it may be configured to be straight in its relaxed state. In this respect, as... Figure 2 As shown in the exemplary arrangement, the cannula 210 may have a single pre-formed anatomical bend 218 based on the portion of the left heart in which it is intended to be operated. Despite this bend 218, the cannula 210 may still be flexible and therefore may be able to straighten (e.g., during insertion via a guidewire) or be further bent (e.g., in patients with a tighter anatomical structure). Furthermore, at this point, the cannula 210 may include a shape memory material configured to allow the cannula 210 to be of different shapes at room temperature (e.g., straight or substantially straight), and the bend 218 forms once the shape memory material is exposed to heat from the patient's body.

[0035] The non-invasive extension 216 can assist in stabilizing and positioning the intracardiac pump assembly 200 in the correct location within the patient's heart. The non-invasive extension 216 can be solid or tubular. If tubular, the non-invasive extension 216 can be configured to allow a guidewire to pass through it to further assist in the positioning of the intracardiac pump assembly 200. The non-invasive extension 216 can be of any suitable size. For example, the non-invasive extension 216 can have an outer diameter ranging from 4 to 8 Fr. The non-invasive extension 216 can be at least partially constructed of a flexible material and can be of any suitable shape or construction, such as a straight construction, a partially curved construction, or... Figure 2 Examples include pigtail-shaped structures. The non-traumatic extension 216 can also have segments with varying degrees of stiffness. For instance, the non-traumatic extension 216 may include a proximal segment that is stiff enough to prevent bending, thus holding blood flow into the cage 214 in the desired position, and a softer, less stiff distal segment that provides a non-traumatic tip for contact with the patient's heart wall and allows for guidewire loading. In this case, the proximal and distal segments of the non-traumatic extension 216 may be composed of different materials, or they may be composed of the same material (where the proximal and distal segments are treated to provide different stiffness).

[0036] Although mentioned above, the non-invasive extension 216 is an optional configuration. In this respect, this technology can also be used with intracardiac pump assemblies and other intracardiac devices that include extensions of different types, shapes, materials, and qualities. Similarly, this technology can be used with intracardiac pump assemblies and other intracardiac devices that do not have any kind of distal extension.

[0037] As described herein, the intracardiac pump assembly 200 can be percutaneously inserted. For example, when used for left ventricular support, the intracardiac pump assembly 200 can be inserted into the aorta via a catheterization procedure through the femoral or axillary artery, through the aortic valve, and into the left ventricle. Once positioned in this manner, the intracardiac pump assembly 200 can deliver blood from the blood inflow cage 214 located inside the left ventricle to the blood outflow cage 206 located inside the ascending aorta via cannula 210. In some aspects of this technology, the intracardiac pump assembly 200 can be configured such that when the intracardiac pump assembly 200 is in the desired position, the bend 218 will abut against a predetermined portion of the patient's heart. Similarly, the non-invasive extension 216 can be configured such that when the intracardiac pump assembly 200 is in the desired position, it abuts against different predetermined portions of the patient's heart.

[0038] Figure 3 An exemplary intracardiac blood pump assembly 300 suitable for right heart support is depicted according to aspects of this disclosure. (See also:) Figure 3As shown, an intracardiac blood pump assembly suitable for right heart support may include an elongated catheter 302, a motor 304, a cannula 310, a blood inflow cage 314 disposed at or near the proximal end 308 of the cannula 310, a blood outflow cage 306 disposed at or near the distal end 312 of the cannula 310, and an optional non-invasive extension 316 disposed at the distal end of the blood outflow cage 306.

[0039] as Figure 2 Similar to the exemplary components, motor 304 can be configured to rotatably drive an impeller (not shown) to generate sufficient suction to draw blood into cannula 310 through blood inflow cage 314 and expel blood from cannula 310 through blood outflow cage 306. In this respect, the impeller can be positioned distal to the blood inflow cage 314, for example, within the proximal end 308 of cannula 310 or within a pump housing 307 coupled to the proximal end 308 of cannula 310. Again, in some aspects of this technology, instead of being driven by an onboard motor 304, the impeller can be coupled to an elongated drive shaft (or drive cable) driven by a motor located outside the patient's body.

[0040] Figure 3 The sleeve 310 can be used for the same purpose and can have the same features as described above. Figure 2 The same properties and characteristics are described for sleeve 210. However, as Figure 3 As shown in the exemplary arrangement, the cannula 310 may have two pre-formed anatomical bends 318 and 320 based on the portion of the right heart in which it is intended to be operated. Again, despite the presence of bends 318 and 320, the cannula 310 may still be flexible and therefore may be able to straighten (e.g., during insertion via a guidewire) or be further bent (e.g., in patients with a tighter anatomical structure). Furthermore, at this point, the cannula 310 may include a shape memory material configured to allow the cannula 310 to be of different shapes at room temperature (e.g., straight or substantially straight), and the bends 318 and / or 320 form once the shape memory material is exposed to heat from the patient's body.

[0041] Figure 3 The catheter 302 and the non-invasive extension 316 can be used for the same purpose and can have the same features as described above. Figure 2 The catheter 202 and the non-invasive extension 216 have the same properties and characteristics as described. Similarly, except that they are located in the same... Figure 2 Beyond the ends of those sleeves relative to each other, Figure 3 The blood flowing into cage 314 and the blood flowing out of cage 306 can be similar to Figure 2Blood flows into cage 214 and blood flows out of cage 206, and therefore can have the same properties and characteristics as described above.

[0042] and Figure 2 Similar to the exemplary components, Figure 3 The intracardiac pump assembly 300 can also be percutaneously inserted. For example, when used for right ventricular support, the intracardiac pump assembly 300 can be inserted via a catheterization procedure through the femoral vein into the inferior vena cava, through the right atrium, across the tricuspid valve, into the right ventricle, through the pulmonary valve, and into the pulmonary artery. Once positioned in this manner, the intracardiac pump assembly 300 can deliver blood from the blood inflow cage 314 located inside the inferior vena cava to the blood outflow cage 306 located inside the pulmonary artery via a cannula 310.

[0043] Figure 4 A flowchart illustrating a process 400 for generating and / or updating an algorithm or model to output a decoupling prediction, according to some embodiments of the present disclosure, is shown. Process 400 may begin at action 410, in which patient data is received. The received patient data may be associated with a patient who has undergone a medical procedure (e.g., non-cardiac surgery) during which an intracardiac blood pump was used. Patient data may be received in any suitable manner. For example, in some embodiments, patient data may be received from one or more electronic patient medical records (e.g., electronic health records (EHRs)). Patient data may include one or more patient characteristics and pressure information sensed and / or derived from one or more pressure sensors of an MCS device including an intracardiac blood pump. Pressure information may include left ventricular pressure (LVP) information and aortic pressure (AOP) information, which may be used to detect decoupling during medical procedures, as discussed in further detail herein.

[0044] Process 400 can then proceed to action 420, where patient data can be analyzed to detect decoupling during non-cardiac procedures. Figure 5 The illustration shows an example of pressure information that can be included as patient data in action 410 and analyzed in action 420 to detect decoupling. Figure 5 As shown, the pressure information includes AOP information 510 and LVP information 520. Each of AOP information 510 and LVP information 520 is shown as a pressure waveform sensed over time by one or more pressure sensors. Figure 5In the example shown, AOP information 510 is sensed by an optical pressure sensor disposed on the intracardiac pump, and LVP information 520 is derived based on AOP information 510 and other information (e.g., the motor current and / or motor speed of the intracardiac pump). However, it should be understood that more than one sensor may be used to sense pressure information, and the embodiments of this disclosure are not limited in this respect.

[0045] Figure 6 Example techniques for detecting decoupling during non-cardiac medical procedures are illustrated schematically according to some embodiments of this disclosure. Figure 6 In the example shown, a peak-to-peak comparison of the AOP signal 610 and the LVP signal 620 is performed during a cardiac cycle, and decoupling is detected when the peak AOP signal is greater than the peak LVP signal during the cardiac cycle. Figure 6 The example portion of the waveform where the peak AOP signal is greater than the peak LVP signal is shown using bounding box 630. Figure 6 The start and end times of non-cardiac medical procedures (e.g., non-cardiac surgery) are also shown, collectively forming a procedure window. In some embodiments, such as indicator 640, the amount of decoupling detected within the procedure window (e.g., percentage) is shown. As described herein, a higher percentage of decoupling within the procedure window may indicate increased patient dependence on intracardiac pumps during the procedure. It should be understood that although only pressure data for a single patient is shown, the process of detecting decoupling based on pressure signals in received patient data can be performed on multiple patients who have undergone the same or different non-cardiac procedures.

[0046] Returning to process 400, after analyzing patient data to detect decoupling in action 420, process 400 may proceed to action 430, where a decoupling prediction algorithm or model may be updated, at least in part, based on the decoupling detection and one or more patient characteristics included in the patient data received in action 410. In some embodiments, the decoupling prediction algorithm / model may be updated to correlate (e.g., by setting specific weights) various patient characteristics (e.g., age, sex, weight, history of cardiac events, baseline ejection fraction, required medications, medical history, stress values ​​measured from previous medical procedures, ECG values, etc.) to provide a decoupling prediction as output. The decoupling prediction can take any suitable form. For example, in some embodiments, the decoupling prediction may include a number (e.g., a percentage) associated with a predicted amount of decoupling for patients with certain characteristics and / or for patients who have undergone different types of non-cardiac medical procedures. In some embodiments, the decoupling prediction may include recommending the use of an intracardiac pump for non-cardiac medical procedures based on the predicted amount of decoupling (e.g., when the predicted amount of decoupling is above a certain threshold). In other embodiments, decoupling prediction may include a number (e.g., a percentage) associated with the predicted chance of decoupling during non-cardiac medical procedures.

[0047] When trained on a large amount of patient data, the algorithm / model can learn (e.g., by changing the weights in the algorithm / model) to predict patient characteristics of decoupling amounts during non-cardiac procedures. It should be understood that when implemented as a machine learning model, the decoupling prediction model can be implemented as any suitable type of model, examples of which include neural network-based models (e.g., deep learning), random forest classifier models, decision tree models (e.g., gradient boosting decision tree models), or logistic regression models. Process 400 can then proceed to action 440, where the updated model is output for use in predicting the decoupling amount of a patient not included in the algorithm / model's updated patient data.

[0048] Figure 7This is a flowchart of a process 700 for predicting the amount of decoupling for a patient scheduled for a non-cardiac procedure (e.g., non-cardiac surgery). Process 700 begins at action 710, in which an indication for predicting the amount of decoupling for the non-cardiac procedure is received. Process 700 may then proceed to action 720, in which one or more patient characteristics (e.g., age, sex, weight, history of cardiac events, baseline ejection fraction, required medications, medical history, stress values ​​measured from previous medical procedures, ECG values, etc.) are provided as input to an algorithm or model (e.g., a machine learning model) trained to output a decoupling prediction for the patient. In some embodiments, one or more patient characteristics from an electronic medical record associated with the patient may be provided as input to the algorithm / model. In some embodiments, the decoupling prediction may indicate the amount of decoupling expected to occur during the non-cardiac medical procedure. Process 700 may then proceed to action 730, in which the indication of the decoupling prediction output by the algorithm / model may be displayed to a user, such as a physician. In some embodiments, decoupling prediction may include recommendations regarding whether the use of an intracardiac blood pump would be beneficial for patients associated with non-cardiac medical procedures. In some embodiments, such recommendations may include indications of the time periods during which the intracardiac blood pump should be used (e.g., before, during, and / or after non-cardiac medical procedures).

[0049] It should be understood that decoupling prediction for a patient can be performed at any appropriate time using one or more of the techniques described herein. For example, in some embodiments, decoupling prediction can be performed when a non-cardiac procedure is scheduled. Such information can enable physicians to determine whether an intracardiac pump should be used during a non-cardiac procedure to reduce the risk of cardiac complications the patient will have during that procedure. For example, the predicted amount of decoupling can be used to assess the likelihood of a patient's survival with or without the use of an intracardiac pump during a non-cardiac medical procedure.

[0050] Figure 8 This is a flowchart of process 800 for treating a patient using an intracardiac blood pump associated with a non-cardiac medical procedure. In this respect, if, according to process 700, it has been determined that the patient will benefit from the use of an intracardiac blood pump during a non-cardiac medical procedure (e.g., because decoupling prediction indicates that the expected amount of decoupling is above a certain threshold), then process 800 can be performed.

[0051] In action 810, a time period can be defined during which the patient will benefit from receiving support from the intracardiac pump. This time period can be one or more of the following: before, during, and after a non-cardiac medical procedure. At this point, the intracardiac pump can be used in a variety of ways to reduce and / or eliminate the risk of the patient experiencing cardiac events during or after the medical procedure. For example, the intracardiac pump can be used before a medical procedure to allow the heart to rest before the procedure, potentially reducing the risk of the heart subsequently being overwhelmed by the trauma of the procedure. Similarly, the intracardiac pump can be used during a medical procedure to reduce the load on the heart and maintain blood flow through the body, potentially reducing the risk of ventricular fibrillation, worsening ischemia, myocardial ischemia, pulmonary edema, hemodynamic failure, cardiac arrest, death, etc., which may occur during the procedure. Furthermore, the intracardiac pump can be used after a medical procedure to allow the heart to recover and thus mitigate the risk of postoperative cardiac events such as heart attack, ventricular fibrillation, worsening ischemia, myocardial ischemia, pulmonary edema, hemodynamic failure, cardiac arrest, death, etc. Therefore, the intracardiac pump can be used as follows, depending on the circumstances: (a) only before the procedure; (b) before and during the procedure; (c) before, during and after the procedure; (d) only before and after the procedure, but not during the procedure; (e) only during the procedure; (f) only during and after the procedure; or (g) only after the procedure.

[0052] In action 820, an intracardiac pump may be inserted into the patient to provide cardiac support for the time period defined in action 810. In this regard, any suitable method of inserting, positioning, and providing cardiac support using an intracardiac pump may be used, including the methods described herein.

[0053] In action 830, a medical procedure can be performed. In some aspects of this technology, actions 820 and 830 can occur simultaneously, or their order can be the reverse of that shown in exemplary procedure 800. For example, if it is determined in action 810 that the intracardiac pump will only be used after the medical procedure, the medical procedure (action 830) can occur before inserting the intracardiac pump (action 820). Similarly, if the intracardiac pump is to be used during the medical procedure, it can still be inserted into the patient at some point after the medical procedure (action 830) has begun (action 820). In some aspects of this technology, performing a medical procedure may include or begin by placing the patient under anesthesia.

[0054] Therefore, having described several aspects and embodiments of the technology set forth in this disclosure, it is to be understood that various changes, modifications, and improvements will readily occur to those skilled in the art. Such changes, modifications, and improvements are intended to fall within the spirit and scope of the technology described herein. For example, those skilled in the art will readily conceive of a wide variety of other means and / or structures for performing functions and / or obtaining results and / or one or more of the advantages described herein, and each of such changes and / or modifications is considered to be within the scope of the embodiments described herein. Those skilled in the art will recognize or be able to determine many equivalents of the specific embodiments described herein using only conventional experimentation. Therefore, it is to be understood that the foregoing embodiments are presented by way of example only, and inventive embodiments may be practiced in other ways than as specifically described. Furthermore, any combination of such features, systems, articles, materials, kits, and / or methods is included within the scope of this disclosure if two or more features, systems, articles, materials, kits, and / or methods described herein are not contradictory.

[0055] The above embodiments can be implemented in any of a variety of ways. One or more aspects and embodiments of this disclosure relating to the execution of a process or method can be performed or controlled by program instructions executable by a device (e.g., a computer, processor, or other device). In this regard, various inventive concepts can be embodied in a computer-readable storage medium (or multiple computer-readable storage media) (e.g., a computer memory, one or more floppy disks, optical disks, optical discs, magnetic tapes, flash memory, field-programmable gate arrays, or other semiconductor devices, circuit configurations, or other tangible computer storage media) encoding one or more programs that, when executed on one or more computers or other processors, perform one or more methods implementing the various embodiments described above. The one or more computer-readable media may be transportable, such that one or more programs stored thereon can be loaded onto one or more different computers or other processors to implement the various aspects described above. In some embodiments, the computer-readable medium may be a non-transitory medium.

[0056] The embodiments of this technology described above can be implemented in any of a number of ways. For example, the embodiments can be implemented using hardware, software, or a combination thereof. When implemented in software, the software code can execute on any suitable processor or set of processors, whether provided in a single computer or distributed across multiple computers. It should be understood that any component or set of components performing the above functions can generally be considered as a controller controlling the above functions. The controller can be implemented in many ways, such as utilizing dedicated hardware or general-purpose hardware (e.g., one or more processors) programmed to perform the above functions using microcode or software, and when the controller corresponds to multiple components of a system, the controller can be implemented in a combination of ways.

[0057] Furthermore, it should be understood that a computer can be embodied in any of a variety of forms, such as rack-mounted computers, desktop computers, laptop computers, or tablet computers, as non-limiting examples. Additionally, a computer can be embedded in a device that is not typically considered a computer but has appropriate processing capabilities, including personal digital assistants (PDAs), smartphones, or any other suitable portable or fixed electronic device.

[0058] In addition, a computer may have one or more input and output devices. These devices, among other things, can be used to present a user interface. Examples of output devices that can be used to provide a user interface include printers or displays for visual presentation of output, and speakers or other sound-generating devices for audible presentation of output. Examples of input devices that can be used for a user interface include keyboards and pointing devices such as mice, touchpads, and digitizers. As another example, a computer may receive input information through speech recognition or in other audible formats.

[0059] Such computers can be interconnected via one or more networks in any suitable form, including local area networks (LANs) or wide area networks (WANs), such as enterprise networks and intelligent networks (INs) or the Internet. Such networks can be based on any suitable technology and can operate according to any suitable protocol, and can include wireless networks, wired networks, or fiber optic networks.

[0060] Furthermore, as described, some aspects can be embodied in one or more methods. Actions performed as part of a method can be ordered in any suitable manner. Therefore, embodiments in which actions are performed in an order different from that illustrated can be constructed, which may include performing certain actions simultaneously, even if they are shown as sequential actions in the illustrative embodiments.

[0061] All definitions defined and used herein should be understood as control dictionary definitions, definitions incorporated by reference in other documents, and / or the general meaning of the defined terms.

[0062] The indefinite articles “a” and “one” used in this specification should be understood to mean “at least one” unless explicitly indicated otherwise.

[0063] The phrase “and / or” as used herein should be understood to mean “any one or two” of the elements so linked (i.e., elements that exist together in some cases and separately in others). Multiple elements listed with “and / or” should be interpreted in the same way, i.e., “one or more” of the elements so linked. In addition to the elements specifically identified by the “and / or” clause, other elements may optionally exist, whether related to or unrelated to those specifically identified. Thus, as a non-limiting example, when used in conjunction with open-ended language such as “including,” a reference to “A and / or B” may in one embodiment refer only to A (optionally including elements other than B); in another embodiment only to B (optionally including elements other than A); in yet another embodiment both A and B (optionally including other elements); and so on.

[0064] As used herein, with respect to a list of one or more elements, the phrase "at least one" should be understood to mean at least one element selected from any one or more elements in the list of elements, but does not necessarily include at least one of every element specifically listed in the list of elements, and does not exclude any combination of elements in the list of elements. This definition also allows elements to optionally exist, whether related to or unrelated to those specifically identified elements, in addition to those specifically identified in the list of elements referred to by the phrase "at least one". Thus, as a non-limiting example, "at least one of A and B" (or equivalently, "at least one of A or B", or equivalently "at least one of A and / or B") may in one embodiment refer to at least one A, optionally including more than one A, where B is absent (and optionally including elements other than B); in another embodiment refer to at least one B, optionally including more than one B, where A is absent (and optionally including elements other than A); in yet another embodiment refer to at least one A, optionally including more than one A, and at least one B, optionally including more than one B (and optionally including other elements); and so on.

[0065] Furthermore, the wording and terminology used herein are for descriptive purposes and should not be considered restrictive. The use of “including,” “contains,” or “has,” “includes,” “involves,” and variations thereof in this document means to include the entries listed thereafter and their equivalents, as well as any additional entries.

[0066] In the above instructions, all transition phrases (such as “including,” “contains,” “carries,” “has,” “includes,” “involves,” “retains,” “composes of,” etc.) should be understood as open-ended, meaning including but not limited to. Only the transition phrases “composes of” and “essentially consists of” should be closed or semi-closed transition phrases, respectively.

Claims

1. A computer-implemented method, comprising: Receive one or more patient characteristics associated with a patient scheduled for a non-cardiac medical procedure; The one or more patient characteristics are provided as input to a machine learning model that is trained to output decoupled predictions. Using at least one computer processor, the machine learning model is used to process the one or more patient characteristics to output a decoupled prediction for the patient, wherein the decoupled prediction is associated with the non-cardiac medical procedure; as well as Display on the user interface an indication of the decoupled prediction for the patient, output from the machine learning model.

2. The computer-implemented method according to claim 1, wherein, The one or more patient characteristics include one or more of the following: physical characteristics, medical history, medication information, or physiological indicators associated with the patient.

3. The computer-implemented method according to claim 1, wherein, Receiving one or more patient characteristics includes receiving the one or more patient characteristics from an electronic medical record associated with the patient.

4. The computer-implemented method according to claim 1, wherein, The machine learning model is trained using training data from multiple patients who have undergone different types of non-cardiac medical procedures.

5. The computer-implemented method according to claim 1, wherein, Indicators showing decoupling predictions for the patient include showing the predicted amount of decoupling the patient may experience during the non-cardiac medical procedure.

6. The computer-implemented method according to claim 5, wherein, The non-cardiac medical procedure has an expected procedure duration, and wherein displaying the predicted decoupling amount includes displaying the predicted decoupling amount as a percentage of the expected procedure duration.

7. The computer-implemented method according to claim 5, further comprising: When the predicted decoupling amount exceeds a threshold, it is determined that the patient will benefit from using an intracardiac blood pump associated with the non-cardiac medical procedure; as well as Recommendations associated with using the intracardiac blood pump are displayed on the user interface.

8. The computer-implemented method according to claim 7, wherein, The recommendation includes a time period during which using the intracardiac blood pump would be beneficial to the patient.

9. The computer-implemented method according to claim 8, wherein, The time period includes one or more of the following: before, during, or after the non-cardiac medical procedure.

10. A controller for a heart pump system, the controller comprising: At least one hardware processor is configured as follows: Receive one or more patient characteristics associated with a patient scheduled for a non-cardiac medical procedure; The one or more patient characteristics are provided as input to a machine learning model, which is trained to output a decoupled prediction for the patient, wherein the decoupled prediction is associated with the non-cardiac medical procedure; and The decoupled predictions for the patient, output from the machine learning model, are displayed on the user interface associated with the heart pump system.

11. A method of treating a patient, the method comprising: Receive decoupled predictions for patients scheduled for non-cardiac medical procedures, wherein, in response to providing one or more patient characteristics associated with the patient as input, the decoupled predictions are output from a trained machine learning model. The decoupling prediction is used to determine the period of time during which the patient will benefit from using the intracardiac blood pump associated with the non-cardiac medical procedure; During the defined time period, the intracardiac blood pump is inserted into the patient's heart; and Perform the aforementioned non-cardiac medical procedures to treat the patient.

12. The method according to claim 11, wherein, The one or more patient characteristics include one or more of the following: physical characteristics, medical history, medication information, or physiological indicators associated with the patient.

13. The method according to claim 11, wherein, One or more patient characteristics are provided as input from the electronic medical record associated with the patient to the trained machine learning model.

14. The method according to claim 11, wherein, The trained machine learning model is trained using training data from multiple patients who have undergone different types of non-cardiac medical procedures.

15. The method according to claim 11, wherein, The time period includes one or more of the following: before, during, or after the non-cardiac medical procedure.

16. The method of claim 11, further comprising: The amount of decoupling experienced by the patient during the non-cardiac medical procedure is tracked; as well as The trained machine learning model is retrained at least in part based on the amount of decoupling experienced by the patient and one or more patient characteristics.

17. A computer-implemented method for training a machine learning model to predict decoupling during non-cardiac medical procedures, the computer-implemented method comprising: Receive patient data from multiple patients, each of whom has undergone a non-cardiac medical procedure associated with an intracardiac blood pump; The patient data is analyzed by at least one hardware processor to detect one or more decoupling events during non-cardiac medical procedures associated with each of the plurality of patients; The machine learning model is trained based at least in part on one or more decoupling events detected during non-cardiac medical procedures and one or more patient characteristics associated with patients involved in the non-cardiac medical procedures. as well as The output is a decoupled, trained machine learning model used to predict one or more patients not associated with the patient data.

18. The computer-implemented method according to claim 17, wherein, The patient data includes one or more patient characteristics associated with the patient, and pressure information sensed by one or more pressure sensors of a mechanical circulatory support device including the intracardiac pump and / or derived from the one or more pressure sensors.

19. The computer-implemented method according to claim 18, wherein, The pressure information includes left ventricular pressure signals and aortic pressure signals.

20. The computer-implemented method according to claim 19, wherein, Analyzing the patient data to detect one or more decoupling events includes detecting a decoupling event when the peak value of the aortic pressure signal is greater than the peak value of the left ventricular pressure signal during a cardiac cycle.

21. The computer-implemented method according to claim 20, further comprising: The amount of decoupling of the patient during the non-cardiac medical procedure is determined based on one or more decoupling events detected. Training the machine learning model based at least in part on the one or more decoupling events includes training the machine learning model based at least in part on the amount of decoupling of the patient during the non-cardiac medical procedure.

22. The computer-implemented method according to claim 17, wherein, Receiving patient data from multiple patients includes receiving the patient data from corresponding electronic medical records associated with the multiple patients.

Images