Detection of severity of a fall

Abstract

An example medical device system includes a medical device including an accelerometer, one or more memories, and processing circuitry. The processing circuitry is configured to detect, based on the one or more accelerometer signals, an impact event occurring at a point in time. The processing circuitry is configured to process the one or more accelerometer signals to determine a plurality of first characteristics corresponding to a first period of time before the impact event. The processing circuitry is configured to process the one or more accelerometer signals to determine a plurality of second characteristics corresponding to a second period of time following the impact event. The processing circuitry is configured to determine a severity of the impact event based on the first characteristics and the second characteristics and to determine whether to output an alert based on the severity of the impact event.

Description

DETECTION OF SEVERITY OF A FALL

[0001] This application claims the benefit of U.S. Provisional Patent Application Serial No. 63 / 562,508, filed March 7, 2024, the entire content of which is incorporated herein by reference.TECHNICAL FIELD

[0002] The disclosure relates generally to devices, systems, and techniques, and more particularly to devices, systems, and techniques to detect a fall, such as of a person, and determine the severity of the fall.BACKGROUND

[0003] Implantable medical devices (IMDs) including implantable pacemakers, implantable cardioverter-defibrillators (ICDs), and insertable cardiac monitors without therapies (e.g., Medtronic Reveal LINQ™ or LINQ II™), record cardiac electrogram (EGM) signals for sensing cardiac events, e.g., P-waves and R-waves. IMDs detect episodes of bradycardia, tachycardia and / or fibrillation from the sensed cardiac events, and some IMDs respond to the episodes as needed with pacing therapy or high-voltage antitachyarrhythmia shocks, e.g., cardioversion or defibrillation shocks. These and other medical devices may include, or be part of a system that includes, sensors that generate other physiological -based signals, such as signals that vary based on patient movement or activity, cardiovascular pressure, blood oxygen saturation, edema, or thoracic impedance.SUMMARY

[0004] In general, this disclosure is directed to techniques for determining an occurrence of a fall and the severity of the fall, such as of a person. While the techniques are primarily described herein as being applied with respect to a person, such techniques may be used to detect the severity of falls of other objects, including inanimate objects or animate objects. The techniques described herein include the detection of an impact event (e.g., an impact with the ground) and the processing of one or more accelerometer signals to determine first characteristics of the one or more accelerometer signals before the impact event and second characteristics of the one or more accelerometer signals after the impact event. The first characteristics and the second characteristics may be used to determine a severity of theimpact event. For example, the first characteristics and second characteristics may be described as mechanical, as they are derived from one or more accelerometer signals which may be indicative of mechanical motion or activity.

[0005] As used herein a time of an impact event is intended to be a specific point in time and not a time span. It should be understood that the use of “before,” “pre,” “post,” “following,” “after,” or similar terminology herein is not intended to exclude the specific point in time. For example, characteristics corresponding to a time period before an impact event may include a characteristic at the specific point in time of the event, but not characteristics after the specific point in time of the event.

[0006] As a severity of an impact event on a patient may not necessarily be directly correlated with a level of mechanical impact, it may be desirable to take physiological parameters into account when determining a severity of an impact event. As such, the techniques may also include processing a set of physiological signals collected over time to determine a respective baseline for each of the physiological signals. The techniques may include processing the set of physiological signals to determine one or more third characteristics before the impact event and processing the set of physiological signals to determine one or more fourth characteristics after the impact event. Deviation (or lack thereof) of the one or more third characteristics or the one or more fourth characteristics, from the baseline, or from each other, may also be used to determine the severity of the impact event. Whether the one or more fourth characteristics are within corresponding acceptable clinical range(s) may also be used to determine the severity of the impact event.

[0007] The determined severity of the impact event may be used to determine whether to generate and / or send an alert regarding the fall to, for example, a caregiver, a clinician, the patient, a patient monitoring service, or the like. In some examples, the severity of impact may be further determined based on physiological parameters sensed before and after the impact event.

[0008] In one example, a system includes one or more sensors, wherein the medical device is configured to generate, using an accelerometer of the one or more sensors, one or more accelerometer signals indicating a motion of a patient; one or more memories; and processing circuitry in communication with the one or more memories, wherein the processing circuitry is configured to: detect, based on the one or more accelerometer signals, an impact event corresponding to the patient, the impact event occurring at a pointin time; process the one or more accelerometer signals to determine a plurality of first characteristics corresponding to a first period of time before the point in time corresponding to the impact event, wherein the plurality of first characteristics comprise at least one of a maximum acceleration, a pre-impact morphology, a pre-impact posture, a pre-impact rate of acceleration change, or a pre-impact duration of acceleration change; process the one or more accelerometer signals to determine a plurality of second characteristics corresponding to a second period of time following the point in time corresponding to the impact event, the plurality of second characteristics comprising at least one of a post-impact morphology, a post-impact posture, or a post impact activity level; determine a severity of the impact event based on the first characteristics and the second characteristics; and determine whether to output an alert based on the severity of the impact event.

[0008] In other examples, a method includes detecting, by processing circuitry of a medical device system and based on the one or more accelerometer signals, an impact event corresponding to the patient, the impact event occurring at a point in time; processing, by the processing circuitry, the one or more accelerometer signals to determine a plurality of first characteristics corresponding to a first period of time before the point in time corresponding to the impact event, wherein the plurality of first characteristics comprise at least one of a maximum acceleration, a pre-impact morphology, a pre-impact posture, a pre-impact rate of acceleration change, or a preimpact duration of acceleration change; processing, by the processing circuitry, the one or more accelerometer signals to determine a plurality of second characteristics corresponding to a second period of time following the point in time corresponding to the impact event, the plurality of second characteristics comprising at least one of a postimpact morphology, a post-impact posture, or a post impact activity level; determining a severity of the impact event based on the first characteristics and the second characteristics; and determining whether to output an alert based on the severity of the impact event.

[0009] In other examples, non-transitory computer-readable storage media includes instructions, that when executed by processing circuitry, cause the processing circuitry to: detect, based on one or more accelerometer signals, an impact event corresponding to a patient, the impact event occurring at a point in time; process the one or moreaccelerometer signals to determine a plurality of first characteristics corresponding to a first period of time before the point in time corresponding to the impact event, wherein the plurality of first characteristics comprise at least one of a maximum acceleration, a preimpact morphology, a pre-impact posture, a pre-impact rate of acceleration change, or a pre-impact duration of acceleration change; process the one or more accelerometer signals to determine a plurality of second characteristics corresponding to a second period of time following the point in time corresponding to the impact event, the plurality of second characteristics comprising at least one of a post-impact morphology, a post-impact posture, or a post impact activity level; determine a severity of the impact event based on the first characteristics and the second characteristics; and determine whether to output an alert based on the severity of the impact event.

[0010] This summary is intended to provide an overview of the subject matter described in this disclosure. It is not intended to provide an exclusive or exhaustive explanation of the apparatus and methods described in detail within the accompanying drawings and description below. The details of one or more aspects of the disclosure are set forth in the accompanying drawings and the description below.BRIEF DESCRIPTION OF DRAWINGS

[0011] FIG. l is a conceptual drawing illustrating an example medical device system in conjunction with a patient.

[0012] FIG. 2 is a conceptual drawing illustrating another example medical device system in conjunction with a patient.

[0013] FIG. 3A is a conceptual drawing illustrating an example configuration of the implantable medical device of FIG. 2.

[0014] FIG. 3B is a perspective drawing illustrating another example configuration of the implantable medical device of FIG. 2.

[0015] FIGS. 4A-4C are a front-view, side-view, and top-view conceptual drawings, respectively, illustrating another example medical device system in conjunction with a patient.

[0016] FIG. 5 is a conceptual drawing illustrating another example medical device system that includes an extracardiovascular implantable cardioverter defibrillator system and an intracardiac pacing device implanted within a person.

[0017] FIG. 6 is a conceptual drawing illustrating an example configuration of the intracardiac pacing device of FIG. 5.

[0018] FIG. 7 is a functional block diagram illustrating an example configuration of an implantable medical device.

[0019] FIG. 8 is a functional block diagram illustrating an example configuration of an external device.

[0020] FIG. 9 is a functional block diagram illustrating an example system that includes remote computing devices, such as a server and one or more other computing devices, that are connected to an implantable medical device and / or external device via a network.

[0021] FIGS. 10A-10C are graphical diagrams illustrating example accelerometer signals, and corresponding spectrograms, of an accelerometer of an implantable medical device surrounding various forward fall events.

[0022] FIG. 11 is a flow diagram illustrating example techniques for determining a severity of an impact event and determining whether to send an alert according to one or more aspects of this disclosure.

[0023] FIG. 12A is a flow diagram illustrating example techniques for determining a severity of an impact according to one or more aspects of this disclosure.

[0024] FIG. 12B is a flow diagram illustrating other example techniques for determining a severity of an impact according to one or more aspects of this disclosure

[0025] FIG. 13 is a graphical diagram illustrating example spike magnitudes and associated posture angle changes for a variety of fall types and activities of daily living as determined by an implantable medical device from accelerometer signals.

[0026] FIG. 14 is a flowchart illustrating an example of techniques for determining a severity of an impact event in accordance with one or more aspects of this disclosure.

[0027] FIG. 15 is a conceptual diagram illustrating an example machine learning model configured to determine a severity of an impact event, such as a fall.

[0028] FIG. 16 is an example of a machine learning model being trained using supervised and / or reinforcement learning techniques.DETAILED DESCRIPTION

[0029] Falling and physical impact is a serious risk for many patients, especially elderly patients and disabled patients. Medical conditions, such as osteoporosis, may weaken a patient’s bones such that fracture or other serious injury is likely, or more likely, when the patient falls or sustains significant impact. Some patients may be unable to return to a standing or upright position after a fall. This means that a patient might become stranded without an ability to move or seek assistance after a fall.

[0030] However, simply identifying a fall may not provide sufficient information to determine whether or not the patient may need assistance. All falls are not equal. Falling onto a hardwood floor can easily cause broken bones, while stumbling into a bed or sofa is less likely to cause harm. The techniques disclosed herein include techniques for identifying patient fall events, determining a severity of fall events, and, potentially, alerting medical professionals in response to detecting fall events may ensure that patients receive prompt assistance and medical care following falls. In addition, the techniques disclosed herein reduce false alerts by including physiological indications for severity assessment, as well as providing long-term trend data for identification of increased risk of falls.

[0031] As a patient is likely to spend the majority of their time outside of a hospital setting, it may be desirable to be able to continuously monitor patient falls during daily life and quickly determine a severity of a fall such that an alert may be issued if the fall is sufficiently severe that the patient may need assistance, even when the patient is outside of the hospital setting.

[0032] This disclosure sets forth techniques for determining whether a patient has fallen, determining a severity of a fall in response to detecting the fall, and, in some instances, generating indications, such as alerts, of the fall. Many current techniques for detecting patient falls require the patient to actively report or verify a fall after the fall occurs. This means that to generate an alert, many techniques require that the patient is conscious and within physical range of one or more devices to report a fall. When a fall renders a patient unconscious or otherwise unable to report the fall, these techniques may fail to generate an alert and leaving the patient stranded without medical care.

[0033] Example techniques of this disclosure include automatically detecting a patient fall based on data collected by one or more sensors / devices and processing the data to determine a severity of a fall without requiring the patient to actively report information.For example, a patient at risk of falling may possess one or more implantable medical devices (IMDs). An IMD may include one or more sensors configured to collect data over an extended period of time. For example, an IMD may include an accelerometer configured to generate one or more motion signals indicating motion of the patient over the extended period of time. An IMD may also be configured to generate one or more signals indicating cardiac activity, respiratory activity, and other biometric functions of the patient. The system may process signals collected by an IMD to determine that a patient has fallen and, if the patient has fallen, determine a severity of the fall.

[0034] In another example, a patient at risk of falling may possess one or more external medical devices, such as a patch, a wearable defibrillator vest, a holter monitor. Such external medical devices may be attached or affixed to a patient in a relatively stable manner to a torso of the patient. An external medical device may include one or more sensors configured to collect data over an extended period of time. For example, an external medical device may include an accelerometer configured to generate one or more motion signals indicating motion of the patient over the extended period of time. An external medical device may also be configured to generate one or more signals indicating cardiac activity, respiratory activity, and other biometric functions of the patient. The system may process signals collected by an external medical device to determine that a patient has fallen and, if the patient has fallen, determine a severity of the fall.

[0035] Other systems may attempt to detect a fall through the use of an accelerometer located within a smartwatch. However, indications of movement of a patient may be much less accurate when taken based on motion of a wrist of a patient than based on motion of an implanted device which may be more securely implanted in a torso of a patient. For example, a person having a smartwatch may remain in one position while waving their arm, such that the accelerometer may erroneously detect a fall.

[0036] The system may process one or more accelerometer signals to identify impact events and determine patient motion responses to impact events. For example, one or more accelerometer signals generated by an IMD (or an appropriate external medical device) may indicate an event where a patient physically collides with an object. When a patient is struck by a swinging door, for example, the system may identify the collision in the one or more accelerometer signals generated by the IMD (or an appropriate external medical device). The system may also detect a response of the patient to the collision based on the one ormore accelerometer signals. For example, the one or more accelerometer signals may indicate that the patient stumbled and returned to a standing position following the collision or may indicate that the patient fell, collided with another surface, such as the floor, and remained in a prone position following the collision.

[0037] In some examples, the system may process one or more accelerometer signals to determine one or more characteristics of the one or more accelerometer signals before a detected impact event. In some examples, the system may process one or more accelerometer signals to determine one or more characteristics of the one or more accelerometer signals after the detected impact event. The system may transform the one or more accelerometer signals, for example, into indications of characteristics, indications of severity of the impact event, impact severity score components, and / or the like.

[0038] In some examples, the system may analyze the rate of acceleration change and duration of acceleration change before a fall as part of determining the severity of a fall. This is because stronger impacts may result in a greater rate of acceleration change. The system may also determine, based on the accelerometer signal, a posture of a patient before a fall and a posture of a patient after a fall to determine a trajectory of the fall and. / or which body part(s) was most likely to first strike the ground during the fall. The trajectory of the fall and / or the body part(s) first to strike the ground may be important for determining a severity of the fall, because injury to some body parts might cause greater harm to the patient than injury to other body parts. For example, if a patient falls backwards, the patient may be less likely to brace the fall with their hand(s) and / or arm(s) and there may be a higher likelihood of a head impact with the ground.

[0039] The techniques of this disclosure are not limited to analyzing accelerometer data. In some examples, the system may process one or more signals generated by an IMD (or an appropriate external medical device) to determine changes in biometric signals due to patient falls. For example, changes in heart rate, heart rate variability, respiration rate, blood pressure, temperature, oxygen saturation, fluid status, pulse transit time, and / or pulse wave velocity due to a patient fall may indicate a severity of the fall. Based on detecting a patient fall and determining that the fall is severe, the system may output one or more alerts so that the patient receives prompt assistance and medical care.

[0040] IMDs including implantable pacemakers and implantable cardioverterdefibrillators (ICDs), and external, e.g., wearable devices, may record cardiac electrogram(EGM) signals (which, as used herein, may include electrocardiogram (ECM) signals) for sensing cardiac events, e.g., P-waves and R- waves. IMDs detect episodes of bradycardia, tachycardia and / or fibrillation from the sensed cardiac events, and some IMDs respond to the episodes as needed with pacing therapy or high-voltage anti -tachyarrhythmia shocks, e.g., cardioversion or defibrillation shocks. These and other devices may include, or be part of a system that includes, sensors to generate at least one accelerometer signal from which characteristics may be determined to determine a severity of an impact event.

[0041] According to the features or aspects of this disclosure, processing circuitry may detect, based on one or more accelerometer signals, an impact event corresponding to the patient, the impact event occurring at a point in time. The processing circuitry may process the one or more accelerometer signals to determine a plurality of first characteristics corresponding to a first period of time before the point in time corresponding to the impact event, the plurality first characteristics comprising at least one of a maximum acceleration, a pre-impact morphology, a pre-impact posture, a pre-impact rate of acceleration change, or a pre-impact duration of acceleration change. The processing circuitry may process the one or more accelerometer signals to determine a plurality of second characteristics corresponding to a second period of time following the point in time corresponding to the impact event, the plurality of second characteristics comprising at least one of a post-impact morphology, a post-impact posture, or a post impact activity level. The processing circuitry may determine a severity of the impact event based on the first characteristics and the second characteristics. The processing circuitry may determine whether to output an alert based on the severity of the impact event.

[0042] FIG. 1 is a conceptual drawing illustrating an example medical device system 8A in conjunction with a patient 14A. Medical device system 8A is an example of a medical device system configured to implement the techniques described herein for determining a severity of a fall. Although not depicted in FIG. 1, one or more devices of medical device system 8A may include accelerometer circuitry configured to generate one or more signals indicating motion of a patient over a period of time. In the illustrated example, medical device system 8 A includes an implantable medical device (IMD) 10A coupled to a ventricular lead 20 and an atrial lead 21. IMD 10A is an implantable cardioverterdefibrillator (ICD) capable of delivering pacing, cardioversion and defibrillation therapy to the heart 16A of a patient 14A, and will be referred to as ICD 10A hereafter.

[0043] Ventricular lead 20 and atrial lead 21 are electrically coupled to ICD 10A and extend into the heart 16A of patient 14 A. Ventricular lead 20 includes electrodes 22 and 24 shown positioned on the lead in the right ventricle (RV) of patient 14A for sensing ventricular EGM signals and pacing in the RV. Atrial lead 21 includes electrodes 26 and 28 positioned on the lead in the patient’s right atrium (RA) for sensing atrial EGM signals and pacing in the RA.

[0044] Ventricular lead 20 additionally carries a high voltage coil electrode 42, and atrial lead 21 carries a high voltage coil electrode 44, used to deliver cardioversion and defibrillation shocks. The term “anti -tachyarrhythmia shock” may be used herein to refer to both cardioversion shocks and defibrillation shocks. In other examples, ventricular lead 20 may carry both of high voltage coil electrodes 42 and 44, or may carry a high voltage coil electrode in addition to those illustrated in the example of FIG. 1.

[0045] ICD 10A may use both ventricular lead 20 and atrial lead 21 to acquire EGM signals from patient 14A and to deliver therapy in response to the acquired data. Medical device system 8A is shown as having a dual chamber ICD configuration, but other examples may include one or more additional leads, such as a coronary sinus lead extending into the right atrium, through the coronary sinus and into a cardiac vein to position electrodes along the left ventricle (LV) for sensing LV EGM signals and delivering pacing pulses to the LV. In other examples, a medical device system may be a single chamber system, or otherwise not include atrial lead 21.

[0046] Processing circuitry, sensing circuitry, and other circuitry configured for performing the techniques described herein are housed within a sealed housing 12A. Housing 12A (or a portion thereof) may be conductive so as to serve as an electrode for pacing or sensing or as an active electrode during defibrillation. As such, housing 12A is also referred to herein as “housing electrode” 12A.

[0047] ICD 10A may transmit EGM signal data and cardiac rhythm episode data acquired by ICD 10 A, as well as data regarding delivery of therapy by ICD 10 A, as well as accelerometer and / or other sensor data in manipulated and / or in raw form, possibly compressed, encoded, and / or the like, associated with patient movement, such as associated with an impact event, to external device 30 A. External device 30A may be a computing device, e.g., used in a home, ambulatory, clinic, or hospital setting, to communicate with ICD 10A via wireless telemetry. External device 30A may be coupled to a remotemonitoring system, such as Carelink®, available from Medtronic pic, of Dublin, Ireland. External device 30A may be, for example, a programmer, external monitor, or consumer device, e.g., a wearable device or a smart phone, such as the iPhone® by Apple Inc. of Cupertino, CA. In some examples, ICD 10A may include the accelerometer circuitry, one or more sensors configured to sense physiological signals indicative of physiological parameters, and / or the like.

[0048] External device 30A may be used to program commands or operating parameters into ICD 10A for controlling its functioning, e.g., when configured as a programmer for ICD 10A. External device 30A may be used to interrogate ICD 10A to retrieve data, including device operational data as well as physiological data accumulated in memory of ICD 10 A, such as data associated with an impact event like one or more accelerometer signals, physiological signals, and / or ECG signals. The interrogation may be automatic, e.g., according to a schedule, or in response to a remote or local user command. Programmers, external monitors, and consumer devices are examples of external devices 30A that may be used to interrogate ICD 10 A. Examples of communication techniques used by ICD 10A and external device 30A include radiofrequency (RF) telemetry, which may be an RF link established via Bluetooth®, wireless local area network, wireless wide area network, medical implant communication service (MICS) or other wireless connection.

[0049] In some examples, as illustrated in FIG. 1, medical device system 8 A may also include a pressure-sensing IMD 50. In the illustrated example, pressure-sensing IMD 50 is implanted in the pulmonary artery of patient 14 A. In some examples, one or more pressuresensing IMDs 50 may additionally or alternatively be implanted within a chamber of heart 16A, or generally at other locations in the circulatory system.

[0050] In one example, pressure-sensing IMD 50 is configured to sense blood pressure of patient 14A. For example, pressure-sensing IMD 50 may be arranged in the pulmonary artery and be configured to sense the pressure of blood flowing from the right ventricle outflow tract (RVOT) from the right ventricle through the pulmonary valve to the pulmonary artery. Pressure-sensing IMD 50 may therefore directly measure pulmonary artery diastolic pressure (PAD) of patient 14A. The PAD value is a pressure value that can be employed in patient monitoring. For example, PAD may be used as a basis for evaluating congestive heart failure in patient 14 A.

[0051] In other examples, however, pressure-sensing IMD 50 may be employed to measure blood pressure values other than PAD. For example, pressure-sensing IMD 50 may be arranged in the right ventricle of heart 16A to sense RV systolic or diastolic pressure, or may sense systolic or diastolic pressures at other locations of the cardiovascular system, such as within the pulmonary artery. As shown in FIG. 1, pressure-sensing IMD 50 is positioned in the main trunk of pulmonary artery 39. In other examples, a sensor, such as pressure-sensing IMD 50 may be either positioned in the right or left pulmonary artery beyond the bifurcation of the pulmonary artery.

[0052] Moreover, the placement of pressure-sensing IMD 50 is not restricted necessarily to the pulmonary side of the circulation. The pressure-sensing IMD 50 could potentially be placed in the systemic side of the circulation. For example, under certain conditions and with appropriate safety measures, pressure-sensing IMD 50 could even be placed in the left atrium, left ventricle, or aorta. Additionally, pressure-sensing IMD 50 is not restricted to placement within the cardiovascular system. For example, the pressuresensing IMD 50 might be placed in the renal circulation. Placement of pressure-sensing IMD 50 in the renal circulation may be beneficial, for example, to monitor the degree of renal insufficiency in patient 14A based on the monitoring of pressure or some other indication of renal circulation by pressure-sensing IMD 50.

[0053] In some examples, pressure-sensing IMD 50 includes a pressure sensor configured to respond to the absolute pressure inside the pulmonary artery of patient 14A. Pressure-sensing IMD 50 may be, in such examples, any of a number of different types of pressure sensors. One form of pressure sensor that may be useful for measuring blood pressure is a capacitive pressure sensor. Another example pressure sensor is an inductive sensor. In some examples, pressure-sensing IMD 50 may also comprise a piezoelectric or piezoresistive pressure transducer. In some examples, pressure-sensing IMD 50 may comprise a flow sensor.

[0054] In one example, pressure-sensing IMD 50 comprises a leadless pressure sensor including capacitive pressure sensing elements configured to measure blood pressure within the pulmonary artery. Pressure-sensing IMD 50 may be in wireless communication with ICD 10A and / or external device 30A, e.g., in order to transmit blood pressure measurements to one or both of the devices. Pressure-sensing IMD 50 may employ, e.g., radio frequency (RF) or other telemetry techniques for communicating with ICD 10A and other devices,including, e.g., external device 30A. In another example, pressure-sensing IMD 50 may include a tissue conductance communication (TCC) system by which the device employs tissue of patient 14A as an electrical communication medium over which to send and receive information to and from ICD 10A and / or external device 30A. In examples, where pressuresensing IMD 50 is not included in system 8 A, IMD 10A may include one or more optical sensors that may be configured to monitor the blood pressure of patient 14 A.

[0055] Medical device system 8A is an example of a medical device system configured for determining a severity of an impact event. Such techniques as contemplated may be performed by processing circuitry of medical device system 8A, such as processing circuitry of one or both of ICD 10A and external device 30 A, individually, or collectively. Other example medical device systems that may be configured to implement the techniques are also described herein.

[0056] FIG. 2 is a conceptual drawing illustrating another example medical device system 8B in conjunction with a patient 14B. Medical device system 8B is another example of a medical device system configured to implement the techniques described herein for determining a severity of an impact event. In the illustrated example, medical device system 8B includes an IMD 10B and an external device 30B. Although not depicted in FIG. 2, one or more devices of medical device system 8B may include accelerometer circuitry configured to generate one or more signals indicating motion of a patient over a period of time.

[0057] IMD 10B is an insertable cardiac monitor (ICM) capable of sensing and recording cardiac EGM signals from a position outside of heart 16B, and will be referred to as ICM 10B hereafter. Further, ICM 10B is capable of implementing one or more techniques for determining a severity of an impact event in accordance with the present disclosure. In some examples, ICM 10B includes or is coupled to one or more additional sensors that generate one or more other physiological signals, such as signals that vary based on patient motion and / or posture (e.g., accelerometer circuitry), heart rate, heart rate variability, respiration rate, blood pressure, temperature, oxygen saturation, fluid status, pulse transit time, and / or pulse wave velocity. ICM 10B may be implanted outside of the thorax of patient 14B, e.g., subcutaneously or submuscularly, such as the pectoral location illustrated in FIG. 2. In some examples, ICM 10B may take the form of a Medtronic Reveal LINQ™ or LINQ II™ ICM, available from Medtronic, Inc. of Minneapolis, Minnesota.

[0058] External device 30B may be configured in a manner substantially similar to that described above with respect to external device 30A and FIG. 1. External device 30B may wirelessly communicate with ICM 10B, e.g., to program the functionality of the ICM, and to retrieve recorded physiological signals, patient parameter values or scores, and / or alerts, such as a severity of an impact event and / or an alert regarding an impact event, and / or other data derived from signals from the ICM. Both ICM 10B and external device 30B include processing circuitry, and the processing circuitry of either or both devices may perform the techniques described herein for determining a severity of an impact event, as discussed in further detail below.

[0059] Although not illustrated in the example of FIG. 2, a system configured to implement the techniques of this disclosure may include one or more implanted or external medical devices in addition to or instead of ICM 10B. For example, a medical device system may include a pressure sensing IMD 50, vascular ICD (e.g., ICD 10A of FIG. 1), extravascular ICD (e.g., ICD 10C of FIGS. 4A-5), or cardiac pacemaker (e.g., IPD 10D of FIGS. 4A-6 or a cardiac pacemaker implanted outside the heart but coupled to intracardiac or epicardial leads). One or more such devices may generate accelerometer signals, physiological signals, and / or an ECG of patient 14B, and include processing circuitry configured to perform, in whole or in part, the techniques described herein for determining a severity of an impact event. The implanted devices may communicate with each other and / or an external device 30, and one of the implanted or external devices may ultimately determine the severity of the impact event.

[0060] FIG. 3 A is a conceptual drawing illustrating an ICM 10B1, which may be an example configuration of ICM 10B of FIG. 2. In the example shown in FIG. 3 A, may be implemented as a monitoring device having housing 62, proximal electrode 64 and distal electrode 66. Housing 62 may further comprise first major surface 68, second major surface 70, proximal end 72, and distal end 74. Housing 62 encloses electronic circuitry located inside the ICM 10B 1 and protects the circuitry contained therein from body fluids. Electrical feedthroughs provide electrical connection of electrodes 64 and 66.

[0061] In the example shown in FIG. 3 A, ICM 10B1 is defined by a length L, a width W and thickness or depth D and is in the form of an elongated rectangular prism wherein the length L is much larger than the width W, which in turn is larger than the depth D. In one example, the geometry of the ICM 10B 1 - in particular a width W greater than the depthD - is selected to allow ICM 10B1 to be inserted under the skin of the patient using a minimally invasive procedure and to remain in the desired orientation during insertion. For example, the device shown in FIG. 3 A includes radial asymmetries (notably, the rectangular shape) along the longitudinal axis that maintains the device in the proper orientation following insertion. For example, the spacing between proximal electrode 64 and distal electrode 66 may range from 30 millimeters (mm) to 55mm, 35mm to 55mm, and from 40mm to 55mm and may be any range or individual spacing from 25mm to 60mm. In addition, ICM 10B1 may have a length L that ranges from 30mm to about 70mm. In other examples, the length L may range from 40mm to 60mm, 45mm to 60mm and may be any length or range of lengths between about 30mm and about 70mm. In addition, the width W of major surface 68 may range from 3mm to 10mm and may be any single or range of widths between 3mm and 10mm. The thickness of depth D of ICM 10B1 may range from 2mm to 9mm. In other examples, the depth D of ICM 10B1 may range from 2mm to 5mm and may be any single or range of depths from 2mm to 9mm. In addition, ICM 10B1 according to an example of the present disclosure is has a geometry and size designed for ease of implant and patient comfort. Examples of ICM 10B1 described in this disclosure may have a volume of three cubic centimeters (cm) or less, 1.5 cubic cm or less or any volume between three and 1.5 cubic centimeters.

[0062] In the example shown in FIG. 3 A, once inserted within the patient, the first major surface 68 faces outward, toward the skin of the patient while the second major surface 70 is located opposite the first major surface 68. In addition, in the example shown in FIG. 3 A, proximal end 72 and distal end 74 are rounded to reduce discomfort and irritation to surrounding tissue once inserted under the skin of the patient. ICM 10B1, including instrument and method for inserting ICM 10B1 is described, for example, in U.S. Patent No. 11,311,312, issued on April 26, 2022, which is herein incorporated by reference in its entirety.

[0063] Proximal electrode 64 and distal electrode 66 are used to sense cardiac signals, e.g., EGM signals, intra-thoracically or extra-thoracically, which may be sub-muscularly or subcutaneously. EGM signals may be stored in a memory of ICM 10B1, and data may be transmitted via integrated antenna 82 to another medical device, which may be another implantable device or an external device, such as external device 30B. In some example, electrodes 64 and 66 may additionally or alternatively be used for sensing any bio-potentialsignal of interest, which may be, for example, an EGM, EEG, or a nerve signal, from any implanted location.

[0064] In the example shown in FIG. 3 A, proximal electrode 64 is in close proximity to the proximal end 72 and distal electrode 66 is in close proximity to distal end 74. In this example, distal electrode 66 is not limited to a flattened, outward facing surface, but may extend from first major surface 68 around rounded edges 76 and / or end surface 78 and onto the second major surface 70 so that the electrode 66 has a three-dimensional curved configuration. In some examples, electrode 66 is an uninsulated portion of a metallic, e.g., titanium, part of housing 62.

[0065] In the example shown in FIG. 3 A, proximal electrode 64 is located on first major surface 68 and is substantially flat, and outward facing. However, in other examples proximal electrode 64 may utilize the three-dimensional curved configuration of distal electrode 66, providing a three-dimensional proximal electrode (not shown in this example). Similarly, in other examples distal electrode 66 may utilize a substantially flat, outward facing electrode located on first major surface 68 similar to that shown with respect to proximal electrode 64.

[0066] The various electrode configurations allow for configurations in which proximal electrode 64 and distal electrode 66 are located on both first major surface 68 and second major surface 70. In other configurations, such as that shown in FIG. 3 A, only one of proximal electrode 64 and distal electrode 66 is located on both major surfaces 68 and 70, and in still other configurations both proximal electrode 64 and distal electrode 66 are located on one of the first major surface 68 or the second major surface 70 (e.g., proximal electrode 64 located on first major surface 68 while distal electrode 66 is located on second major surface 70). In another example, ICM 10B1 may include electrodes on both major surface 68 and 70 at or near the proximal and distal ends of the device, such that a total of four electrodes are included on ICM 10B1. Electrodes 64 and 66 may be formed of a plurality of different types of biocompatible conductive material, e.g., stainless steel, titanium, platinum, iridium, or alloys thereof, and may utilize one or more coatings such as titanium nitride or fractal titanium nitride. It should be noted that, in the example of FIG. 3 A, ICM 10B1 does not include an atrial electrode.

[0067] In the example shown in FIG. 3 A, proximal end 72 includes a header assembly 80 that includes one or more of proximal electrode 64, integrated antenna 82, anti -migrationprojections 84, and / or suture hole 86. Integrated antenna 82 is located on the same major surface (e.g., first major surface 68) as proximal electrode 64 and is also included as part of header assembly 80. Integrated antenna 82 allows ICM 1 OB 1 to transmit and / or receive data. In other examples, integrated antenna 82 may be formed on the opposite major surface as proximal electrode 64, or may be incorporated within the housing 62 of ICM 10B1. In the example shown in FIG. 3A, anti-migration projections 84 are located adjacent to integrated antenna 82 and protrude away from first major surface 68 to prevent longitudinal movement of the device. In the example shown in FIG. 3 A, anti -migration projections 84 include a plurality (e.g., nine) small bumps or protrusions extending away from first major surface 68. As discussed above, in other examples anti-migration projections 84 may be located on the opposite major surface as proximal electrode 64 and / or integrated antenna 82. In addition, in the example shown in FIG. 3A, header assembly 80 includes suture hole 86, which provides another means of securing ICM 10B1 to the patient to prevent movement following insertion. In the example shown, suture hole 86 is located adjacent to proximal electrode 64. In one example, header assembly 80 is a molded header assembly made from a polymeric or plastic material, which may be integrated or separable from the main portion of ICM 10B1.

[0068] FIG. 3B is a perspective drawing illustrating another ICM 10B2, which may be another example configuration of ICM 10B from FIG. 2. ICM 10B2 of FIG. 3B may be configured substantially similarly to ICM 10B1 of FIG. 3 A, with differences between them discussed herein.

[0069] ICM 10B2 may include a leadless, subcutaneously-implantable monitoring device, e.g. an ICM. ICM 10B2 includes housing having a base 90 and an insulative cover 36. Proximal electrode 32 and distal electrode 34 may be formed or placed on an outer surface of cover 36. Various circuitries and components of ICM 10B2, e.g., described below with respect to FIG. 3B, may be formed or placed on an inner surface of cover 36, or within base 90. In some examples, a battery or other power source of ICM 10B2 may be included within base 90. In the illustrated example, antenna 96 is formed or placed on the outer surface of cover 36, but may be formed or placed on the inner surface in some examples. In some examples, insulative cover 36 may be positioned over an open base 90 such that base 90 and cover 36 enclose the circuitries and other components and protect them from fluids such as body fluids.

[0070] Circuitries and components may be formed on the inner side of insulative cover 36, such as by using flip-chip technology. Insulative cover 36 may be flipped onto a base 90. When flipped and placed onto base 90, the components of ICM 10B2 formed on the inner side of insulative cover 36 may be positioned in a gap 38 defined by base 90. Electrodes 32 and 34 and antenna 96 may be electrically connected to circuitry formed on the inner side of insulative cover 36 through one or more vias (not shown) formed through insulative cover 36. Insulative cover 36 may be formed of sapphire (i.e., corundum), glass, parylene, and / or any other suitable insulating material. Base 90 may be formed from titanium or any other suitable material (e.g., a biocompatible material). Electrodes 32 and 34 may be formed from any of stainless steel, titanium, platinum, iridium, or alloys thereof. In addition, electrodes 32 and 34 may be coated with a material such as titanium nitride or fractal titanium nitride, although other suitable materials and coatings for such electrodes may be used.

[0071] In the example shown in FIG. 3B, the housing of ICM 10B2 defines a length Z, a width W and thickness or depth D and is in the form of an elongated rectangular prism wherein the length L is much larger than the width W, which in turn is larger than the depth D, similar to ICM 10B1 of FIG. 3A. For example, the spacing between electrodes 32 and 34 may range from 30 millimeters (mm) to 50mm, from 35mm to 45mm, or be approximately 40mm. In addition, ICM 10B2 may have a length L that ranges from 30mm to about 70mm. In other examples, the length L may range from 40mm to 60mm, 45mm to 55mm, or be approximately 45mm. In addition, the width may range from 3mm to 10mm, such as approximately 8mm. The thickness of depth D of ICM 10B2 may range from 2mm to 9mm, from 3 to 5mm, or be approximately 4mm. ICM 10B2 may have a volume of three cubic centimeters (cm) or less, or 1.5 cubic cm or less, such as approximately 1.4 cubic cm.

[0072] In the example shown in FIG. 3B, once inserted subcutaneously within the patient, outer surface of cover 36 generally faces outward, toward the skin of the patient. In addition, as shown in FIG. 3B, proximal end 94 and distal end 92 are rounded to reduce discomfort and irritation to surrounding tissue once inserted under the skin of the patient. In addition, edges of ICM 10B2 may be rounded. In some cases, ICM 10B2 may flip within a patient such that the outer surface of cover 36 may not face outward, toward the skin of the patient. The techniques of this disclosure may be implementable whether the outer surface of cover 36 faces outward, toward the skin of the patient or not. For example, accelerometersignals may be affected by the orientation of ICM 10B2. In some examples, ICM 10B2 may implement a device flip detection feature which may determine that ICM 10B2 has flipped based on one or more accelerometer signals and may process such accelerometer signals to adjust for the flipping of ICM 10B2.

[0073] FIGS. 4A-4C are front-view, side-view, and top-view conceptual drawings, respectively, illustrating another example medical device system 8C in conjunction with a patient 14C. Medical device system 8C is another example of a medical device system configured to implement the one or more techniques described herein for determining a severity of an impact event. Although not depicted in FIGS. 4A-4C, one or more devices of medical device system 8C may include accelerometer circuitry configured to generate one or more signals indicating motion of a patient over a period of time.

[0074] In the illustrated example, medical device system 8C includes an extracardiovascular ICD system 100A implanted within a patient 14C. ICD system 100A includes an IMD 10C, which is an ICD and is referred to hereafter as ICD 10C, connected to at least one implantable cardiac defibrillation lead 102A. ICD 10C is configured to deliver high-energy cardioversion or defibrillation pulses to a heart 16C of patient 14C when atrial or ventricular fibrillation is detected. Cardioversion shocks are typically delivered in synchrony with a detected R-wave when fibrillation detection criteria are met. Defibrillation shocks are typically delivered when fibrillation criteria are met, and the R-wave cannot be discerned from signals sensed by ICD 10C.

[0075] ICD 10C is implanted subcutaneously or submuscularly on the left side of patient 14C above the ribcage. Defibrillation lead 102A may be implanted at least partially in a substemal location, e.g., between the ribcage and / or sternum 110 and heart 16C. In one such configuration, a proximal portion of lead 102A extends subcutaneously from ICD 10C toward sternum 110 and a distal portion of lead 102 A extends superior under or below the sternum 110 in the anterior mediastinum 112 (FIG. 4C). The anterior mediastinum 112 is bounded laterally by the pleurae 116 (FIG. 1C), posteriorly by the pericardium 114 (FIG. 4C), and anteriorly by the sternum 110. In some instances, the anterior wall of the anterior mediastinum may also be formed by the transversus thoracis and one or more costal cartilages. The anterior mediastinum includes a quantity of loose connective tissue (such as areolar tissue), some lymph vessels, lymph glands, substemal musculature (e.g., transverse thoracic muscle), branches of the internal thoracic artery, and the internal thoracic vein. Inone example, the distal portion of lead 102A extends along the posterior side of the sternum 110 substantially within the loose connective tissue and / or substemal musculature of the anterior mediastinum. Lead 102A may be at least partially implanted in other intrathoracic locations, e.g., other non-vascular, extra-pericardial locations, including the gap, tissue, or other anatomical features around the perimeter of and adjacent to, but not attached to, the pericardium or other portion of the heart and not above the sternum 110 or ribcage.

[0076] In other examples, lead 102A may be implanted at other extracardiovascular locations. For example, defibrillation lead 102A may extend subcutaneously above the ribcage from ICD 10C toward a center of the torso of patient 14C, bend or turn near the center of the torso, and extend subcutaneously superior above the ribcage and / or sternum 110. Defibrillation lead 102A may be offset laterally to the left or the right of the sternum 110 or located over the sternum 110. Defibrillation lead 102A may extend substantially parallel to the sternum 110 or be angled lateral from the sternum 110 at either the proximal or distal end.

[0077] Defibrillation lead 102 A includes an insulative lead body having a proximal end that includes a connector 104 configured to be connected to ICD 10C and a distal portion that includes one or more electrodes. Defibrillation lead 102A also includes one or more conductors that form an electrically conductive path within the lead body and interconnect the electrical connector and respective ones of the electrodes.

[0078] Defibrillation lead 102 A includes a defibrillation electrode that includes two sections or segments 106 A and 106B, collectively (or alternatively) defibrillation electrode 106. The defibrillation electrode 106 is toward the distal portion of defibrillation lead 102A, e.g., toward the portion of defibrillation lead 102A extending along the sternum 110. Defibrillation lead 102A is placed below and / or along sternum 110 such that a therapy vector between segments 106 A or 106B and a housing electrode formed by or on ICD 10C (or other second electrode of the therapy vector) is substantially across a ventricle of heart 16C. The therapy vector may, in one example, be viewed as a line that extends from a point on defibrillation electrode 106 (e.g., a center of one of the segments 106A or 106B) to a point on the housing electrode of ICD 10C. Defibrillation electrode 106 may, in one example, be an elongated coil electrode.

[0079] Defibrillation lead 102A may also include one or more sensing electrodes, such as sensing electrodes 108A and 108B (individually or collectively, “sensing electrode(s)108”), located along the distal portion of defibrillation lead 102A. In the example illustrated in FIG. 4 A and FIG. 4B, sensing electrodes 108 A and 108B are separated from one another by defibrillation electrode 106. In other examples, however, sensing electrodes 108 A and 108B may be both distal of defibrillation electrode 106 or both proximal of defibrillation electrode 106. In other examples, lead 102A may include more or fewer electrodes at various locations proximal and / or distal to defibrillation electrode 106. In the same or different examples, ICD 10C may include one or more electrodes on another lead (not shown).

[0080] ICD system 100A may sense electrical signals via one or more sensing vectors that include combinations of electrodes 108A and 108B and the housing electrode of ICD 10C. In some instances, ICD 10C may sense cardiac electrical signals using a sensing vector that includes one of the segments 106 A and 106B and one of sensing electrodes 108 A and 108B or the housing electrode of ICD 9. The sensed electrical intrinsic signals may include electrical signals generated by cardiac muscle and indicative of depolarizations and repolarizations of heart 16C at various times during the cardiac cycle. ICD 10C analyzes the electrical signals sensed by the one or more sensing vectors to detect tachyarrhythmia, such as ventricular tachycardia or ventricular fibrillation. In response to detecting the tachyarrhythmia, ICD 10C may begin to charge a storage element, such as a bank of one or more capacitors, and, when charged, deliver one or more defibrillation pulses via defibrillation electrode 106 of defibrillation lead 102A if the tachyarrhythmia is still present.

[0081] Medical device system 8C also includes an IMD 10D, which is implanted within heart 16C and configured to deliver cardiac pacing to the heart, e.g., is an intracardiac pacing device (IPD). IMD 10D is referred to as IPD 10D hereafter. In the illustrated example, IPD 10D is implanted within the right ventricle of heart 16C. However, in other examples, system 8C may additionally or alternatively include one or more IPDs 10D within other chambers of heart 16C, or similarly configured pacing devices attached to an external surface of heart 16C (e.g., in contact with the epicardium) such that the pacing device is disposed outside of heart 16C.

[0082] IPD 10D is configured to sense electrical activity of heart 16C and deliver pacing therapy, e.g., bradycardia pacing therapy, cardiac resynchronization therapy (CRT), antitachycardia pacing (ATP) therapy, and / or post-shock pacing, to heart 16C. IPD 10D may be attached to an interior wall of heart 16C via one or more fixation elements that penetrate thetissue. These fixation elements may secure IPD 10D to the cardiac tissue and retain an electrode (e.g., a cathode or an anode) in contact with the cardiac tissue.

[0083] IPD 10D may be capable sensing electrical signals using the electrodes carried on the housing of IPD 10D. These electrical signals may be electrical signals generated by cardiac muscle and indicative of depolarizations and repolarizations of heart 16C at various times during the cardiac cycle. IPD 10D may analyze the sensed electrical signals to detect bradycardia and tachyarrhythmias, such as ventricular tachycardia or ventricular fibrillation. In response to detecting bradycardia, IPD 10D may deliver bradycardia pacing via the electrodes of IPD 10D. In response to detecting tachyarrhythmia, IPD 10D may, e.g., depending on the type of tachyarrhythmia, deliver ATP therapy via the electrodes of IPD 10D. In some examples, IPD 10D may deliver post-shock pacing in response to determining that another medical device, e.g., ICD 10C, delivered an anti -tachyarrhythmia shock.

[0084] IPD 10D and ICD 10C may be configured to coordinate their arrhythmia detection and treatment activities. In some examples IPD 10D and ICD 10C may be configured to operate completely independently of one another. In such a case, IPD 10D and ICD 10C are not capable of establishing telemetry communication sessions with one another to exchange information about sensing and / or therapy using one-way or two-way communication. Instead, each of IPD 10D and ICD 10C analyze the data sensed via their respective electrodes to make tachyarrhythmia detection and / or therapy decisions. As such, each device does not know if the other will detect the tachyarrhythmia, if or when it will provide therapy, and the like. In some examples, IPD 10D may be configured to detect antitachyarrhythmia shocks delivered by ICD system 100 A, which may improve the coordination of therapy between subcutaneous ICD 10C and IPD 10D without requiring device-to-device communication. In this manner, IPD 10D may coordinate the delivery of cardiac stimulation therapy, including the termination of ATP and the initiation of the delivery of post-shock pacing, with the application of an anti-tachyarrhythmia shock merely through the detection of defibrillation pulses and without the need to communicate with the defibrillation device applying the anti -tachyarrhythmia shock.

[0085] In other examples, IPD 10D and ICD 10C may engage in communication to facilitate the appropriate detection of arrhythmias and / or delivery of therapy. The communication may include one-way communication in which one device is configured to transmit communication messages and the other device is configured to receive thosemessages. The communication may instead include two-way communication in which each device is configured to transmit and receive communication messages.

[0086] External device 30C may be configured substantially similarly to external device 30A described above with respect to FIG. 1. External device 30C may be configured to communicate with one or both of ICD 10C and IPD 10D. In examples where external device 30C only communicates with one of ICD 10C and IPD 10D, the non-communicative device may receive instructions from or transmit data to the device in communication with external device 30C. In some examples, a user may interact with device 30C remotely via a networked computing device. The user may interact with external device 30C to communicate with IPD 10D and / or ICD 10C.

[0087] For example, the user may interact with external device 30C to send an interrogation request and retrieve sensed accelerometer data, physiological data, ECG data, and / or therapy delivery data stored by one or both of ICD 10C and IPD 10D, and program or update therapy parameters that define therapy, or perform any other activities with respect to ICD 10C and IPD 10D. Although the user is a physician, technician, surgeon, electrophysiologist, or other healthcare professional, the user may be patient 14C in some examples. For example, external device 30C may allow a user to program any coefficients, weighting factors, or techniques for determining a severity of an impact event. As another example, external device 30C may be used to program commands or operating parameters into ICD 10C for controlling its functioning. External device 30C may be used to interrogate ICD 10C to retrieve data, including device operational data as well as accelerometer data, physiological data, and / or ECG data accumulated in IMD memory, such as data associated with an impact event. ICD 10C may be configured to implement the various features or aspects of the present disclosure for determining a severity of an impact event.

[0088] Medical device system 8C is an example of a medical device system configured to determine a severity of an impact event. Such techniques as contemplated may be performed by processing circuitry of medical device system 8C, such as processing circuitry of one or both of ICD 10C and external device 30C, individually, or collectively, as discussed in further detail below following a description provided in connection with FIG. 10. Other example medical device systems that may be configured to implement the techniques are described herein.

[0089] FIG. 5 is a conceptual drawing illustrating another example medical device system 8D that includes an extracardiovascular ICD system 100B and IPD 10D implanted within a person. Medical device system 8D may be configured to perform any of the techniques described herein with respect to medical device system 8C of FIGS. 4A-4C. Components with like numbers in FIGS. 4A-4C and FIG. 5 may be similarly configured and provide similar functionality. Although not depicted in FIG. 5, one or more devices of medical device system 8C may include accelerometer circuitry configured to generate one or more signals indicating motion of a patient over a period of time.

[0090] In the example of FIG. 5, extracardiovascular ICD system 100B includes ICD 10C coupled to a defibrillation lead 102B. Unlike defibrillation lead 102A of FIGS. 4A - 4C, defibrillation lead 102B extends subcutaneously above the ribcage from ICD 10C. In the illustrated example, defibrillation lead 102B extends toward a center of the torso of patient 14D, bends or turns near the center of the torso, and extends subcutaneously superior above the ribcage and / or sternum 110. Defibrillation lead 102B may be offset laterally to the left or the right of sternum 110 or located over sternum 110. Defibrillation lead 102B may extend substantially parallel to sternum 102 or be angled lateral from the sternum at either the proximal or distal end.

[0091] Defibrillation lead 102B includes an insulative lead body having a proximal end that includes a connector 104 configured to be connected to ICD 10C and a distal portion that includes one or more electrodes. Defibrillation lead 102B also includes one or more conductors that form an electrically conductive path within the lead body and interconnect the electrical connector and respective ones of the electrodes. In the illustrated example, defibrillation lead 102B includes a single defibrillation electrode 106 toward the distal portion of defibrillation lead 102B, e.g., toward the portion of defibrillation lead 102B extending along sternum 110. Defibrillation lead 102B is placed along sternum 110 such that a therapy vector between defibrillation electrode 106 and a housing electrode formed by or on ICD 10C (or other second electrode of the therapy vector) is substantially across a ventricle of heart 16D.

[0092] Defibrillation lead 102B may also include one or more sensing electrodes, such as sensing electrodes 108 A and 108B, located along the distal portion of defibrillation lead 102B. In the example illustrated in FIG. 5, sensing electrodes 108A and 108B are separated from one another by defibrillation electrode 106. In other examples, however, sensingelectrodes 108 A and 108B may be both distal of defibrillation electrode 106 or both proximal of defibrillation electrode 106. In other examples, lead 102B may include more or fewer electrodes at various locations proximal and / or distal to defibrillation electrode 106, and lead 102B may include multiple defibrillation electrodes, e.g., segments 106A and 106B as illustrated in the example of FIGS. 4A-4C.

[0093] Medical device system 8D is an example of a medical device system configured to determine a severity of an impact event. Such techniques as contemplated may be performed by processing circuitry of medical device system 8D, such as processing circuitry of one or both of ICD 10C and external device 30D, individually, or collectively, as discussed in further detail below.

[0094] FIG. 6 is a conceptual drawing illustrating an example configuration of IPD 10D. As shown in FIG. 6, IPD 10D includes case 130, cap 138, electrode 140, electrode 132, fixation mechanisms 142, flange 134, and opening 136. Together, case 130 and cap 138 may be considered the housing of IPD 10D. In this manner, case 130 and cap 138 may enclose and protect the various electrical components, e.g., circuitry, within IPD 10D. Case 130 may enclose substantially all of the electrical components, and cap 138 may seal case 130 and create the hermetically sealed housing of IPD 10D. Although IPD 10D is generally described as including one or more electrodes, IPD 10D may typically include at least two electrodes (e.g., electrodes 132 and 140) to deliver an electrical signal (e.g., therapy such as cardiac pacing) and / or provide at least one sensing vector.

[0095] Electrodes 132 and 140 are carried on the housing created by case 130 and cap 138. In this manner, electrodes 132 and 140 may be considered leadless electrodes. In the example of FIG. 6, electrode 140 is disposed on the exterior surface of cap 138. Electrode 140 may be a circular electrode positioned to contact cardiac tissue upon implantation. Electrode 132 may be a ring or cylindrical electrode disposed on the exterior surface of case 130. Both case 130 and cap 138 may be electrically insulating.

[0096] Electrode 140 may be used as a cathode and electrode 132 may be used as an anode, or vice versa, for delivering cardiac pacing such as bradycardia pacing, CRT, ATP, or post-shock pacing. However, electrodes 132 and 140 may be used in any stimulation configuration. In addition, electrodes 132 and 140 may be used to detect intrinsic electrical signals from cardiac muscle.

[0097] Fixation mechanisms 142 may attach IPD 10D to cardiac tissue. Fixation mechanisms 142 may be active fixation tines, screws, clamps, adhesive members, or any other mechanisms for attaching a device to tissue. As shown in the example of FIG. 6, fixation mechanisms 142 may be constructed of a memory material, such as a shape memory alloy (e.g., nickel titanium), that retains a preformed shape. During implantation, fixation mechanisms 142 may be flexed forward to pierce tissue and allowed to flex back towards case 130. In this manner, fixation mechanisms 142 may be embedded within the target tissue.

[0098] Flange 144 may be provided on one end of case 130 to enable tethering or extraction of IPD 10D. For example, a suture or other device may be inserted around flange 144 and / or through opening 146 and attached to tissue. In this manner, flange 144 may provide a secondary attachment structure to tether or retain IPD 10D within heart 16C (or 16D) if fixation mechanisms 142 fail. Flange 144 and / or opening 146 may also be used to extract IPD 10D once the IPD needs to be explanted (or removed) from patient 14D if such action is deemed necessary.

[0099] Referring back to FIGS. 4A-5, medical device systems 8C and 8D are examples of medical device systems configured to determine a severity of an impact event. Such techniques may be performed by processing circuitry of medical device system 8C or 8D, such as processing circuitry of one or more of ICD 10C, IPD 10D, and external device 30C or 30D, individually, or collectively. Although the example medical devices systems 8C and 8D of FIGS. 4A-5 are illustrated as including both ICD 10C and IPD 10D, other examples may include only one of ICD 10C or IPD 10D, alone, or in combination with other implanted or external devices.

[0100] FIG. 7 is a functional block diagram illustrating an example configuration of an IMD 10. IMD 10 may correspond to any of ICD 10A, ICM 10B, ICD 10C, IPD 10D, or another IMD configured to implement techniques for determining a severity of an impact event as described in this disclosure. While described as IMD 10, in some examples, FIG. 7 may represent an external medical device, such as a patch, a wearable defibrillator vest, a holter monitor, or another external medical device, such as an external medical device configured to be attachable or afflxable to a torso of a patient. IMD 10 may implement the techniques for any of patients 14A-14D (hereinafter patient 14) having hearts 16A-16D (hereinafter heart 16). In the illustrated example, IMD 10 includes processing circuitry 160and an associated memory 170, sensing circuitry 162, therapy delivery circuitry 164, sensors 166, and communication circuitry 168. However, ICD 10A, ICM 10B, ICD 10C, and IPD 10D need not include all of these components, or may include additional components. For example, ICM 10B may not include therapy delivery circuitry 164, in some examples (illustrated by intermittent line).

[0101] Memory 170 includes computer-readable instructions that, when executed by processing circuitry 160, cause IMD 10 and processing circuitry 160 to perform various functions attributed to IMD 10 and processing circuitry 160 herein (e.g., determine a severity of an impact event). Memory 170 may include any volatile, non-volatile, magnetic, optical, or electrical media, such as a random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), electrically-erasable programmable ROM (EEPROM), flash memory, or any other digital or analog media. Memory 170 may store machine learning model 174, threshold(s) 176, severity of impact event 178, and characteristics 179. Characteristics 179 may include accelerometer signals, other sensor signals, determined characteristics based on such signals, acceptable clinical range(s) for various physiological parameter(s), and / or the like. An acceptable clinical range may be a range commonly accepted by the medical community as being normal for a human being.

[0102] Processing circuitry 160 may include fixed function circuitry and / or programmable processing circuitry. Processing circuitry 160 may include any one or more of a microprocessor, a controller, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or equivalent discrete or analog logic circuitry. In some examples, processing circuitry 160 may include multiple components, such as any combination of one or more microprocessors, one or more controllers, one or more DSPs, one or more ASICs, or one or more FPGAs, as well as other discrete or integrated logic circuitry. The functions attributed to processing circuitry 160 herein may be embodied as software, firmware, hardware or any combination thereof. Processing circuitry 160 may be communicatively coupled elements of IMD 10 such as memory 170, sensor(s) 166, communication circuitry 168, etc.

[0103] Sensing circuitry 162 and therapy delivery circuitry 164 may be coupled to electrodes 190. Electrodes 190 illustrated in FIG. 7 may correspond to, for example: electrodes 12, 22, 24, 26, 28, 44, and 44 of ICD 10A (FIG. 1); electrodes 64 and 66 of ICM 10B1 (FIG. 3 A); electrodes 32 and 34 of ICM 10B2 (FIG. 3B); electrodes 106, 108, and oneor more housing electrodes of ICD IOC (FIGS. 4A-5); or electrodes 132 and 140 of IPD 10D (FIG. 6).

[0104] Sensing circuitry 162 monitors signals from a selected two or more of electrodes 190 in order to monitor electrical activity of heart 16, impedance, or other physical phenomenon. Sensing of a cardiac electrical signal may be done to determine heart rates or heart rate variability, or to detect arrhythmias (e.g., tachyarrhythmias or bradycardia) or other electrical signals. In some examples, sensing circuitry 162 may include one or more filters and amplifiers for filtering and amplifying a signal received from electrodes 190.

[0105] The resulting cardiac electrical signal may be passed to cardiac event detection circuitry that detects a cardiac event when the cardiac electrical signal crosses a sensing threshold. The cardiac event detection circuitry may include a rectifier, filter and / or amplifier, a sense amplifier, comparator, and / or analog-to-digital converter. Sensing circuitry 162 outputs an indication to processing circuitry 160 in response to sensing of a cardiac event (e.g., detected P-waves or R-waves).

[0106] In this manner, processing circuitry 160 may receive detected cardiac event signals corresponding to the occurrence of detected R-waves and P-waves in the respective chambers of heart 16. Indications of detected R-waves and P-waves may be used for detecting ventricular and / or atrial tachyarrhythmia episodes, e.g., ventricular or atrial fibrillation episodes. Some detection channels may be configured to detect cardiac events, such as P- or R-waves, and provide indications of the occurrences of such events to processing circuitry 160.

[0107] Sensing circuitry 162 may also include a switch module to select which of the available electrodes 190 (or electrode polarities) are used to sense the heart activity. In examples with several electrodes 190, processing circuitry 160 may select the electrodes that function as sense electrodes, i.e., select the sensing configuration, via the switch module within sensing circuitry 162. Sensing circuitry 162 may also pass one or more digitized EGM signals (such as an ECG signal) to processing circuitry 160 for analysis, e.g., for use in cardiac rhythm discrimination.

[0108] Processing circuitry 160 may implement programmable counters. If IMD 10 is configured to generate and deliver pacing pulses to heart 16, such counters may control the basic time intervals associated with bradycardia pacing (e.g., DDD, WI, DVI, VDD, AAI, DDI, DDDR, VVIR, DVIR, VDDR, AAIR, DDIR pacing) and other modes of pacing.Intervals defined by processing circuitry 160 may include atrial and ventricular pacing escape intervals, refractory periods during which sensed P-waves and R-waves are ineffective to restart timing of the escape intervals, and the pulse widths of the pacing pulses. The durations of these intervals may be determined by processing circuitry 160 in response to pacing mode parameters stored in memory 170.

[0109] Interval counters implemented by processing circuitry 160 may be reset upon sensing of R-waves and P-waves with detection channels of sensing circuitry 162, or upon the generation of pacing pulses by therapy delivery circuitry 164, and thereby control the basic timing of cardiac pacing functions, including bradycardia pacing, CRT, ATP, or postshock pacing. The value of the count present in the interval counters when reset by sensed R-waves and P-waves may be used by processing circuitry 160 to measure the durations of R-R intervals, P-P intervals, P-R intervals and R-P intervals, which are measurements that may be stored in memory 170. Processing circuitry 160 may use the count in the interval counters to detect a tachyarrhythmia event, such as atrial fibrillation (AF), atrial tachycardia (AT), VF, or VT. These intervals may also be used to detect the overall heart rate, ventricular contraction rate, and heart rate variability. A portion of memory 170 may be configured as a plurality of recirculating buffers, capable of holding series of measured intervals, which may be analyzed by processing circuitry 160 in response to the occurrence of a pace or sense interrupt to determine whether the heart 16 of patient 14 is presently exhibiting atrial or ventricular tachyarrhythmia.

[0110] In some examples, processing circuitry 160 may determine that tachyarrhythmia has occurred by identification of shortened R-R (or P-P) interval lengths. Generally, processing circuitry 160 detects tachycardia when the interval length falls below 220 milliseconds and fibrillation when the interval length falls below 180 milliseconds. In other examples, processing circuitry 160 may detect ventricular tachycardia when the interval length falls between 330 milliseconds and ventricular fibrillation when the interval length falls below 240 milliseconds. These interval lengths are merely examples, and a user may define the interval lengths as desired, which may then be stored within memory 170. This interval length may need to be detected for a certain number of consecutive cycles, for a certain percentage of cycles within a running window, or a running average for a certain number of cardiac cycles, as examples. In other examples, additional patient parameters may be used to detect an arrhythmia. For example, processing circuitry 160 may analyzeone or more morphology measurements, impedances, or any other physiological measurements to determine that patient 14 is experiencing a tachyarrhythmia.[oni] In addition to detecting and identifying specific types of cardiac events, e.g., cardiac depolarizations, sensing circuitry 162 may also sample the detected intrinsic signals to generate an electrogram or other time-based indication of cardiac events. Sensing circuitry 162 may include an analog-to-digital converter or other circuitry configured to sample and digitize the electrical signal sensed via electrodes 190. Processing circuitry 160 may analyze the digitized signal for a variety of purposes, including morphological identification or confirmation of tachyarrhythmia of heart 16. As another example, processing circuitry 160 may analyze the digitized cardiac electrogram signal to identify and measure a variety of morphological features of the signal.

[0112] In some examples, sensing circuitry 162 is configured to sense other physiological signals of patient 14. For example, sensing circuitry 162 may be configured to sense signals that vary with changing thoracic impedance of patient 14. The thoracic impedance may vary based on fluid volume or edema in patient 14.

[0113] Sensing circuitry 162 may use any two or more of electrodes 190 to sense thoracic impedance. As the tissues within the thoracic cavity of patient 14 change in fluid content, the impedance between two electrodes may also change. For example, the impedance between a defibrillation coil electrode (42, 44, 106) and the housing electrode may be used to monitor changing thoracic impedance.

[0114] In some examples, processing circuitry 160 measured thoracic impedance values to determine a fluid index. As more fluid is retained within patient 14, e.g., edema increases, and the thoracic impedance decreases or remains relatively high, the fluid index increases. Conversely, as the thoracic impedance increases or remains relatively low, the fluid index decreases.

[0115] The thoracic impedance may also vary with patient respiration. In some examples, processing circuitry 160 may determine values of one or more respiration-related patient parameters based on thoracic impedance sensed by sensing circuitry 162. Respiration-related patient parameters may include, as examples, respiration rate, respiration depth, or the occurrence or magnitude of dyspnea or apneas.

[0116] The magnitude of the cardiac electrogram may also vary based on patient respiration, e.g., generally at a lower frequency than the cardiac cycle. In some examples,processing circuitry 160 and / or sensing circuitry 162 may filter the cardiac electrogram to emphasize the respiration component of the signal. Processing circuitry 160 may analyze the filtered cardiac electrogram signal to determine values of respiration-related patient parameters.

[0117] In the example of FIG. 7, IMD 10 includes sensors 166 coupled to sensing circuitry 162. Although illustrated in FIG. 7 as included within IMD 10, sensors 166 may be external to IMD 10, e.g., coupled to IMD 10 via one or more leads, or configured to wirelessly communicate with IMD 10. In some examples, sensors 166 transduce a signal indicative of a patient parameter, which may be amplified, filtered, or otherwise processed by sensing circuitry 162. In such examples, processing circuitry 160 determines values of patient parameters based on the physiological signals. In some examples, sensors 166 determine the patient parameter values, and communicate them, e.g., via a wired or wireless connection, to processing circuitry 160.

[0118] In some examples, sensors 166 include accelerometer circuitry 167, e.g., one or more 3-axis accelerometers. Signals generated by the accelerometer circuitry 167, such as one or more of a sagittal axis signal, a vertical axis signal and a transverse axis signal, may be indicative of, as examples, gross body movement (e.g., activity) of patient 14, such as patient 14 walking, falling, impacting a surface, patient posture, heart sounds or other vibrations or movement associated with the beating of the heart, respiration, coughing, rales, or other respiration abnormalities. In some examples, accelerometer 167 has a working range of at least + / - 3G (g-forces). In some examples, accelerometer 167 has a working range of at least + / - 10G. In some examples, accelerometer 167 is a digital accelerometer and has a 12-bit or better resolution.

[0119] In some examples, sensors 166 include one or more microphones (e.g., microphone 165) configured to detect heart sounds, respiration abnormalities, and / or other sensors configured to detect patient activity or posture, such as gyroscopes and / or strain gauges. In some examples, sensors 166 may include sensors configured to transduce physiological signals indicative of heart rate, heart rate variability, R-wave amplitude, respiration rate, respiration effort, blood pressure, temperature, oxygen saturation, fluid status, arterial blood pulsation amplitude, pulse transit time, pulse wave velocity, and / or the like. Processing circuitry 160 may determine patient parameters values based on these physiological signals.

[0120] In some examples, sensors 166 include one or more pressure sensors that transduce one or more signals indicative of blood pressure, and processing circuitry 160 determines one or more patient parameter values based on the pressure signals. Patient parameter values determined based on pressure may include, as examples, systolic or diastolic pressure values, such as pulmonary artery diastolic pressure values. In some examples, a separate pressure-sensing IMD 50 includes one or more sensors and sensing circuitry configured to generate a pressure signal, and processing circuitry 160 determines patient parameter values related to blood pressure based on information received from IMD 50.

[0121] Therapy delivery circuitry 164 is configured to generate and deliver electrical therapy to the heart. Therapy delivery circuitry 164 may include one or more pulse generators, capacitors, and / or other components capable of generating and / or storing energy to deliver as pacing therapy, defibrillation therapy, cardioversion therapy, other therapy or a combination of therapies. In some instances, therapy delivery circuitry 164 may include a first set of components configured to provide pacing therapy and a second set of components configured to provide anti-tachyarrhythmia shock therapy. In other instances, therapy delivery circuitry 164 may utilize the same set of components to provide both pacing and anti -tachyarrhythmia shock therapy. In still other instances, therapy delivery circuitry 164 may share some of the pacing and shock therapy components while using other components solely for pacing or shock delivery.

[0122] Therapy delivery circuitry 164 may include charging circuitry, one or more charge storage devices, such as one or more capacitors, and switching circuitry that controls when the capacitor(s) are discharged to electrodes 190 and the widths of pulses. Charging of capacitors to a programmed pulse amplitude and discharging of the capacitors for a programmed pulse width may be performed by therapy delivery circuitry 164 according to control signals received from processing circuitry 160, which are provided by processing circuitry 160 according to parameters stored in memory 170. Processing circuitry 160 controls therapy delivery circuitry 164 to deliver the generated therapy to the heart via one or more combinations of electrodes 190, e.g., according to parameters stored in memory 170. Therapy delivery circuitry 164 may include switch circuitry to select which of the available electrodes 190 are used to deliver the therapy, e.g., as controlled by processing circuitry 160.

[0123] In some examples, processing circuitry 160 may use intrinsic responses of accelerometer 167 to various mechanical inputs (e.g., an impact) and / or an accelerometerbased estimate of impact energy to detect a fall and classify the severity of the impact event. In some examples, to properly capture all the energy involved in the mechanical impacts, a sampling rate of accelerometer 167 may be higher than 256Hz, such as 1000Hz or higher.

[0124] In some examples, processing circuitry 160 may determine a pre-impact posture of patient 14 and a post-impact posture of patient 14 to determine a trajectory of the patient during the fall and / or to estimate which body part of patient 14 may have been the first body part to touch the surface (e.g. the ground) which is impacted during the fall. In estimating which body part of patient 14 may have been the first body part to touch the surface, processing circuitry may determine which body part of all body parts (or a subset thereof) is most likely to have been the first body part to touch the surface. For example, processing circuitry 160 may determine a trajectory of patient 14 and estimate a posture at the time of impact based on the pre-impact posture and the post-impact posture. Processing circuitry 160 may use the estimated posture at the time of impact to estimate which body part was the first body part to touch the surface. That body part may be a body part involved in the impact. For example, processing circuitry 160 may analyze accelerometer signals for a few seconds prior to impact to determine or estimate how quickly the body of patient 14 falls right before impact. For example, a slowing of the body right before impact may be indicative of patient 14 bracing themselves, such as with a hand or arm prior to impact. In some examples, processing circuitry 160 may execute machine learning model 174 to estimate which body part first touched the surface.

[0125] In some examples, processing circuitry 160 may first assess the impact of the fall using other accelerometer information, such as a maximum acceleration, a pre-impact morphology, a pre-impact posture, a pre-impact rate of acceleration change, a pre-impact duration of acceleration change, duration of accelerometer signal oscillation or ringing after impact, a post-impact morphology, a post-impact posture, or a post impact activity level, and then and use the trajectory and / or estimate of the first body part to touch the surface to refine the assessed severity of impact.

[0126] In some examples, processing circuitry 160 may sense and / or determine changes in physiological parameters, such as heart rate, heart rate variability, R-wave amplitude, respiration rate, respiration effort, blood pressure, temperature, oxygen saturation, fluidstatus, arterial blood pulsation amplitude, pulse transit time, and / or pulse wave velocity before and after an impact event as part of determining the severity of the impact event and / or to enhance the reliability of the fall detection and the reliability of the severity classification.

[0127] In some examples, processing circuitry 160 may sense and / or determine changes in physiological parameters between baseline physiological parameters and physiological parameters after an impact event and / or between clinically acceptable ranges and physiological parameters after the impact event as part of determining the severity of the impact event and / or to enhance the reliability of the fall detection and the reliability of the severity classification.

[0128] In some examples, processing circuitry 160 may determine an occurrence of a fall using a threshold that is dependent on a combination of the maximum amplitude of an accelerometer signal (e.g., “spike” amplitude) and a posture derived angle concurrently. In some examples, fall occurrence thresholds for either spike amplitude or posture derived angle may be linearly or polynomially related such that their values change dependent to one another's absolute value.

[0129] In some examples, processing circuitry 160 may analyze pre-impact activity and post-impact activity based on the one or more accelerometer signals from accelerometer 167 and if the pre-impact activities are similar to the post-impact activities (e.g., within a threshold difference of each other), processing circuitry 160 may determine that no fall occurred. For example, a person jogging may have stumbled, but regained their balance and continued j ogging.

[0130] For example, certain spans of spike amplitude range may have static posture derived angle thresholds, where other spans of spike amplitude range may have dynamic posture derived angle thresholds that are linearly dependent on spike amplitude, and vice versa. In some examples, higher order polynomial relationships between metrics may be used to realize complex curves for fall occurrence threshold boundaries in the spike-posture angle plane. The use of such thresholds may improve sensitivity and reduce the false positive rate of a fall detection algorithm. An example fall detection threshold is described hereinafter with respect to FIG. 13.

[0131] During an impact of patient 14 with a surface (e.g., the ground, another person, a piece of furniture, etc.), accelerometer signals of accelerometer 167 may include arelatively large change or displacement. This may be referred to as a spike. The spike amplitude (or the maximum positive or maximum negative displacement from average) may be compared to a threshold to determine whether the spike met a threshold magnitude and / or whether the follow-on spikes remained above a threshold magnitude (e.g., remain high) for a certain period of time. For example, processing circuitry 160 may compare a height of an accelerometer signal and / or a width of a portion of an accelerometer signal above a certain threshold to thresholds. Alternatively, the duration of the accelerometer signal oscillation or ringing as a result of the impact can be used to assess the severity of the impact event, with a longer oscillation or ringing associated with a more severe impact.

[0132] In some examples, one or more of threshold(s) 176 may be patient-specific thresholds, established based on long-term data collected for that patient. For example, IMD 10 may collect data, such as heart rate, blood pressure, R-wave amplitude, pulse wave velocity of patient 14 over a period of time, such as a recent month or week, and determine one or more of threshold(s) 176 based on the collected data. For example, corresponding post-impact parameters outside a predefined range (e.g. with a less than 0.5% probability based on the normal distribution of the parameters), may result in IMD 10 assigning a higher severity score to the fall.

[0133] In some examples, one or more of threshold(s) 176 may be defined based on a reasonable clinical range. For example, when post-impact blood pressure dropped too low (e.g., outside of the reasonable clinical range), IMD 10 may assign a higher severity and potentially issue an alert.

[0134] In some examples, processing circuitry 160 may determine that a fall has occurred and / or the severity of the fall by executing machine learning model 174. For example, machine learning model 174 may be trained to determine a fall has occurred based on one or more accelerometer signals, and / or a severity of the fall based on one or more accelerometer signals, one or more physiological signals, and / or an EGM signal (e.g., an ECG signal). It should be understood that an artificial intelligence model may be used as, or in place of, machine learning model 174 and still fall within the scope of this disclosure.

[0135] In some examples, machine learning model 174 may be run in parallel with a non-machine learning algorithm which may be configured to detect a fall and / or to determine a severity of a fall. The output of the non-machine learning algorithm may be used as training data for machine learning model 174. In some examples, machine learningmodel 174 may be trained based on electronic health record data. In some examples, processing circuitry 160 may collect sensor data for a period of time, such as for a week or longer and use such collected data to establish patient-specific statistics for one of more of threshold(s) 176, to train machine learning model 174, and / or as input to a formula for determining impact severity and / or whether to send an alert.

[0136] In some examples, processing circuitry 160 may determine a likely reason for a fall. For example, processing circuitry 160 may analyze at least one of the one or more accelerometer signals, the one or more physiological parameters, an ECG, an identification of a medication of patient 14 (e.g., which may be stored in memory 170), information from electronic healthcare records, and / or the like to estimate a reason for the fall. In estimating the reason for the fall, processing circuitry 160 may determine a likely, more likely, or most likely reason, among a plurality of reasons, for the fall. In some examples, processing circuitry 160 may execute machine learning model 174 to estimate a reason for a fall. For example, based on an ECG signal and / or one or more physiological signals, processing circuitry may be able to discriminate between a fall due to cardiac reasons or non-cardiac reasons. For example, based on one or more accelerometer signals, processing circuitry 160 may be able to determine a fall was likely due to a collision with another object. For example, there may be a pre-fall impact evident in the accelerometer signals. In some examples, processing circuitry 160 may use additional information, such as information from electronic health records and / or information regarding medications that patient 14 is taking to determine a likely reason for a fall. For example, if a patient 14 is taking hypertension medication, the fall may be more likely due to blood pressure reasons. Processing circuitry 160 may also check pre-impact blood pressure and / or post-impact blood pressure when estimating that a fall may be due to blood pressure reasons.

[0137] Communication circuitry 168 includes any suitable hardware, firmware, software or any combination thereof for communicating with another device, such as an external device 30 or another IMD or sensor. Under the control of processing circuitry 160, communication circuitry 168 may receive downlink telemetry from and send uplink telemetry to external device 30 or another device with the aid of an antenna, which may be internal and / or external. In some examples, communication circuitry 168 may communicate with a local external device, and processing circuitry 160 may communicate with a networked computing device via the local external device and a computer network, such asthe Medtronic CareLink® Network developed by Medtronic, Inc. In some examples, processing circuitry 160 may control communication circuitry 168 to send collected sensor data (such as for establishing a baseline), and / or information associated with a fall to external device 30, based on the occurrence of a fall, periodically, at the request of external device 30, or the like. Such information associated with a fall may include a time and / or date of the fall, pre-impact signals, signals during impact, and post-impact signals. For example, signals may include accelerometer signal(s) and / or other sensor signals. In some examples, the information may include determined first characteristics, determined second characteristics, sensed physiological parameter(s), posture angle(s), change in posture angle, trajectory, estimated body part to first contact a surface, and / or the like.

[0138] A clinician or other user may retrieve data from IMD 10 using external device 30 or another local or networked computing device configured to communicate with processing circuitry 160 via communication circuitry 168. The clinician may also program parameters of IMD 10 using external device 30 or another local or networked computing device.

[0139] FIG. 8 is a functional block diagram illustrating an example configuration of an external device 30. In some examples, external device 30 is configured to communicate with one or more HMDs 10. In some examples, external device 30 is not configured to communicate with one or more HMDs 10. In some examples, external device 30 is configured to implement the techniques of this disclosure. In some examples, external device 30 may be a smart phone or a wearable device, such as a smart watch or fitness tracker.

[0140] In the example of FIG. 8, external device 30 includes processing circuitry 200, memory 202, user interface (UI) 204, communication circuitry 206, and sensors 208. In some examples, external device 30 may correspond to any of external devices 30A-30D described with respect to FIGS. 1, 2, and 4A-5. External device 30 may be a dedicated hardware device with dedicated software for the programming and / or interrogation of an IMD 10. Alternatively, external device 30 may be an off-the-shelf computing device, e.g., a smart phone running a mobile application that enables external device 30 to program and / or interrogate IMD 10. In some examples where external device 30 is a smart phone, external device 30 may include a mobile application to facilitate interaction with IMD 10.In some examples, external device 30 may be a wearable device, such as a smart watch or fitness tracker.

[0141] In some examples, a user of external device 30 may be clinician, physician, heath care giver, patient, family member of the patient or friend of the patient. In some examples, a user uses external device 30 to select or program any of the values for operational parameters of IMD 10. In some examples, a user uses external device 30 to receive data collected by IMD 10, such as accelerometer signals, physiological signals, ECG data, an indication of a severity of an impact event, and / or an alert regarding an impact event, or other operational and performance data of IMD 10. In some examples, the user may also receive alerts provided by IMD 10 that indicate that a severity of a fall event. In some examples, IMD 10 may transmit an alert and the user may receive the alert that the patient has experienced a relatively severe impact event (e.g., a severe fall) and is likely to need medical attention. The user may interact with external device 30 via UI 204, which may include a display to present a graphical user interface to a user, and a keypad or another mechanism (such as a touch sensitive screen) for receiving input from a user. External device 30 may communicate wirelessly with IMD 10 using communication circuitry 206, which may be configured for RF communication with communication circuitry 168 of IMD 10. Sensors 208 may include accelerometer circuitry 212 (shown as ACCLRM 212). Accelerometer circuitry 212 may be configured to generate at one or more signals indicative of motion and / or posture of a person.

[0142] Processing circuitry 200 may include any combination of integrated circuitry, discrete logic circuitry, analog circuitry, such as one or more microprocessors, digital signal processors (DSPs), application specific integrated circuits (ASICs), or field-programmable gate arrays (FPGAs). In some examples, processing circuitry 200 may include multiple components, such as any combination of one or more microprocessors, one or more DSPs, one or more ASICs, or one or more FPGAs, as well as other discrete or integrated logic circuitry, and / or analog circuitry. Processing circuitry 200 may be communicatively coupled elements of external device 30 such as memory 202, accelerometry circuitry 212, and microphone 210.

[0143] Memory 202 may store program instructions, which may include one or more program modules, which are executable by processing circuitry 200. When executed by processing circuitry 200, such program instructions may cause processing circuitry 200 andexternal device 30 to provide the functionality ascribed to them herein. The program instructions may be embodied in software, firmware and / or RAMware. Memory 202 may include any volatile, non-volatile, magnetic, optical, or electrical media, such as a random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), electrically-erasable programmable ROM (EEPROM), flash memory, or any other digital media.

[0144] In some examples, processing circuitry 200 of external device 30 may be configured to provide some or all of the functionality ascribed to processing circuitry 160 (FIG. 7) of IMD 10 herein. For example, processing circuitry 200 may process the one or more accelerometer signals from accelerometer 212 or from accelerometer 167 (FIG. 7) to determine first characteristics corresponding to a first period of time before the point in time corresponding to the impact event and to determine second characteristics corresponding to a second period of time after the point in time corresponding to the impact event. Processing circuitry 200 may determine a severity of the impact event based on the first characteristics and the second characteristics. Processing circuitry 200 may determine whether to output an alert based on the severity of the impact event.

[0145] FIG. 9 is a functional block diagram illustrating an example system that includes external computing devices, such as external device 224, which may include one or more servers, and one or more other computing devices 230A-230N, that are coupled to IMD 10 and external device 30 via a network 222. In this example, IMD 10 may use its communication circuitry 168 to, e.g., at different times and / or in different locations or settings, communicate with external device 30 via a first wireless connection, and to communication with an access point 220 via a second wireless connection. In the example of FIG. 9, access point 220, external device 30, external device 224, and computing devices 230A-230N are interconnected, and able to communicate with each other, through network 222.

[0146] Access point 220 may comprise a device that connects to network 222 via any of a variety of connections, such as telephone dial-up, digital subscriber line (DSL), or cable modem connections. In other examples, access point 220 may be coupled to network 222 through different forms of connections, including wired or wireless connections. In some examples, access point 220 may be co-located with patient 14. Access point 220 may interrogate IMD 10, e.g., periodically or in response to a command from patient 14 ornetwork 222, to retrieve accelerometer signals, physiological signals, ECG data, a severity of an impact event, alerts regarding impact events, alerts of acute cardiac events, and / or other operational or patient data from IMD 10. Access point 220 may provide the retrieved data to external device 224 via network 222.

[0147] In some cases, external device 224 may be configured to provide a secure storage site for data that has been collected from IMD 10 and / or external device 30. In some cases, external device 224 may assemble data in web pages or other documents for viewing by trained professionals, such as clinicians, via computing devices 230A-230N. The illustrated system of FIG. 9 may be implemented, in some aspects, with general network technology and functionality similar to that provided by the Medtronic CareLink® Network developed by Medtronic pic, of Dublin, Ireland.

[0148] In some examples, one or more of access point 220, external device 224, or computing devices 230 may be configured to perform, e.g., may include processing circuitry configured to perform, some or all of the techniques described herein, e.g., with respect to processing circuitry 160 of IMD 10 and processing circuitry 200 of external device 30, relating to determining a severity of a fall. In the example of FIG. 9, external device 224 includes a memory 226 to store signals / statistics or a severity of risk of impact event 178 received from IMD 10 and / or external device 30, and processing circuitry 228, which may be configured to provide some or all of the functionality ascribed to processing circuitry 160 of IMD 10 and processing circuitry 200 of external device 30 herein. For example, processing circuitry 228 may determine severity of impact event 178 (FIGS. 7 and 8), and / or may obtain severity of impact event 178 from one or more IMDs 10 or external device 30. Processing circuitry 228 may determine the first characteristics, determine the second characteristics, and / or determine severity of impact event 178 in the manner described above with respect to processing circuitry 160 (FIG. 7) of IMD 10. In some examples, processing circuitry 228 may train and / or execute machine learning model 174 (FIGS. 7 and 8), for example, using data from a plurality of patients and / or electronic healthcare records, which may be stored in memory 226.

[0149] FIGS. 10A-10C are graphical diagrams illustrating example accelerometer signals and spectrograms of an accelerometer of an IMD surrounding various forward fall events. FIG. 10A depicts examples of a forward fall onto a 4-inch-thick mat. FIG. 10B depicts examples of a forward fall onto a 2-inch-thick mat. FIG. 10C depicts examples of aforward fall onto hard ground. As can be seen, there are differences between the both the accelerometer signals and the spectrogram data of the accelerometer signals surrounding the impact event (e.g., impact with the mat or the hard ground). As such, it may be possible for a system to use characteristics of the accelerometer signals to determine a severity of a fall, as a forward fall onto a hard surface is likely more severe than a forward fall onto a 4-inch mat. Specific characteristics may include the duration for which the acceleration exceed a threshold, or the duration of accelerometer signal oscillation or ringing as a result of the impact. As can be seen, there is more oscillation or ringing in the accelerometer signals of FIG. IOC than in the accelerometer signals of FIG. 10A, which may be indicative of a more severe impact event.

[0150] FIG. 11 is a flow diagram illustrating example techniques for determining a severity of an impact event and determining whether to send an alert according to one or more aspects of this disclosure. The techniques of FIG. 11 are described with respect to IMD 10 (FIG. 7), but may be practiced by any device or system capable of doing so. In some examples, the techniques of FIG. 11 may be performed by an external medical device.

[0151] IMD 10 may determine whether an impact is detected (240). For example, IMD 10 may monitor one or more accelerometer signals from accelerometer 167 to determine whether the accelerometer signal(s) include a spike whose magnitude meets a threshold and / or include a spike whose width meets a threshold for at least a predetermined period to time. If an impact is not detected (the “NO” path from box 240), IMD 10 may continue to monitor for potential impacts. If an impact is detected (the “YES” path from box 240), IMD 10 may perform any of, any combination of, or all of, the following: analyze the impact response (242), analyze pre-impact posture (244), analyze pre-impact rate and duration of acceleration change (246), analyze post-impact activity (248), and / or analyze post-impact vital signs (250). In some examples, IMD 10 may execute machine learning model 174 to perform any of, any combination of, or all of analyses 242, 244, 246, 248, and / or 250. While shown in FIG. 11 as being performed in parallel, any of these steps may be performed in parallel or in series.

[0152] For example, IMD 10 may analyze the impact response (242). IMD 10 may process one or more accelerometer signals (e.g., from accelerometer 167) as part of analyzing the impact response. For example, IMD 10 may determine a maximum acceleration of an accelerometer signal (e.g., a maximum of an acceleration “spike”)associated with the impact. For example, IMD 10 may analyze a morphology of one or more accelerometer signals, such as to determine to what extent there is a “ringing” in the signal after a detected impact where an accelerometer signal oscillates for a period of time after the initial impact. A larger maximum acceleration and / or a greater oscillation (in terms of length of time and / or magnitude) may be indicative of a more severe impact. In some examples, IMD 10 may analyze a morphology of one or more accelerometer signals to determine whether there is an indication that patient 14 braced themselves during the fall, such as with their hand, arm, shoulder, buttocks, etc. This bracing may present itself in the one or more accelerometer signals as individual or overlapping impacts. Bracing, particularly by the hand, arm and / or shoulder, may be indicative of a less severe impact.

[0153] IMD 10 may analyze pre-impact posture (244). For example, IMD 10 may determine a pre-impact posture of patient 14 based on the one or more accelerometer signals. For example, signals from a 3-axis accelerometer may be indicative of a posture of patient 14. The pre-impact posture of patient 14 may be indicative of the trajectory of the fall and / or a body part most likely to impact the surface of the object impacted first. Such information may be indicative of the severity of a fall. For example, if a person falls forward, they are more likely to be able to brace themselves with their hands and / or arms if they are conscious. If a person falls sideways their shoulder is more likely to be the body part first impacting the surface, at least partially protecting their head from impacting the surface. However, if a person falls backwards, they are less likely to be able to brace themselves and, even if their buttocks are the first body part to impact the surface, they are more likely to strike their head on the surface (assuming they do not end up in a seated position and avoid a prone position), than if they had fallen forward or sideways.

[0154] The pre-impact posture of patient 14 may also be indicative of a distance of the fall. For example, if patient 14 is sitting pre-impact, the distance of the fall is likely to be shorter than if the patient 14 was upright. The longer the distance of the fall may be indicative of a more severe impact.

[0155] IMD 10 may analyze pre-impact rate and duration of acceleration change (246). For example, IMD 10 may process the one or more accelerometer signals from accelerometer 167 to determine, during a period of time before the impact, a rate of acceleration change and a duration of acceleration change. The period of time may be of a predetermined length. IMD 10 may determine the rate of acceleration change and / or theduration of acceleration time based on the one or more accelerometer signals during the period of time. For example, a lower acceleration for a longer duration may result in a higher pre-impact velocity which may cause a relatively high impact severity. A higher rate of acceleration for a shorter duration may also result in a high pre-impact velocity. In some examples, the dynamic range of an accelerometer such as accelerometer 167 may be limited, in which case accelerometer signal(s) may be saturated and a true maximum acceleration may not be accurately determinable from the accelerometer signals. As such, IMD 10 may use the derivative of the acceleration (e.g., the rate of acceleration change) to assess impact severity. For example, IMD 10 may estimate a maximum acceleration based on the rate of acceleration change, which IMD 10 may use when assessing the impact severity.

[0156] IMD 10 may analyze post-impact activity (248). For example, IMD 10 may analyze the one or more accelerometer signals to determine any activity and / or posture of patient 14 after the impact. No activity and / or a prone posture may be more indicative of a severe impact than activity and / or an upright or seated posture.

[0157] IMD 10 may analyze post-impact vital signs (250). A severity of an impact event on a patient may not necessarily be directly correlated with a level of mechanical impact (e.g., as determined based on one or more accelerometer signals). For example, a patient may fall onto a softer surface, but fall in such a way as to severely injure themselves. A patient may also fall onto a harder surface, but fall in such a way as to not severely injure themselves. As such, it may be desirable to take physiological parameters into account when determining a severity of an impact event. For example, IMD 10 may sense signals indicative of a number of physiological parameters and may analyze such physiological parameters when determining a severity of an impact event. IMD 10 may analyze heart rate, heart rate variability, R-wave amplitude, respiration rate, respiration effort, blood pressure, temperature, oxygen saturation, fluid status (e.g., perfusion), arterial blood pulsation amplitude, pulse transit time, pulse wave velocity, and / or the like. In some examples, IMD 10 may compare post-impact vital signs with pre-impact vital signs, baseline vital signs, and / or clinically acceptable ranges for such vital signs. Certain values of post-impact vital signs (e.g., no heart rate) and / or changes in values from pre-impact vital signs to post-impact vital signs (e.g., an relatively large increase in heart rate variability), from baseline vital signs to post-impact vital signs, post-impact vital signs outside of clinically acceptable ranges may be indicative of a more severe impact than others.

[0158] IMD 10 may assign a severity of the impact event (252). For example, IMD 10 may utilize any of, any combination of, or all of analyses 242, 244, 246, 248, and / or 250 to determine a severity of the impact event or to adjust a determination of a severity of the impact event. For example, each of the analyses 242, 244, 246, 248, and / or 250 that may be used in the determination of the severity of the impact may assign the impact a value and IMD 10 may apply a formula to the values to determine a severity of the impact event. In some examples, the severity of the impact event may comprise an impact score. In some examples, IMD 10 may execute machine learning model 174 to determine the severity of the impact event based on any of, any combination of, or all of analyses 242, 244, 246, 248, and / or 250.

[0159] IMD 10 may record the incidence of the impact, including the severity of the impact event (254). For example, IMD 10 may record the incidence of the impact in severity of impact event 178 in memory 170.

[0160] IMD 10 may determine whether to send an alert or notification of the impact / fall to another device. For example, IMD 10 may compare the severity of the impact event to a threshold. If the severity of the impact event meets the threshold (e.g., is greater than, or greater than or equal to the threshold), IMD 10 may notify the other device of the fall (256). This notification may include an alert. The alert may include information regarding the fall, such as time, date, location, the determined severity of the impact event, accelerometer signals, physiological parameters and / or physiological signals, fall trajectory, estimation of first body part to contact the surface, ECG data, and / or any other data collected by IMD 10 which may be useful to a clinician, emergency medical technician, or other device that may evaluate the data.

[0161] FIG. 12A is a flow diagram illustrating example techniques for determining a severity of an impact according to one or more aspects of this disclosure. The techniques of FIG. 12A are described with respect to IMD 10 (FIG. 7) and external device 30 (FIG. 8), but may be practiced by any devices capable of doing so. For example, an external medical device may perform some or all of the actions attributed to IMD 10. For example, external device 224 may perform some or all of the actions attributed to external device 30. Additionally, the division of steps between IMD 10 and external device 30 may be different than depicted in FIG. 12A and still remain within the scope of this disclosure. In the example of FIG. 12 A, the determination of an impact severity index occurs on anexternal device, not on IMD 10. In some examples, a more complicated algorithm may be used to determine the impact severity index in the example of FIG. 12Athan if the determination of the impact severity index were to occur on IMD 10, because more processing power may be available on the external device.

[0162] IMD 10 may log diagnostic data (260). For example, IMD 10 may monitor various physiological signals from sensors 166 and / or other sensors during normal daily activities of patient 14. IMD 10 may send such diagnostic data to external device 30. External device 30 may use the diagnostic data to determine a baseline for each diagnostic parameter (262). For example, external device 30 may determine a respective baseline for each of a plurality of physiological parameters of patient 14 based on the logged diagnostic data. The diagnostic data used to determine the baseline may be collected over a period of one or more weeks to one or more years. A baseline may be patient-specific because the baseline is based on the diagnostic data logged by IMD 10 which is specific to the patient. In some examples, each respective baseline may represent a range of values of a physiological parameter. These ranges may or may not be as extensive as all values of the physiological parameter collected over the period of time. For example, external device 30 may discard or ignore extraneous values when determining a respective baseline.

[0163] IMD 10 may monitor accelerometer signals (264). For example, IMD 10 may monitor accelerometer signals of accelerometer 167. IMD 10 may determine whether the accelerometer data of the accelerometer signals meets a threshold (266). For example, IMD 10 may compare the accelerometer data to a threshold indicative of a fall. If the accelerometer data does not meet the threshold (the “NO” path from box 266), IMD 10 may determine a fall has not occurred and may continue to monitor accelerometer signals (264). If the accelerometer data does meet the threshold (the “YES” path from box 266), IMD 10 may analyze pre- and post-impact activity levels (268). For example, IMD 10 may analyze the accelerometer signals to determine pre-impact activity levels and post-impact activity levels. IMD 10 may send the results of the analysis to external device 30.

[0164] IMD 10 may determine whether the post-impact activity level or a change in impact activity level meets a threshold (270). Meeting a threshold may include being greater than, greater than or equal to, less than, or less than or equal to based on the usage. For example, IMD 10 may compare the post-impact activity level or the change in impact activity level to the threshold. If the post-impact or change in impact activity level does notmeet the threshold (the “NO” path from box 270), IMD 10 may determine the impact to not be severe and may continue to monitor the accelerometer signals (264).

[0165] If the post-impact or change in impact activity level does meet the threshold (the “YES” path from box 270), IMD 10 may estimate which part of the body of patient 14 was impacted (272). For example, IMD 10 may utilize pre-impact posture and post-impact posture to determine a likely body part to be impacted during the fall. IMD 10 may send the estimate to external device 30.

[0166] IMD 10 may analyze pre- and post-impact diagnostic data (274). For example, IMD 10 may determine various physiological parameters pre- and / or post-impact and send such information to external device 30.

[0167] External device 30 may receive, from IMD 10, the results of the analysis of the pre- and post-impact activity levels (268), the estimate of which body part was impacted (272), and the results of the analysis of the pre- and post-impact diagnostic data (274). External device 30 may compute an impact severity index (276). For example, external device 30 may calculate an impact severity score based on the pre- and post-impact activity levels, the estimate of which part of the body was impacted, the pre- and post-impact diagnostic data, and, in some examples, the baseline for the diagnostic parameters. In some examples, external device 30 may execute machine learning model 174 to compute the impact severity index.

[0168] External device 30 may determine whether the impact severity index meets a threshold (278). If the impact severity index does meet the threshold (e.g., is greater than, or greater than or equal to the threshold) (the “YES” path from box 278), external device 30 may send a notification (279). For example, external device 30 may send an alert to a user of external device 30, to external device 224 and / or any of computing devices 230 (FIG. 9). The alert may be indicative that patient 14 has had a relatively severe fall.

[0169] In some examples, external device 30 may log event information associated with a fall whether or not the impact severity index meets the threshold. For example, external device 30 may store information associated with the fall in memory 202 (e.g., characteristics 179). Such information may include a time and / or date of the fall, pre-impact signals, signals during impact, and post-impact signals. For example, signals may include accelerometer signal(s) and / or other sensor signals. In some examples, the information may include determined first characteristics, determined second characteristics, sensed physiologicalparam eter(s), posture angle(s), change in posture angle, trajectory, estimated body part to first contact a surface, and / or the like. External device 30 may use such information for longterm monitoring and / or predicting of a risk of a fall and / or a severity of a fall. In some examples, external device 30 may send such information to external device 224 which may use such information for long-term monitoring and / or predicting of a risk of a fall and / or a severity of a fall.

[0170] FIG. 12B is a flow diagram illustrating other example techniques for determining a severity of an impact according to one or more aspects of this disclosure. The techniques of FIG. 12B are described with respect to IMD 10 (FIG. 7) and external device 30 (FIG. 8), but may be practiced by any devices capable of doing so. For example, an external medical device may perform some or all of the actions attributed to IMD 10. For example, external device 224 may perform some or all of the actions attributed to external device 30. Additionally, the division of steps between IMD 10 and external device 30 may be different than depicted in FIG. 12B and still remain within the scope of this disclosure. In the example of FIG. 12B, the determination of an impact severity index and whether to send a notification occurs on IMD 10, not on an external device. In some examples, because the decision of whether to send a notification occurs on IMD 10, the notification may be sent earlier, than if that determination were to be made by an external device. For example, if IMD 10 may send the notification (e.g., to external device 30 and / or external device 224) upon determination that a notification should be sent, rather than send data during a next periodic transmission.

[0171] It should be noted that the example of FIG. 12A and the example of FIG. 12B are not mutually exclusive. In other words, a system may be implemented where a first patient may have a first IMD which interacts with an external device in the manner of FIG. 12A and a second patient may have a second IMD which interacts with the external device (or a different external device) in the manner of FIG. 12B.

[0172] IMD 10 may log diagnostic data (280). For example, IMD 10 may monitor various physiological signals from sensors 166 and / or other sensors during normal daily activities of patient 14. IMD 10 may send such diagnostic data to external device 30. External device 30 may use the diagnostic data to determine a baseline for each diagnostic parameter (282). For example, external device 30 may determine a baseline for a pluralityof physiological parameters of patient 14 based on the logged diagnostic data, as discussed above with respect to box 262 of FIG. 12A.

[0173] IMD 10 may monitor accelerometer signals (284). For example, IMD 10 may monitor accelerometer signals of accelerometer 167. IMD 10 may determine whether the accelerometer data of the accelerometer signals meets a threshold (286). For example, IMD 10 may compare the accelerometer data to a threshold indicative of a fall. If the accelerometer data does not meet the threshold (the “NO” path from box 286), IMD 10 may determine a fall has not occurred and may continue to monitor accelerometer signals (284). If the accelerometer data does meet the threshold (the “YES” path from box 286), IMD 10 may analyze pre- and post-impact activity levels (288). For example, IMD 10 may analyze the accelerometer signals to determine pre-impact activity levels and post-impact activity levels. IMD 10 may send the results of the analysis to external device 30.

[0174] IMD 10 may determine whether the post-impact activity level or a change in impact activity level meets a threshold (290). Meeting a threshold may include being greater than, greater than or equal to, less than, or less than or equal to based on the usage. For example, IMD 10 may compare the post-impact activity level or the change in impact activity level to the threshold. If the post-impact or change in impact activity level does not meet the threshold (the “NO” path from box 290), IMD 10 may determine the impact to not be severe and may continue to monitor the accelerometer signals (284).

[0175] If the post-impact or change in impact activity level does meet the threshold (the “YES” path from box 290), IMD 10 may estimate which part of the body of patient 14 was impacted (292). For example, IMD 10 may utilize pre-impact posture and post-impact posture to determine a likely body part to be impacted during the fall. IMD 10 may send the estimate to external device 30.

[0176] IMD 10 may analyze pre- and post-impact diagnostic data (294). For example, IMD 10 may determine various physiological parameters pre- and / or post-impact and send such information to external device 30.

[0177] IMD 10 may receive, from external device 30, baseline statistics for each diagnostic parameter. IMD 10 may store the baseline statistics for each diagnostic parameter (295), for example, in memory 170. IMD 10 may, based on the results of the analysis of the pre- and post-impact activity levels (288), the estimate of which body part was impacted (292), and / or the results of the analysis of the pre- and post-impact diagnostic data (294),compute an impact severity index (296). For example, external device 30 may calculate an impact severity score based on the pre- and post-impact activity levels, the estimate of which part of the body was impacted, the pre- and post-impact diagnostic data, and, in some examples, the baseline for the diagnostic parameters. In some examples, external device 30 may execute machine learning model 174 to compute the impact severity index.

[0178] IMD 10 may determine whether the impact severity index meets a threshold (298). If the impact severity index does meet the threshold (e.g., is greater than, or greater than or equal to the threshold) (the “YES” path from box 298), IMD 10 may send a notification (299). For example, IMD 10 may send an alert to a user of external device 30, to external device 224 and / or any of computing devices 230 (FIG. 9). The alert may be indicative that patient 14 has had a relatively severe fall.

[0179] In some examples, IMD 10 may log event information associated with a fall whether or not the impact severity index meets the threshold. For example, IMD 10 may store information associated with the fall in memory 170. Such information may include a time and / or date of the fall, pre-impact signals, signals during impact, and post-impact signals. For example, signals may include accelerometer signal(s) and / or other sensor signals. In some examples, the information may include determined first characteristics, determined second characteristics, sensed physiological parameter(s), posture angle(s), change in posture angle, trajectory, estimated body part to first contact a surface, and / or the like. IMD 10 may use such information for long-term monitoring and / or predicting of a risk of a fall and / or a severity of a fall. In some examples, IMD 10 may send such information to external device 30 and / or external device 224 which may use such information for longterm monitoring and / or predicting of a risk of a fall and / or a severity of a fall.

[0180] FIG. 13 is a graphical diagram illustrating example spike magnitudes and associated posture angle changes for a variety of fall types and activities of daily living (ADL) as determined by an IMD from accelerometer signals. In some examples, IMD 10 may employ a relatively simple version of a fall detection algorithm to determine whether a fall has occurred. For example, IMD 10 may compare an acceleration magnitude in an accelerometer signal to a fixed threshold. When IMD 10 does determine that an acceleration magnitude meets the fixed threshold (e.g., is greater than, is greater than or equal to, is less than, or is less than or equal to depending on the usage), IMD 10 then compare a posture angle change between a posture angle before the spike in the accelerometer signal(s) to aposture angle after the spike to a fixed threshold. However, to improve discrimination between falls and ADL, it may be desirable for the spike threshold to depend on the angle change threshold, and vice versa.

[0181] As such, IMD 10 may use a more formulaic threshold. In a relatively simple example, IMD 10 may employ a threshold 300. The application of threshold 300 may be as shown in Table 1.Table 1

[0182] For example, IMD 10 may determine whether the acceleration magnitude of a spike in an accelerometer signal has a maximum that is > 200. If the acceleration magnitude of the spike in the accelerometer signal is not > 200, IMD 10 may determine no fall has occurred and continue to monitor acceleration magnitude of the accelerometer signals. If the acceleration magnitude of the spike in the accelerometer signal is > 200, IMD 10 may perform a posture change detection which may determine a posture change angle between a posture angle before in the accelerometer signal(s) and a posture angle after the spike.

[0183] If the acceleration magnitude of the spike in the accelerometer signal is between 200 and 350, IMD 10 may determine whether the posture change angle is < 180 — (ac cel spike mag) . (accel spike mag)- - - , it the posture change angle is < 180 - - - , IMD 10 maydetermine no fall has occurred and continue to monitor acceleration magnitude of the accelerometer signals. If IMD 10 determine that the posture change angle is > 180 — (accei spike mag)may determine that a fall has occurred.

[0184] If the acceleration magnitude of the spike in the accelerometer signal is > 350, then IMD 10 may determine whether the posture change angle is > 70 degrees. If the posture change angle is > 70 degrees, IMD 10 may determine that a fall has occurred. If the posture change angle is not > 70 degrees, IMD 10 may determine a fall has not occurred and continue to monitor acceleration magnitude of the accelerometer signals.

[0185] In another example, IMD 10 may determine whether all of the following conditions are true: 1) the spike in an accelerometer signal has a maximum that is > 200;. . . . (accel spike maq) . . .2) the posture change angle is > 180 - - - ; or 3) the posture change angle is> 70 degrees. If all of the conditions are true, IMD 10 may determine that a fall has occurred. If any one of the conditions is not true, IMD 10 may determine a fall has not occurred and continue to monitor acceleration magnitude of the accelerometer signals.

[0186] FIG. 14 is a flowchart illustrating an example of techniques for determining a severity of an impact event in accordance with one or more aspects of this disclosure. This example may be implemented by any one of the IMDs discussed above, because each one of the same is configured to include accelerometer circuitry, as well as communication and processing circuitry (see FIG. 7 and corresponding description) to facilitate determining a severity of an impact event based on accelerometer circuitry-generated signal(s). This example may also be implemented by any device having an accelerometer that is capable of implementing the techniques of this disclosure. For example, the techniques of FIG. 14 may be performed by an external medical device.

[0187] IMD 10 may detect, based on the one or more accelerometer signals, an impact event corresponding to patient 14, the impact event occurring at a point in time (310). For example, IMD 10 may monitor the one or more accelerometer signals from accelerometer 167 and compare a maximum acceleration indicated in an accelerometer signal and / or a change of posture angle to a threshold of threshold(s) 176 to detect an impact event.

[0188] IMD 10 may process the one or more accelerometer signals to determine a plurality of first characteristics corresponding to a first period of time before the point in time corresponding to the impact event (312). The first period of time may be in the order of one or more seconds to one or more minutes. The plurality of first characteristics may include at least one of a maximum acceleration, a pre-impact morphology, a pre-impact posture, a pre-impact rate of acceleration change, or a pre-impact duration of acceleration change.

[0189] IMD 10 may process the one or more accelerometer signals to determine a plurality of second characteristics corresponding to a second period of time following the point in time corresponding to the impact event (314). The second period of time may be in the order of one or more seconds to one or more minutes. The plurality of secondcharacteristics comprising at least one of a post-impact morphology, a post-impact posture, or a post impact activity level.

[0190] IMD 10 may determine a severity of the impact event based on the first characteristics and the second characteristics (316). For example, IMD 10 may generate a severity of impact score for the impact event based on the first characteristics and the second characteristics.

[0191] IMD 10 may determine whether to output an alert based on the severity of the impact event (318). For example, IMD 10 may compare the severity of impact score to a threshold of threshold(s) 176 to determine whether the severity of impact score meets the threshold. If the severity of impact score meets the threshold, IMD 10 may determine to output the alert, generate the alert, and output the alert to external device 30, external device 224, and / or another device.

[0192] In some examples, IMD 10 may, as part of detecting the impact event, compare at least one of an amplitude of an accelerometer signal or a change in posture angle from a first posture angle at a first time to a second posture angle at a second time to a threshold, wherein the threshold includes an amplitude component and a change in posture angle component.

[0193] In some examples, IMD 10 may generate, using the one or more sensors, a set of physiological signals, wherein each physiological signal of the set of physiological signals indicates one or more physiological parameters of the patient. In some examples, IMD 10 may process the set of physiological signals to determine one or more third characteristics each corresponding to a respective third period of time before the point in time. In some examples, the respective third periods of time may be different for each characteristic of the one or more characteristics. The respective third periods of time may be in the range of one or more seconds to one or more minutes.

[0194] IMD 10 may process the set of physiological signals to determine one or more fourth characteristics each corresponding to a respective fourth period of time following the point in time. In some examples, the respective fourth periods of time may be different for each characteristic of the one or more characteristics. The respective fourth periods of time may be in the range of one or more seconds to one or more minutes.

[0195] IMD 10 may determine the severity of the impact event based on the plurality of first characteristics, the plurality of second characteristics, the one or more thirdcharacteristics, and the one or more fourth characteristics. In some examples, the one or more physiological parameters include at least one of heart rate, heart rate variability, R- wave amplitude, respiration rate, respiration effort, blood pressure, temperature, oxygen saturation, fluid status, arterial blood pulsation amplitude, pulse transit time, or pulse wave velocity.

[0196] In some examples, IMD 10 may process the set of physiological signals to determine one or more baseline characteristics each corresponding to a respective fifth period of time before the point in time, wherein each respective baseline characteristic defines a respective range. In some examples, the respective fifth periods of time may be different for each characteristic of the one or more characteristics. The respective fifth periods of time may be in the range of one or more weeks to one or more years. Each respective range may be a normal range of a corresponding physiological parameter for the patient. These respective ranges may be patient-specific as they may be determined based on the set of physiological signals of the patient. It should be understood that the respective ranges may or may not include all values of a particular physiological parameter represented corresponding physiological signal(s) during the respective fifth period of time. For example, IMD 10 may remove outlying values when determining the one or more baseline characteristics. In some examples, IMD 10 may determine the severity of the impact event based on the plurality of first characteristics, the plurality of second characteristics, the one or more baseline characteristics, and the one or more fourth characteristics.

[0197] In some examples, each of the one or more fourth characteristics has a corresponding acceptable clinical range. In some examples, IMD 10 may determine the severity of the impact event based on the plurality of first characteristics, the plurality of second characteristics, the one or more fourth characteristics, and each acceptable clinical range.

[0198] In some examples, IMD 10 may estimate a body part that first contacted a surface that was impacted during the impact event based on the pre-impact posture and the postimpact posture, and wherein the determined severity of the impact event is at least partially based on the estimate of the body part that first contacted the surface. In some examples, IMD 10 may determine a traj ectory of the patient prior to the impact event based on the preimpact posture and the post-impact posture, and wherein the determined severity of the impact event is at least partially based on the trajectory.

[0199] In some examples, IMD 10 may include machine learning model 174, and processing circuitry 160 is configured to execute machine learning model 174 to determine the severity of the impact event.

[0200] In some examples, the impact event is a result of a fall. IMD 10 may estimate a reason for the fall based on at least one of the one or more accelerometer signals, the one or more physiological parameters, an electrocardiogram sensed by one or more of the one or more sensors, or an identification of a medication of the patient. In some examples, the reason for the fall includes at least one of a cardiac reason, or a non-cardiac reason. In some examples, the non-cardiac reason includes a collision with an object.

[0201] In some examples, accelerometer 167 has a measurement range of at least plus or minus 10G and at least 12-bit resolution. In some examples, accelerometer 167 has a sampling rate of the accelerometer is at least 1000 Hz.

[0202] FIG. 15 is a conceptual diagram illustrating an example machine learning model 320 configured to determine a severity of an impact event, such as a fall. Machine learning model 320 may be an example of machine learning model 174 of FIGS. 7 and 8. In some examples, machine learning model 320 is an example of a set of rules implemented by processing circuitry 160 or processing circuitry 200. Machine learning model 320 may be an example of a deep learning model, or deep learning algorithm, trained to determine a severity of an impact event based on one or more accelerometer signals as described herein.

[0203] One or more of IMD 10, external device 30, and / or external device 224 may train, store, and / or utilize machine learning model 320, but other devices may apply inputs associated with a particular patient to machine learning model 320 in other examples. In some examples, machine learning model 320 may include a convolutional neural network (CNN) model of ResNet-18. Some non-limiting examples of models that may be used for transfer learning include AlexNet, VGGNet, GoogleNet, ResNet50, or DenseNet, etc. Some non-limiting examples of machine learning techniques include Support Vector Machines, K-Nearest Neighbor algorithm, and Multi-layer Perceptron.

[0204] As shown in the example of FIG. 15, machine learning model 320 may include three layers. These three layers include input layer 322, hidden layer 324, and output layer 326. Output layer 326 comprises the output from the transfer function 325 of output layer 326. Input layer 322 represents each of the input values XI through X4 provided tomachine learning model 320. The number of inputs may be equal to, less than, or greater than 4, including much greater than 4, e.g., hundreds or thousands. In some examples, the input values may any of the of values input into a machine learning model, as described above. In some examples, input values may include samples of one or more accelerometer signals, one or more physiological signals, an ECG signal, and / or other information useful in determining a severity of an impact. In addition, in some examples input values of machine learning model 320 may include additional data, such medications patient 14 may be taking, personal health information from electronic healthcare records, which may be useful for determining a likely reason for a fall.

[0205] Each of the input values for each node in the input layer 322 is provided to each node of hidden layer 324. In the example of FIG. 15, hidden layers 324 include two layers, one layer having four nodes and the other layer having three nodes, but fewer or greater number of nodes may be used in other examples. Each input from input layer 322 is multiplied by a weight and then summed at each node of hidden layers 324. During training of machine learning model 320, the weights for each input are adjusted to establish the relationship between the inputs, e.g., one or more accelerometer signals, to determining severity of an impact event. In some examples, one hidden layer may be incorporated into machine learning model 320, or three or more hidden layers may be incorporated into machine learning model 320, where each layer includes the same or different number of nodes.

[0206] The result of each node within hidden layers 324 is applied to the transfer function of output layer 326. The transfer function may be liner or non-linear, depending on the number of layers within machine learning model 320. Example non-linear transfer functions may be a sigmoid function or a rectifier function. The output 327 of the transfer function may be a classification that indicates a severity of an impact event. In some examples, output 327 may include respective probabilities for a plurality of classifications.

[0207] By applying the accelerometer signal data and / or other patient data to a machine learning model, such as machine learning model 320, processing circuitry, such as processing circuitry 160 of IMD 10 and / or processing circuitry 200 of external device 30, is able to determine a severity associated with an impact event. Processing circuitry 160 and / or processing circuitry 200 may use this determined severity to determine whether to send an alert to a user, a clinician, a caregiver, an EMT, a monitoring service, and / or thelike, indicative of the patient experiencing a severe fall. This may lead to the rendering of prompt assistance, treatment, and / or hospitalization, thereby improving patient outcomes.

[0208] FIG. 16 is an example of a machine learning model 320 being trained using supervised and / or reinforcement learning techniques. Machine learning model 320 may be implemented using any number of models for supervised and / or reinforcement learning, such as but not limited to, an artificial neural network, a decision tree, naive Bayes network, support vector machine, or k-nearest neighbor model, to name only a few of the examples discussed above. In some examples, processing circuitry one or more of IMD 10, external device 30, and / or external device 224 initially trains the machine learning model 320 based on training set data 400 including numerous instances of input data corresponding to falls and / or ADLs not constituting falls, e.g., as labeled by an expert. A prediction or classification by the machine learning model 320 may be compared 404 to the target output 403, e.g., as determined based on the label. Based on an error signal representing the comparison, the processing circuitry implementing a learning / training function 405 may send or apply a modification to weights of machine learning model 320 or otherwise modify / update the machine learning model 320. For example, one or more of IMD 10, external device 30, and / or external device 224 may, for each training instance in the training set 400, modify machine learning model 320 to change a score generated by the machine learning model 320 in response to data applied to the machine learning model 320.

[0209] Various aspects of the techniques may be implemented within one or more processors, including one or more microprocessors, DSPs, ASICs, FPGAs, or any other equivalent integrated or discrete logic circuitry, as well as any combinations of such components, embodied in programmers, such as physician or patient programmers, electrical stimulators, or other devices. The term “processor” or “processing circuitry” may generally refer to any of the foregoing logic circuitry, alone or in combination with other logic circuitry, or any other equivalent circuitry.

[0210] In one or more examples, the functions described in this disclosure may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software, the functions may be stored on, as one or more instructions or code, a computer-readable medium and executed by a hardware-based processing unit. Computer- readable media may include computer-readable storage media forming a tangible, non-transitory medium. Instructions may be executed by one or more processors, such as one or more DSPs, ASICs, FPGAs, general purpose microprocessors, or other equivalent integrated or discrete logic circuitry. Accordingly, the term “processor,” as used herein may refer to one or more of any of the foregoing structure or any other structure suitable for implementation of the techniques described herein.

[0211] In addition, in some aspects, the functionality described herein may be provided within dedicated hardware and / or software modules. Depiction of different features as modules or units is intended to highlight different functional aspects and does not necessarily imply that such modules or units must be realized by separate hardware or software components. Rather, functionality associated with one or more modules or units may be performed by separate hardware or software components, or integrated within common or separate hardware or software components. Also, the techniques could be fully implemented in one or more circuits or logic elements. The techniques of this disclosure may be implemented in a wide variety of devices or apparatuses, including an IMD, an external programmer, a combination of an IMD and external programmer, an integrated circuit (IC) or a set of ICs, and / or discrete electrical circuitry, residing in an IMD and / or external programmer.

[0212] This disclosure includes the following non-limiting examples.

[0213] Example 1. A medical device system comprising: a medical device comprising one or more sensors, wherein the medical device is configured to generate, using an accelerometer of the one or more sensors, one or more accelerometer signals indicating a motion of a patient; one or more memories; and processing circuitry in communication with the one or more memories, wherein the processing circuitry is configured to: detect, based on the one or more accelerometer signals, an impact event corresponding to the patient, the impact event occurring at a point in time; process the one or more accelerometer signals to determine a plurality of first characteristics corresponding to a first period of time before the point in time corresponding to the impact event, wherein the plurality of first characteristics comprise at least one of a maximum acceleration, a pre-impact morphology, a pre-impact posture, a pre-impact rate of acceleration change, or a pre-impact duration of acceleration change; process the one or more accelerometer signals to determine a plurality of second characteristics corresponding to a second period of time following the point in time corresponding to theimpact event, the plurality of second characteristics comprising at least one of a postimpact morphology, a post-impact posture, or a post impact activity level; determine a severity of the impact event based on the first characteristics and the second characteristics; and determine whether to output an alert based on the severity of the impact event.

[0214] Example 2. The medical device system of example 1, wherein as part of detecting the impact event, the processing circuitry is configured to compare at least one of an amplitude of an accelerometer signal or a change in posture angle from a first posture angle at a first time to a second posture angle at a second time to a threshold, wherein the threshold includes an amplitude component and a change in posture angle component.

[0215] Example 3. The medical device system of example 1 or example 2, wherein the medical device is further configured to generate, using the one or more sensors, a set of physiological signals, wherein each physiological signal of the set of physiological signals indicates one or more physiological parameters of the patient, and wherein the processing circuitry is further configured to: process the set of physiological signals to determine one or more third characteristics each corresponding to a respective third period of time before the point in time; process the set of physiological signals to determine one or more fourth characteristics each corresponding to a respective fourth period of time following the point in time; and determine the severity of the impact event based on the plurality of first characteristics, the plurality of second characteristics, the one or more third characteristics, and the one or more fourth characteristics.

[0216] Example 4. The medical device system of example 3, wherein the one or more physiological parameters comprise at least one of heart rate, heart rate variability, R-wave amplitude, respiration rate, respiration effort, blood pressure, temperature, oxygen saturation, fluid status, arterial blood pulsation amplitude, pulse transit time, or pulse wave velocity.

[0217] Example s. The medical device of system example 3 or example 4, wherein the processing circuitry is further configured to: process the set of physiological signals to determine one or more baseline characteristics each corresponding to a respective fifth period of time before the point in time, wherein each respective baseline characteristic defines a respective range; and determine the severity of the impact eventbased on the plurality of first characteristics, the plurality of second characteristics, the one or more baseline characteristics, and the one or more fourth characteristics.

[0218] Example 6. The medical device system of any of examples 3-5, wherein each of the one or more fourth characteristics has a corresponding acceptable clinical range, and wherein the processing circuitry is further configured to determine the severity of the impact event based on the plurality of first characteristics, the plurality of second characteristics, the one or more fourth characteristics, and each acceptable clinical range.

[0219] Example 7. The medical device system of any of examples 1-6, wherein the processing circuitry is further configured to estimate a body part that first contacted a surface that was impacted during the impact event based on the pre-impact posture and the post-impact posture, and wherein the determined severity of the impact event is at least partially based on the estimate of the body part that first contacted the surface.

[0220] Example 8. The medical device system of any of examples 1-7, wherein the processing circuitry is further configured to determine a trajectory of the patient prior to the impact event based on the pre-impact posture and the post-impact posture, and wherein the determined severity of the impact event is at least partially based on the trajectory.

[0221] Example 9. The medical device system of any of examples 1-8, further comprising a machine learning algorithm, wherein the processing circuitry is configured to execute the machine learning algorithm to determine the severity of the impact event.

[0222] Example 10. The medical device system of any of examples 1-9, wherein the impact event is a result of a fall and wherein the processing circuitry is further configured to estimate a reason for the fall based on at least one of the one or more accelerometer signals, the one or more physiological parameters, an electrocardiogram sensed by one or more of the one or more sensors, or an identification of a medication of the patient.

[0223] Example 11. The medical device system of example 10, wherein the reason for the fall comprises at least one of a cardiac reason, or a non-cardiac reason.

[0224] Example 12. The medical device system of example 11, wherein the noncardiac reason comprises a collision with an object.

[0225] Example 13. The medical device system of any of examples 1-12, wherein the accelerometer has a measurement range of at least plus or minus 10G and atleast 12-bit resolution and wherein a sampling rate of the accelerometer is at least 1000 Hz.

[0226] Example 14. The medical device system of any of examples 1-13, wherein the medical device comprises an implantable medical device and the implantable medical device comprises the processing circuitry.

[0227] Example 15. The medical device system of any one of examples 1-13, wherein the medical device comprises an implantable medical device, the medical device system further comprising an external device configured to communicate with the implantable medical device, the external device comprising the processing circuitry.

[0228] Example 16. The medical device system of any one of examples 1-13, wherein the medical device comprises an external medical device and the external medical device comprises the processing circuitry.

[0229] Example 17. A method comprising: detecting, by processing circuitry of a medical device system and based on one or more accelerometer signals, an impact event corresponding to a patient, the impact event occurring at a point in time; processing, by the processing circuitry, the one or more accelerometer signals to determine a plurality of first characteristics corresponding to a first period of time before the point in time corresponding to the impact event, wherein the plurality of first characteristics comprise at least one of a maximum acceleration, a pre-impact morphology, a pre-impact posture, a pre-impact rate of acceleration change, or a pre-impact duration of acceleration change; processing, by the processing circuitry, the one or more accelerometer signals to determine a plurality of second characteristics corresponding to a second period of time following the point in time corresponding to the impact event, the plurality of second characteristics comprising at least one of a post-impact morphology, a post-impact posture, or a post impact activity level; determining a severity of the impact event based on the first characteristics and the second characteristics; and determining whether to output an alert based on the severity of the impact event.

[0230] Example 18. The method of example 17, wherein detecting the impact event comprises comparing at least one an amplitude of an accelerometer signal or a change in posture angle from a first posture angle at a first time to a second posture angle at a second time to a threshold, wherein the threshold includes an amplitude component and a change in posture angle component.

[0231] Example 19. The method of example 17 or example 18, further comprising: obtaining, by the processing circuitry, a set of physiological signals, wherein each physiological signal of the set of physiological signals indicates one or more physiological parameters of the patient; processing, by the processing circuitry, the set of physiological signals to determine one or more third characteristics corresponding to a respective third period of time before the point in time; processing, by the processing circuitry, the set of physiological signals to determine one or more fourth characteristics corresponding to a respective fourth period of time following the point in time; and determining, by the processing circuitry, the severity of the impact event based on the first plurality of characteristics, the second plurality of characteristics, the one or more third characteristics, and the one or more fourth characteristics.

[0232] Example 20. The method of examples 19, further comprising generating, by one or more sensors, the set of physiological signals.

[0233] Example 21. The method of example 19 or example 20, wherein the one or more physiological parameters comprise at least one of heart rate, heart rate variability, R-wave amplitude, respiration rate, respiration effort, blood pressure, temperature, oxygen saturation, fluid status, arterial blood pulsation amplitude, pulse transit time, or pulse wave velocity.

[0234] Example 22. The method of any of examples 19-21, further comprising: processing the set of physiological signals to determine one or more baseline characteristics each corresponding to a respective fifth period of time before the point in time, wherein each respective baseline characteristic defines a respective range; and determining the severity of the impact event based on the plurality of first characteristics, the plurality of second characteristics, the one or more baseline characteristics, and the one or more fourth characteristics.

[0235] Example 23. A non-transitory computer-readable storage medium comprising instructions, that when executed by processing circuitry of a device, cause the device to: detect, based on one or more accelerometer signals, an impact event corresponding to a patient, the impact event occurring at a point in time; process the one or more accelerometer signals to determine a plurality of first characteristics corresponding to a first period of time before the point in time corresponding to the impact event, wherein the plurality of first characteristics comprise at least one of a maximumacceleration, a pre-impact morphology, a pre-impact posture, a pre-impact rate of acceleration change, or a pre-impact duration of acceleration change; process the one or more accelerometer signals to determine a plurality of second characteristics corresponding to a second period of time following the point in time corresponding to the impact event, the plurality of second characteristics comprising at least one of a postimpact morphology, a post-impact posture, or a post impact activity level; determine a severity of the impact event based on the first characteristics and the second characteristics; and determine whether to output an alert based on the severity of the impact event.

[0236] Example 24. A method comprising detecting an impact event corresponding to a patient, the impact event occurring at a point in time, processing the one or more accelerometer signals to determine a plurality of first characteristics corresponding to a first period of time before the point in time corresponding to the impact event, wherein the plurality of first characteristics comprise at least one of a maximum acceleration, a pre-impact morphology, a pre-impact posture, a pre-impact rate of acceleration change, or a pre-impact duration of acceleration change, processing the one or more accelerometer signals to determine a plurality of second characteristics corresponding to a second period of time following the point in time corresponding to the impact event, the plurality of second characteristics comprising at least one of a postimpact morphology, a post-impact posture, or a post impact activity level, determining a severity of the impact event based on the first characteristics and the second characteristics; and determining whether to output an alert based on the severity of the impact event.

[0237] Example 25. The method of Example 24, wherein the method is performed by the medical device system of any one of Examples 1-16.

[0238] Example 26. A computer-readable storage medium comprising instructions, that when executed by processing circuitry of a system, cause the system to perform the method of Examples 17-22, 24 or 25.

[0239] Various aspects of the disclosure have been described. These and other aspects are within the scope of the following claims.

Claims

WHAT IS CLAIMED IS:

1. A medical device system comprising: a medical device comprising one or more sensors, wherein the medical device is configured to generate, using an accelerometer of the one or more sensors, one or more accelerometer signals indicating a motion of a patient; one or more memories; and processing circuitry in communication with the one or more memories, wherein the processing circuitry is configured to: detect, based on the one or more accelerometer signals, an impact event corresponding to the patient, the impact event occurring at a point in time; process the one or more accelerometer signals to determine a plurality of first characteristics corresponding to a first period of time before the point in time corresponding to the impact event, wherein the plurality of first characteristics comprise at least one of a maximum acceleration, a pre-impact morphology, a preimpact posture, a pre-impact rate of acceleration change, or a pre-impact duration of acceleration change; process the one or more accelerometer signals to determine a plurality of second characteristics corresponding to a second period of time following the point in time corresponding to the impact event, the plurality of second characteristics comprising at least one of a post-impact morphology, a post-impact posture, or a post impact activity level; determine a severity of the impact event based on the first characteristics and the second characteristics; and determine whether to output an alert based on the severity of the impact event.

2. The medical device system of claim 1, wherein as part of detecting the impact event, the processing circuitry is configured to compare at least of an amplitude of an accelerometer signal or a change in posture angle from a first posture angle at a first time to a second posture angle at a second time to a threshold, wherein the threshold includes an amplitude component and a change in posture angle component.

3. The medical device system of claim 1 or claim 2, wherein the medical device is further configured to generate, using the one or more sensors, a set of physiological signals, wherein each physiological signal of the set of physiological signals indicates one or more physiological parameters of the patient, and wherein the processing circuitry is further configured to: process the set of physiological signals to determine one or more third characteristics each corresponding to a respective third period of time before the point in time; process the set of physiological signals to determine one or more fourth characteristics each corresponding to a respective fourth period of time following the point in time; and determine the severity of the impact event based on the plurality of first characteristics, the plurality of second characteristics, the one or more third characteristics, and the one or more fourth characteristics.

4. The medical device system of claim 3, wherein the one or more physiological parameters comprise at least one of heart rate, heart rate variability, R-wave amplitude, respiration rate, respiration effort, blood pressure, temperature, oxygen saturation, fluid status, arterial blood pulsation amplitude, pulse transit time, or pulse wave velocity.

5. The medical device system of claim 3 or claim 4, wherein the processing circuitry is further configured to: process the set of physiological signals to determine one or more baseline characteristics each corresponding to a respective fifth period of time before the point in time, wherein each respective baseline characteristic defines a respective range; and determine the severity of the impact event based on the plurality of first characteristics, the plurality of second characteristics, the one or more baseline characteristics, and the one or more fourth characteristics.

6. The medical device system of any of claims 3-5, wherein each of the one or morefourth characteristics has a corresponding acceptable clinical range, and wherein the processing circuitry is further configured to determine the severity of the impact event based on the plurality of first characteristics, the plurality of second characteristics, the one or more fourth characteristics, and each acceptable clinical range.

7. The medical device system of any of claims 1-6, wherein the processing circuitry is further configured to estimate a body part that first contacted a surface that was impacted during the impact event based on the pre-impact posture and the post-impact posture, and wherein the determined severity of the impact event is at least partially based on the estimate of the body part that first contacted the surface.

8. The medical device system of any of claims 1-7, wherein the processing circuitry is further configured to determine a trajectory of the patient prior to the impact event based on the pre-impact posture and the post-impact posture, and wherein the determined severity of the impact event is at least partially based on the trajectory.

9. The medical device system of any of claims 1-8, further comprising a machine learning algorithm, wherein the processing circuitry is configured to execute the machine learning algorithm to determine the severity of the impact event.

10. The medical device system of any of claims 1-9, wherein the impact event is a result of a fall and wherein the processing circuitry is further configured to estimate a reason for the fall based on at least one of the one or more accelerometer signals, the one or more physiological parameters, an electrocardiogram sensed by one or more of the one or more sensors, or an identification of a medication of the patient.

11. The medical device system of claim 10, wherein the reason for the fall comprises at least one of a cardiac reason, or a non-cardiac reason.

12. The medical device system of any of claims 1-12, wherein the accelerometer has a measurement range of at least plus or minus 10G and at least 12-bit resolution and wherein a sampling rate of the accelerometer is at least 1000 Hz.

13. A method compri sing : detecting an impact event corresponding to a patient, the impact event occurring at a point in time; processing the one or more accelerometer signals to determine a plurality of first characteristics corresponding to a first period of time before the point in time corresponding to the impact event, wherein the plurality of first characteristics comprise at least one of a maximum acceleration, a pre-impact morphology, a pre-impact posture, a pre-impact rate of acceleration change, or a pre-impact duration of acceleration change; processing the one or more accelerometer signals to determine a plurality of second characteristics corresponding to a second period of time following the point in time corresponding to the impact event, the plurality of second characteristics comprising at least one of a post-impact morphology, a post-impact posture, or a post impact activity level; determining a severity of the impact event based on the first characteristics and the second characteristics; and determining whether to output an alert based on the severity of the impact event.

14. The method of claim 13, wherein the method is performed by the medical device system of any one of claims 1-12.

15. A computer-readable storage medium comprising instructions, that when executed by processing circuitry of a system, cause the system to perform the method of any one of claims 13 or 14.

Images