Methods for determining recovery from a concussion

Abstract

A system and method for detecting or confirming recovery from a concussion includes a headset or a head-mountable device that is placed on the head of a patient who has suffered a concussion. The device has an accelerometer for detecting and measuring the natural movement of the head due to blood flow to the brain. After a specified number of days following the concussion, when the patient returns to physical activity, the accelerometer data from the natural movement is analyzed to determine whether the concussion persists. Preferably, a whole-motion accelerometer is also included in the head-mountable device to record the overall movement of the head so as to correlate any increase in data indicating a concussion with the return to physical activity.

Description

[Technical Field]

[0001] The present invention relates to a method for determining when a subject has suffered a concussion injury and when the subject has recovered from the injury and can safely return to activity. This is evaluated using a headset system, head pulse signals, and one or more algorithms. The head pulse signals are obtained from brain movement, are non-subjective, do not require cooperation from the subject, and concussion injury detection and recovery from concussion injury can be performed on subjects while they are sleeping, unconscious, or awake. The non-invasive headset can be installed for a single recording, multiple recordings, or worn continuously.

[0002] Another aspect of the present invention is to provide a head movement monitoring device for providing input to an algorithm that determines whether a subject has recovered from a concussion injury.

[0003] Another aspect of the present invention is to ensure complete remote monitoring of concussion recovery. [Background technology]

[0004] The terms “concussion” and “mild traumatic brain injury (mTBI)” are used interchangeably in this document. A concussion is clinically defined and diagnosed by sports medicine trained professionals, including emergency physicians, team doctors, coaches, team trainers, and physical therapists. The term “concussion injury” refers to the pathological changes in the brain following a concussion. “Subconcussion injury” is brain injury caused by external force applied to the brain that does not result in the symptoms of a concussion. In this specification, the unit mg represents milligrams or 0.001g, where g is the acceleration due to gravity, 1g = 9.8m / s². 2 That is the case.

[0005] Traumatic brain injury (TBI) can lead to serious illness and death, with an annual global incidence exceeding 60 million cases. One-third of injuries are sports-related and are more common in young adults. The majority of TBIs are professionally classified as "mild" using current diagnostic criteria. Concussion is a subcategory of TBI, and its symptoms include headache, vestibular dysfunction, visual disturbances, cognitive symptoms, mood swings, and sleep disturbances. A second concussion during recovery can be neurologically detrimental and, rarely, fatal ("second impact syndrome"). Recurrent concussions or asymptomatic subconcussions are also associated with delayed neurobehavioral dysfunction, as well as subsequent neuropathological findings of chronic traumatic encephalopathy (CTE) and other neurodegenerative conditions. Risk factors for recurrent head injuries and CTE are well documented in the literature and include collision sports. To mitigate these risks, most organized sports teams utilize concussion protocols that prohibit return to competition before clinically determined recovery. Objectively identifying concussion injury, asymptomatic subconcussion, and eventual recovery is crucial for mitigating short-term and long-term functional impairment and remains an unmet need.

[0006] Digital biomarkers derived from cranial accelerometry (CA) exist, which apply the principles of the cardiography test, first recognized in the 19th century. Cardiography tests measure the whole-body forces generated by cardiac contraction. Similarly, the movement of the brain due to the cardiac cycle, and the resulting inflow and outflow of blood into the brain, can be measured on the scalp using a highly sensitive accelerometer or other sensor. In this specification, this is referred to as the "head pulse," and it has been previously shown that frequency-domain analysis of the head pulse can aid in the diagnosis of concussion injury and recovery from concussion injury. Abnormalities in the head pulse, when processed through appropriate algorithms, can lead to the diagnosis of a number of neurological conditions, particularly moderate to severe TBI, large vessel occlusion (LVO) stroke, intracranial hemorrhage (ICH), seizures, migraines, cerebral vasospasm, and the onset of edema after cardiac arrest.

[0007] Analysis of head pulse signals can indicate when a person suffered a concussion injury. By processing head pulse signals through a concussion algorithm, various "biometric parameters" that correlate with a recent concussion diagnosis are generated. The inventors use the term "biometric concussion" to indicate when a head pulse recording from an individual significantly exceeds values ​​obtained from individuals who have not experienced a normal concussion, and "biometric onset" to indicate when the initial recording meets the definition of a biometric concussion. This biometric concussion signal appears after a clinically diagnosed concussion and returns to normal after sufficient time has passed and the subject has recovered. The time it takes to return to normal varies depending on the severity of the concussion and the individual's recovery rate.

[0008] The Zurich Protocol was developed to determine when a person who has suffered a concussion can return to activity. This protocol uses a symptom survey of 22 physiological indicators (Sport Concussion Assessment Tool 2 or SCAT2), such as difficulty sleeping, headache, balance, or difficulty concentrating on tasks, to determine if a person has suffered a concussion. If the symptom survey scores return to normal (or baseline) levels, the person is permitted to begin limited physical exercise. If the symptom survey scores rise again, the person is instructed to rest again without physical exercise. Physical exercise is resumed when the person's symptom survey scores return to baseline again. Further, more vigorous physical exercise is permitted if physical exercise does not cause an increase in the symptom survey scores. This continues until vigorous physical exercise no longer causes an increase in the symptom survey scores.

[0009] The Zurich Protocol assumes that the brain is healed if physical activity does not trigger symptoms. The Zurich Protocol mistakenly assumes that physical activity that does not induce symptoms is not harmful. Furthermore, the Zurich Protocol requires self-reporting by the subject, and the protocol can be manipulated if a subject who wishes to return to activity fails to report symptoms despite having them. Similarly, a subject may be permitted to feign concussion symptoms to avoid participating in activities.

[0010] The present invention provides a method for determining when it is safe for a subject to return to activity after a concussion injury, while avoiding the possibility of the subject mistakenly self-reporting symptoms. The present invention may also provide a method for determining whether a subject has suffered a concussion injury. [Overview of the project] [Means for solving the problem]

[0011] The disclosed method of the present invention non-invasively measures head pulse, a physiological parameter that, when processed through a concussion algorithm, can detect when a subject suffered a concussion injury and when they recovered from that injury. Head pulse is a physiological criterion for concussion that does not require the subject's participation. Subjects while awake, asleep, or unconscious can be diagnosed with a concussion and monitored for recovery in a way that is not possible with other measures or methods that depend on the subject's participation.

[0012] Further disclosed, the present invention uses a whole-body motion detector in contact with the head to continuously monitor how much force is transmitted to the subject's brain during normal activity. The motion detection data is processed by a concussion recovery algorithm to determine when it is safe to return to activity.

[0013] Analysis of head pulse biometric parameters indicates a concussion if any of the calculated parameters exceed the concussion threshold. Head pulse concussion biometrics indicates recovery from concussion injury in two stages. In stage 1, recovery is indicated if any of the concussion biometric parameters fall below the concussion threshold and continue to decline over time. In stage 2, recovery is further indicated if the concussion biometric parameters do not correlate with the external force applied to the head. Recovery from a concussion is not observed if movement to the brain, i.e., physical activity measured directly at the head, increases the concussion biometric parameters, even if the recovery in stage 1 is met.

[0014] Concussion biometric parameters can be used as a binary test as described above, and they can also be used to measure increases in concussion biometrics and recovery from concussion, provided that these biometrics fall within the normal range. Even if concussion biometric parameters are below the concussion reference threshold, they will continue to decrease until they reach a stable baseline. Normal subjects and subjects who have recovered from concussion injury will have different baselines, although they will be below the concussion biometric threshold. During recovery from concussion injury, concussion biometric parameters will smoothly return to the baseline. If concussion biometric parameters remain stable over a long period, the subject is recovering from their injury and can return to active physical activity.

[0015] Clinical trials may be conducted to determine the diagnostic sensitivity of the device.

[0016] A group of subjects (control subjects) who have not received a concussion diagnosis for a defined period (e.g., one year) undergo head pulse recording and subsequent algorithmic analysis. Concussion biometric parameters are calculated for this cohort, and the average and variance (standard deviation) of these parameters are derived. Control subjects whose head pulse exceeds a predetermined statistical value (e.g., 2 standard deviations above the mean of the controls, but may be between 1 and 4 standard deviations) of the concussion biometric parameters are considered false positives, while those below this threshold are considered true negatives. The false positive rate is expected to be predictable from the definition of the threshold. For example, assuming a normal distribution of values ​​and using a threshold of 2 standard deviations, an estimated false positive rate of 2.5% is expected, with higher or lower false positive rates applicable to thresholds between 1 and 4 standard deviations.

[0017] A group of subjects clinically diagnosed with concussion undergoes regular head pulse recordings for a specified duration (e.g., one month, which may be two weeks or up to six months) after the concussion event. Subjects whose head pulse exceeds the concussion biometric parameter by a predetermined statistical value (e.g., two standard deviations above the mean of the control) are registered as having a concussion. Such subjects are true positives. Subjects who do not exceed this threshold during the defined observation period (e.g., two weeks, which may be from one week to six months) are considered false negatives. To determine whether head pulse concussion biometrics is actually a useful measure of concussion recovery, the following clinical trials can be conducted.

[0018] Normal subjects without concussion are measured during exercise.

[0019] Using the control group as described above, when head pulse concussion biometrics does not indicate a concussion, this is scored as a true negative.

[0020] When head pulse concussion biometrics indicates a concussion, this is scored as a false positive.

[0021] Subjects with concussion are measured during exercise.

[0022] When head pulse concussion biometrics indicates a concussion, this is scored as a true positive.

[0023] When the head pulse concussion sign does not indicate a concussion, this is scored as a false negative.

[0024] In the above protocol, the determination of when a subject had a concussion is based on a clinical assessment by a health management expert.

[0025] A further problem with the Zurich Protocol is the lack of objective measurement criteria for body movement and its impact on the brain. Physical activities such as raising the hand and jumping or jogging are performed at different intensities, thus making it difficult to determine whether the brain has been shaken severely enough to trigger a recurrence of concussion symptoms.

[0026] The present invention applies a direct method for detecting concussion and a much more sensitive and objective measurement criterion for recovery from concussion. A device that includes a low-level accelerometer and a high-level accelerometer in a single unit, or individual low-level and high-level accelerometers attached together to the subject's head, can record the total amount of movement the brain receives from activities, measure the head pulse when the subject is at rest, and calculate a concussion sign score. The correlation between activities and concussion biometric measurement parameters can be used to determine whether the subject can return to normal and resume activities. Normal means that active activities do not induce an increase in the head pulse concussion biometric measurement parameters.

[0027] When the accelerometer has a sufficient operating range for both low levels (about 98 mm / s 2 ~392 mm / s 2 (about 10 mg~40 mg)) and high levels (4.9 m / s 2 ~980 m / s 2 (0.5 g~100 g)), the accelerometer can be one device. The 24-bit ADC has a resolution of 23 bits for accelerations on one side (+980 m / s 2 (+100 g)), and the least significant bit (LSB) is 117.6 mm / s 2 (12 mg). Systems with these parameters are readily available.

[0028] Dual accelerometer systems, either integrated or combined separately, can be manufactured to be very small and can be fitted into eyeglasses, in the ear canal, in a headset, or attached to the head with adhesive. The devices can be powered by primary or secondary batteries, solar cells, or other power sources. The devices can be designed to remain dormant for most of the time to conserve power and can be powered at predetermined intervals or by detecting motion. Application-Specific Integrated Circuits (ASICs) can be designed and manufactured to reduce cost and size. Advanced power management can be used to maximize wear time between recharges.

[0029] The diagnosis described above is made based on the time course of the paired parameters of physical activity and head pulse concussion biometrics. If the subject who suffered a concussion is back to normal, the head pulse concussion biometric parameters do not increase with physical activity. The subject can safely return to vigorous physical activity, contact sports, or military duties.

[0030] In a further embodiment, the present invention helps ensure that self-managed recordings of cranial acceleration and activity monitoring are reliably recorded from a subject identified as having suffered a concussion.

[0031] In most cases, individuals who have suffered a concussion wish to be closely monitored for recovery and are aware that resuming vigorous activity before the brain has fully healed could result in serious long-term consequences. In other cases, despite still having high symptom assessment scores or high concussion biometric parameters from the head pulse system, there is a strong desire or economic motivation to return to activity.

[0032] Furthermore, there is a strong desire to return to activity even if the symptom assessment score returns to normal and the cranial acceleration concussion biometric parameters return to baseline, but these may increase upon resumption of activity. An increase in either the symptom assessment score or the concussion biometric parameters would restrict the individual's return to activity, particularly military activity or contact sports.

[0033] Having head pulse recordings taken by a medical professional ensures that the recordings belong to someone recovering from a concussion. However, it is not practical for patients to be constantly under the supervision of a medical professional for repeated recordings in order to determine recovery from a concussion, which can take several weeks or more.

[0034] This invention ensures that the time course of cranial acceleration measurement data and head pulse concussion biometric measurements is recorded entirely by the person who suffered the concussion, and not by a substitute.

[0035] The head pulse concussion sign does not require baseline data of the subject and only rises if the subject experiences a concussion. The head pulse is recorded by using an accelerometer or other sensor that comes into contact with the subject's head, preferably the temple area. A smartphone may be used to control the headset, or the headset may be self-contained, with all concussion biometric parameters calculated within the headset itself.

[0036] In some cases, individuals may deliberately fabricate records by applying a headset to individuals who have not suffered a concussion, thereby instructing medical professionals or other record-keepers that the individual is fit to continue vigorous physical activity, contact sports, or military activities. In other cases, individuals may wish to be diagnosed with a concussion when they have not actually suffered one.

[0037] The present invention includes several methods for preventing any of these erroneous diagnoses from being made.

[0038] A medical professional registers the individual who initially suffered the concussion and takes the initial recording. If a smartphone is part of the headset system, facial recognition may be implemented to record the individual's face. If additional head pulse recordings are required over the recovery period, the individual must log in to the system using their face for ID. To further ensure that the recording is being performed by the concussed individual, the R-peak signal of the ECG may be correlated with the photoplethysmography signal, which is obtained from well-confirmed skin redness due to blood flow to the brain and face detected by the smartphone's camera. The head pulse app can utilize the R-peak timing transmitted to the headset by the headset. This timing can be transmitted wirelessly, or the camera may detect a light flash in the headset correlated with the ECG R-peak. The correlation of the photoplethysmography signal with the ECG R-peak, combined with facial recognition, ensures that the recording is being performed by the concussed individual.

[0039] Another method of the present invention to ensure complete head pulse recording is to use an average single-heartbeat time-domain head pulse recording. The head pulse signal has been shown to be unique to each subject. This average signal remains stable over weeks, months, and years and can therefore be used to verify that the recorded signal is from the same subject initially enrolled in the concussion recovery protocol.

[0040] The device is used by healthcare professionals to assess patients after a clinical diagnosis of concussion. These clinical assessments are often performed 7–10 days after traumatic brain injury (TBI), and patients may have a range of symptoms from none to significant. The assessing healthcare professional applies the device to the patient's head, and the device calculates concussion biometrics and compares these results to control data to determine whether a concussion is present or not. The test may be repeated at nursing time if desired. This result helps differentiate between post-traumatic migraine and concussion. The symptoms of migraine and concussion overlap (nausea, photophobia, dizziness, vomiting, headache), making it often difficult to determine which category a patient falls into. Individuals prone to migraines (those with a history of migraines or those exhibiting migraine-like features without headache) tend to overreact to brain injury, particularly because migraine symptoms can be significantly aggravated even by mild TBI. Clinicians treating such patients need to understand whether the patient has migraine, concussion, or a combination of both. The device results help clinicians determine whether migraine treatment or concussion-specific treatment is best for the patient. For subjects with an apparent concussion event, with normal readings or readings indicating a concussion is currently occurring, the device can be prepared for recording during walking to test whether the subject is still overreacting to physical activity, and if so, intervention can be carried out by restricting activity until the device indicates recovery. For subjects where the device does not indicate a concussion, medication or other therapies aimed at treating migraine may be administered along with permission to resume physical activity. Subjects with migraine may also be monitored during walking to check for recurrence of concussions.

[0041] The device is used to assess the combat readiness of military personnel. Combat readiness encompasses several areas, including a history of recent head injuries. For military combatants returning to service, the device assessment helps determine their combat readiness.

[0042] The device is used to identify the cause of a change in mental state in patients who are unable to provide their medical history and for whom there is no third party to explain the events leading up to the change in mental state. Specifically, worldwide, a third party encounters an individual who is unresponsive to calls or in a comatose state and requests an assessment from advanced emergency services. Before hospital arrival, healthcare providers will arrive quickly, assess the patient, and pay attention to the patient's level of consciousness and vital signs (blood pressure, oxygen saturation), and spontaneous respiration. The priority is to medically support the patient until they can receive advanced care at the hospital. After hospital arrival, emergency services will consider many possible causes, ranging from trauma, stroke, seizure, drug toxicity, shock, and metabolic processes. The device has already proven to detect major vascular strokes and massive intracerebral hemorrhages sufficient to cause loss of consciousness, and the application of the device to facilitate stroke triage is the basis for concurrently pending application 17 / 423,281. Shock is ruled out if the patient has normal blood pressure. Drug toxicity is revealed by uremic toxicity tests, and most metabolic processes are revealed by simple basal metabolic profile serum tests. Head trauma may not be detected by brain CT scans (which are usually performed in all cases), and may not show changes even in people with significant traumatic brain injury. The device can reveal the cause of changes in mental state as TBI, while the additional tests mentioned above can reveal whether multiple causes are coexisting. For example, alcohol intoxication with traumatic brain injury is often present in patients brought to the emergency room. Knowing that TBI is a factor or contributes to changes in mental state can help in the clinical management of the patient, and thus maximize recovery. The device is useful for diagnosing TBI in resource-limited environments where CT scans are not readily available, and its portability allows it to be used in the pre-hospital environment to assist in triage decisions. [Brief explanation of the drawing]

[0043] [Figure 1]This is a schematic elevation view indicating the head of an object to which the device of the present invention, which has related analysis and monitoring equipment, is fixed. [Figure 2] This figure shows an example of the headset of the present invention. [Figure 3] This is a time-domain graph of head pulse signals in the X, Y, and Z directions, representing subjects who did not experience a concussion. [Figure 4] This graph shows the FFT of a head pulse signal, illustrating the signal in the frequency domain. It represents a subject immediately after a concussion. [Figure 5] This is a similar frequency domain graph, showing the same subject as in Figure 4, but three days later. [Figure 6] This is a frequency domain plot showing subjects who did not suffer a concussion. [Figure 7] This is a plot of the value identified as R1 for subjects who suffered a concussion, with R1 plotted against time in days. [Figure 8] This is a series of graphs comparing symptom scores or NSI for subjects who suffered a concussion. The NSI graphs are compared to R1 and R2 values ​​over a period of approximately one month, and subjects returned to strenuous athletic activity within a certain number of days. [Figure 9] Figure 8 shows a series of graphs comparing symptom scores or NSI for subjects who suffered concussions, different from those in Figure 8. The NSI graphs are compared to R1 and R2 values ​​over a period of approximately one month, and the subjects returned to strenuous athletic activity within a certain number of days. [Figure 10] This is a similar series of graphs showing another subject who suffered a concussion and returned to competition early during the concussion period. [Figure 11] This graph shows the percentage of the largest control group that exhibited concussion based on biometric measurements over a one-month period. [Figure 12] This graph shows the percentage of concussion-free subjects based on NSI values. The lower line represents the percentage of subjects who achieved a concussion-free state based on biometric measurements over a one-month period. [Modes for carrying out the invention]

[0044] Figure 1 schematically shows a system 10 according to the present invention, with the subject's head shown in profile at 12 and the subject's brain shown by a dashed line at 14. The system 10 includes a headset 16 that is worn on the head like a hairband. The headset 16 has an accelerometer 18 that is firmly held in the headset and positioned near the subject's temple for determining a head pulse signal generated by a series of heartbeats schematically indicated at 20, 22, and 24, due to the contraction of the user's heart indicated at 26. The pulsation of the heartbeat in the head causes slight physical movement of the head, i.e., the skull, in the orthogonal directions of X, Y, and Z, which are longitudinal, anterior-posterior, and lateral. The acceleration is schematically indicated by arrows 28 in the drawing. Acceleration in one, two, or all of these directions can be sensed by the accelerometer 18, and a multi-axis accelerometer is preferably provided (however, a single-axis accelerometer may be used). Importantly, acceleration data can be acquired at a single location on the head.

[0045] The headset is preferably wireless and therefore includes a battery as indicated in 30, however, a non-wireless form of the headset may be provided and capable of receiving power through a cable. The wireless device includes a wireless data transmitter or interface 32 that wirelessly transmits data to the analyzer 34, which may, for example, be included in an app on a nearby smartphone. Alternatively, the analyzer 34 may be contained within the headset, and the processed data may be transmitted to a smartphone by cable or wirelessly for instructions on analysis and results. In another embodiment, the data may be stored in the memory storage of the headset itself and retrieved in a timely manner. The data may be stored in removable memory such as a card, cartridge, or memory stick.

[0046] The drawings also indicate that inspection or clinical data 36 obtained simultaneously through other observations or measurements are supplied to the analyzer 34 for correlation with the head pulse signal, and both are typically converted to the frequency domain.

[0047] As part of the system 10 of the present invention, as indicated by 38, a high-level or overall motion accelerometer is also included in the headset device. This overall motion accelerometer typically has an acceleration of 4.9 m / s 2 ~196 m / s 2 (0.5 g to 20 g) and has the function of sensing and recording high-level motion such as that caused by violent body movements. This information, which is supplied to the high-level analyzer 40 and also transmitted to the analyzer 34, is used to correlate the activity of the subject and the events of that activity with the analyzed head pulse signal for the purposes discussed above, i.e., to determine whether the return to play (RTP) has caused an indication of a concussion recurrence based on biometric measurements. The high-level accelerometer 38 can be placed anywhere in the headset. Alternatively, as discussed above, the high-level accelerometer 38 can be on an individual device such as a FitBit device worn on the wrist or arm. Further, it can be combined with the high-sensitivity accelerometer 18 if a wide-range accelerometer is used. In any case, the head pulse signal and the high-level overall acceleration signal are separate streams of data from the headset.

[0048] Figure 2 shows a head pulse headset device 16 in a preferred embodiment. Two circuit boards are mounted on a headset resembling a plastic headband and interconnected by a ribbon cable extending around the band. A protective cap 42 is shown on one side, but for clarity, the cap is removed on the right side. Each circuit board contains a highly sensitive 3-axis accelerometer, sampled at 250 / second in this embodiment and stored on a microSD card that slides into the unit. The device has a Bluetooth® interface to a custom smartphone app indicated in Figure 3, which also includes NSI (Neuro-behavioral System Inventory) queries. In one embodiment, the subject is required to answer all questions presented by the app in each biometric recording at head pulse. This system is used in subjects in the study reported below. Instructions are shown to the subject in the leftmost panel, Figure 3A. Responses to the NSI questions were reported from cells using a 5-point Likert scale ranging from 0 to 4, where 0 indicated "none" or no system, and 4 indicated very severe symptoms for each question. The overall symptom severity was calculated as the sum of the values ​​provided by subjects, with the probability of a score ranging from 0 to 88.

[0049] The head-mounted device can take various forms. For example, it could simply be a patch (e.g., an adhesive patch) that includes both an accelerometer (or at least a high-sensitivity accelerometer) and Bluetooth or other wireless technology, along with a small battery. The data can be transmitted wirelessly to the cloud for processing, or the processing can be performed on a smartphone or on the device itself. The data can also be transmitted directly from the head-mounted device, or from a smartphone or other device, to a medical professional to assess the patient's recovery, or to provide a record of recovery by correlating high-g accelerometer data with head pulse and low-g accelerometer data.

[0050] Concussion study in Australian athletes This is a prospective cohort study. Participants were enrolled in feasibility assessment (A1) and validation (A2) phases. A1 recruitment took place between August 5, 2021 and September 10, 2021. A2 recruitment took place between May 5, 2022 and September 3, 2022. The study evaluated a longitudinal cohort of concussed and control athletes in amateur Australian Rules Football (ARF) to investigate the time course of changes in head pulse signals observed in athletes who had recently suffered a concussion (Australian University of California San Francisco concussion study in athletes, AUSSIE-1, A1). The model was subsequently validated in a second cohort (AUSSIE-2, A2). ARF athletes did not wear pads or head protection. Concussions are a common complication. Recruitment took place within the Adelaide Football League, the top amateur Australian Rules Football league in South Australia, Australia. Adult male or female athletes participated. Controls were concussed. All participants gave informed consent at the start of the season. Head pulse measurements were obtained using a laboratory-developed headset that utilizes cranial acceleration measurement to transmit head force (head pulse) generated by cardiac contraction. Participants were followed and repeatedly recorded for one month, completing a Neurological Symptom Inventory (NSI) at each recording.

[0051] Head pulse waveform analysis included frequency conversion and subsequent evaluation using a pre-specified algorithm. The calculated waveform patterns of athletes who suffered concussions were compared to controls using Z-scores. Head pulse concussions were characterized as Z-scores > 2 (2 standard deviations from the control head pulse Z-score). Sensitivity, onset time, and duration of head pulse concussions were investigated.

[0052] The study involved 100 participants, including 58 controls and 42 participants who experienced concussions, resulting in a total of 44 concussions. Head pulse biometric parameters were abnormal in 82% of the injured participants (50% by day 2 and 90% by day 14). Head pulse biometrics remained elevated for 12 days longer than the average reported symptom. Exacerbations of head pulse abnormalities occurred in participants who returned to competition or engaged in unsupervised physical activity.

[0053] The results indicate that continuous head pulse measurement has diagnostic value for concussion and recovery, and can identify changes due to potential subconcussion in return to competition within one month. This objective measurement can support clinical decision-making.

[0054] The study was designed to follow specific procedures, including concussion diagnosis and RTP assessment. Academic study coordinators attended games and monitored players who suffered concussions. Attempts were made to record concussions early on. All records except the initial one were taken within one hour of the concussion. Coordinators visited participants' homes every 1–3 days for one month after the injury to obtain additional records. The A2 cohort was encouraged to wear wristband accelerometers to record physical activity after the injury (FitBit, Inspire2, Alphabet, San Francisco, USA).

[0055] Recordings were obtained using a battery-powered, non-invasive headset device as described above, consisting of a highly sensitive accelerometer mounted on the headset. The headset was positioned on the subject's head in the coronal plane (Figures 1 and 2). The head pulse was 98–147 mm / s². 2 (10-15 mg: g, gravitational units, 9.8 m / s) 2It consists of the force on the head resulting from cardiac contraction within the range of ). In A1, bilateral triaxial accelerometers were placed in front of the ears on the temporal bone. In A2, a commercially available device with a unilateral triaxial accelerometer (MindRhythm, Inc., Cupertino, CA) was used.

[0056] The device transmitted an electrocardiogram (ECG) using a standard three-lead system to provide heart rate information. The device connected to an iPhone® using Bluetooth and custom software. Recordings lasted 180 seconds in A1 and 90 seconds in A2, and were obtained in a seated position. Participants were asked to keep their heads still and refrain from speaking or chewing. The A2 headset (MindSafe, MindRhythm, Inc, Cupertino, CA) was designed to provide feedback to participants if movement was detected. Participants completed a Neurobehavioral Symptom Inventory (NSI)23,24, adjusted on the iPhone, during each recording.

[0057] The data files were sent to custom software (MATLAB, MathWorks, Natick, MA) to analyze the frequency data as previously described. Time-domain accelerometer signals from the right-side accelerometer at 45 heartbeats were converted to the frequency domain using Fourier transform.

[0058] The average heart rate during recording was derived from the R-wave analysis of the ECG, obtaining the fundamental and harmonics of the heart rate. Frequency conversion was performed by sampling the fundamental wave and the harmonics 2-9 that are continuous with it. The concussion biometric R1 was calculated as the ratio of the average values ​​of the 5th and 6th harmonics to the maximum values ​​of harmonics 1-3 (R1), and the concussion biometric R2 was calculated as the average values ​​of the 7th and 8th harmonics divided by the maximum values ​​of harmonics 1-3, and R1 and R2 values ​​for the control were calculated.

[0059] The R1 and R2 values ​​of concussion biometrics were analyzed as Z-scores above or below the mean control value. Higher Z-scores for concussion biometrics R1 and R2 represent higher head pulse frequency deviations after TBI. Any record exceeding two standard deviations (SD) of the control was defined as significant, and the earliest value exceeding this threshold was defined as the onset time of abnormal concussion biometrics.

[0060] The R1 and R2 values ​​of concussion biometrics, as well as the corresponding NSI scores, were plotted over time. Kaplan-Meier curves representing the resolution of abnormal biometrics were calculated to show diagnostic sensitivity compared to controls and to indicate the delay time from the concussion event to the first recording of an abnormal device. To record recovery from concussion, the duration from the onset of concussion by biometrics to a zero NSI score was aggregated, and subjects were considered clinically unrecovered if they did not return to a zero NSI score during the recording period. For subjects without a threshold biometric onset time, the biometric duration was defined as zero. For subjects with a threshold biometric onset time, the recovery period was defined as the duration from clinical concussion to the time when all ratios fell below 1 SD from the mean. The selection of 1 SD allowed for objective quantification of biometric parameters based on the observation that concussion biometrics R1 and R2 characteristically decreased over time.

[0061] The study identified characteristic changes in head pulse after acute injury and marked asymptomatic head pulse abnormalities in individuals with RTP. Of 762 athlete volunteers, a total of 100 subjects, including controls (n=58) and subjects with concussions (n=43), underwent head pulse measurements. Of the 44 concussions, one control subject subsequently experienced another concussion (and was therefore counted as two), and one subject with a concussion in group A2 experienced two concussions more than one month apart.

[0062] The subjects had similar years of education. Female athletes had fewer self-reported concussion histories than male athletes, and male athletes had participated in more matches than female athletes (a surrogate indicator of concussion exposure). However, there was no significant difference in the number of matches played between the control group and the concussion-affected group in either sex. Of the subjects who suffered a concussion, 4 (10%) experienced loss of consciousness (LOC), 16 (39%) experienced altered consciousness, and 7 (17%) experienced post-traumatic amnesia (PTA) at the time of injury. Concussions were diagnosed by physicians in 7 (17%), by team / staff / members in 25 (61%), by the subjects themselves in 8 (20%), and the diagnosis was unknown in 1 (2.4%). Concussion diagnoses followed AFL guidelines. None of the participants suffered a concussion within one month of the injury, and most returned to competition within this timeframe.

[0063] In A1, 40 out of 184 records (22%) showed reduced movement. In A2, the improved device (MindSafe device provided by MindRhythm, Inc., Cupertino, CA) provided feedback on excessive body movement, resulting in fewer records being rejected (24 / 276 = 8.7%).

[0064] Figure 3 shows a typical head pulse recording, which is raw time-domain data for each of the three axes from which accelerometer readings are acquired. The first three panels are the head pulse force in mg units (millig on the vertical, anterior-posterior, and lateral axes (i.e., X, Y, and Z axes, respectively)). The bottom recording line in Figure 3 is the ECG.

[0065] Figure 4 shows Fourier transform data from time-domain samples, similar to those in Figure 3, except that it pertains to a subject immediately after concussion. Heart rate harmonics are shown as vertical dotted lines and represented by numbers. The height of the frequency transform at each harmonic is used to calculate the R1 and R2 values ​​of the concussion biometrics. Harmonics 5 and 6 are smaller than the peak values ​​of harmonics 1-3 immediately after concussion, i.e., within a few hours.

[0066] Figure 5 shows the same subject three days later, i.e., three days after the concussion, in the frequency domain. This chart shows that the energy of the 5th and 6th harmonics is significantly higher here than the average of harmonics 1-3, defining a concussion by biometric measurement. The vertical axis of these plots represents the frequency in Hz, i.e., the energy at each frequency. Note that in Figure 5, following the rise in frequency components at the 5th and 6th harmonics, there is a drop to a nearly normal level around the 9th harmonic, and then several rises and falls in harmonics occur up to the 20th harmonic.

[0067] Figure 6 is another plot of energy in the frequency domain, showing subjects without concussions. This is generally similar to the graph in Figure 4 and is typical of subjects without concussions, with the highest energy frequency occurring at approximately the third or fourth harmonic.

[0068] Figure 7 plots the R1 value of concussion biometrics, where the concussion biometric value R1 is the average or mean value of harmonics 5 and 6 divided by the maximum value of harmonics 1-3. The plot shows the concussion biometric R1 value over time on a daily basis. Typically, the sharp rise in the R1 value of concussion biometrics does not occur immediately, but here it begins on day 2 or 3, peaks around day 5, then declines over the next 4 or 5 days, and continues to decline more gradually after day 3 or 24.

[0069] Figures 8 and 9 are a series of graphs comparing NSI scores to biometric data for each of two different subjects. Figure 8 represents a female subject who suffered a concussion during an Australian Rules Football match. The top record line is her NSI score, aligned with each head pulse record over a one-month period. The dashed line on the left vertical axis represents the time of concussion, with the dashed line on May 13 indicating return to play (RTP). The three graph panels below are head pulse records from longitudinal, anterior-posterior, and lateral accelerometers obtained while the subject was resting in a seated position. R1 and R2 values ​​are shown for each record for each of these three axes. The values ​​in the vertical axis are the Z-scores of the R1 and R2 values ​​compared to the control mean, with the horizontal dashed line at level 2 indicating 2 standard deviations above the control mean. The first record that intersects the 2-SD line on any of the three axes is defined as the biometric onset time of concussion, which is typically one or two days later than the actual time of concussion. As was typical in both cohorts (Figures 8 and 9), the R1 and R2 values ​​of concussion biometrics were abnormal on day 2, returned to normal around day 5-7, and then rose and fell again in an "M-shape." This subject remained seated until she returned to competition. As indicated in the graph, she saw an increase in the R1 and R2 values ​​of concussion biometrics on day 13, particularly on the x-axis, and then returned to normal for the rest of the month. This is an example of biometric signals paralleling NSI.

[0070] Figure 9 represents a 21-year-old male with two previous concussions who fell to the ground, stood up, experienced ataxia, and then fell again. His NSI score revealed a minimal self-report system, and his cranial accelerometer records showed a significant rise / fall / rise / fall pattern, as confirmed in the cohort in Figure 8. No concussion symptoms were reported by the NSI, and he returned to competition on day 13, August 28, with no recurrence of symptoms. Despite being asymptomatic, his biometric parameters were elevated by more than two standard deviations (approximately September 1, day 17) and did not return to normal within the measurement period. This is an example of biometric signals being independent of the NSI.

[0071] Figure 10 shows a similar graph comparing NSI to biometrics for a 19-year-old male with a known history of concussions who suffered a concussion during an Australian Rules Football match. The layout of the figure is the same as in Figures 8 and 9, with the same three acceleration axes as in Figures 8 and 9, showing the symptom score, NSI, on the top record line, and concussion biometrics R1 and R2 plotted in the three lower graphs. In Figure 10, the subject's activity level is plotted by date in the bottom graph. In addition to head pulse recording, the subject wore a wrist-mounted accelerometer. Activity in the bottom graph is defined as "fairly active" and "very active" by the device manufacturers (FitBit, Alphabet, San Francisco). The wrist-mounted device recorded significant physical activity in the days leading up to the increase on day 10, before the subject returned to competition on day 14 (as indicated by the top NSI graph). While the other four subjects had good wrist acceleration measurements, all four experienced a deterioration in biometric measurements after the start of physical exercise. Figure 10 shows that, in the case of a 19-year-old male subject, he was not concussion-free at the end of the test duration, which is shown as 28 days. On both the vertical and AP axes, the subject's R1 or R2 values ​​did not return to less than one standard deviation during the period.

[0072] Figure 11 is a graph that includes all subjects registered as having experienced a concussion. It shows the percentage of participants who presented head pulse biometric parameters exceeding 2SD (Z>2) of the control group for more than 30 days. This includes both A1 and A2 cohorts, which have multiple records. Overall, 26 out of 32 (81%) of subjects clinically diagnosed with a concussion had concussion biometric onset within the first 30 days. Of these 26 subjects, 11% were abnormal on the day of concussion (day 0), 50% were abnormal on days 2-3 (lower dashed line), and 90% of subjects in whom a concussion was detected were abnormal on day 14 (upper dashed line). Small circles on the chart indicate censoring of subjects. Censoring means that the subject could not be observed further after that time.

[0073] Figure 12 is another graph showing concussion over time on a daily basis for subjects who recovered using two criteria: the NSI returning to a zero score and the ratio of biometric measurements returning to within 1 SD of the control value. The upper line represents the return of the NSI score to zero, and the lower line represents the return of the ratio of biometric measurements to within 1 SD of the control value. 50% of the subjects had a zero NSI value on day 7, but 50% of the subjects had recovered their ratio of biometric measurements by day 21, 14 days later. This suggests that abnormalities in biometric measurements resolve around 14 days after concussion, according to the conventional definition of concussion recovery.

[0074] The procedures and systems of the present invention significantly improve the assessment of recovery from concussion and indicate that the NSI score is unreliable for determining a safe return to vigorous activity. Premature return to activity (RTP) has been shown to worsen concussion and increase concussion biometrics, indicating that vigorous activity was resumed too early.

[0075] This invention is applicable not only to sports related to concussions, but also to all activities in which head shaking may occur, such as military and industrial activities, automobile accidents, and physical attacks.

[0076] The preferred embodiments described above are intended to illustrate the principles of the present invention, but not to limit its scope. Other embodiments and modifications of these preferred embodiments will be obvious to those skilled in the art and may be made without departing from the spirit and scope of the invention as defined in the following claims.

Claims

1. A method for detecting the recovery status from a concussion in a patient by detecting and measuring the natural movement of the patient's head due to blood flow in the brain, and the resulting movement of the brain tissue, The steps include providing a sensor device that can be attached to the head, which includes a first accelerometer sensor that responds to the natural movement of the head, A step of providing a second accelerometer sensor that can be fixed to the patient, wherein the second accelerometer responds to larger movements associated with shaking, shock, or sudden movements of the patient. The steps include connecting both of the sensors to a computer that receives and records accelerometer signals from the first and second sensors, For a patient in whom a concussion has been confirmed, the steps include attaching the head-mountable sensor device to the patient's head such that the first accelerometer sensor is in contact with the head, in order to record the frequency of natural skull movement in at least one axis at the position of the first sensor in a maximum frequency range of approximately 25 Hz, immediately after the onset of the concussion or a selected number of days thereafter; The steps include attaching the second accelerometer sensor to the patient so that the second sensor detects and records greater acceleration of the patient due to shaking, impact, or sudden movement, The steps include: after the patient wears the first and second accelerometer sensors and engages in physical activity, the computer receives and records signal data from the first and second accelerometer sensors that represent natural movements of the head, as well as movements due to sudden movements, shaking, or impacts; Using the computer, the signal data is compared with data corresponding to non-concussion in order to determine the occurrence and magnitude of any increase in the frequency component of skull movement measured by the first accelerometer sensor in a frequency range exceeding the approximately fourth harmonic of the heartbeat of the patient indicating non-recovery from a concussion, and on the other hand, the patient's physical activity before or at the time of the data measured by the second accelerometer sensor is examined as a factor of any detected increase in the frequency component, If the comparison by the processor indicates non-recovery from a concussion, the steps include restricting the patient's physical activity. Methods that include...

2. The method according to claim 1, further comprising the steps of: continuing to monitor signal data from the first and second sensors if non-recovery from a concussion is indicated; determining whether a concussion is still indicated when the patient returns to physical activity; and repeating this procedure until physical activity no longer indicates non-recovery from a concussion.

3. The method according to claim 1, wherein the second accelerometer sensor is included in the sensor device that can be attached to the head.

4. A method for detecting the recovery status from a concussion in a patient by detecting and measuring the natural movement of the patient's head due to blood flow in the brain, and the resulting movement of the brain tissue, A step comprising including an accelerometer sensor that responds to the natural movement of the head, wherein the sensor is connected to a computer that receives and records acceleration signals from the sensor, For a patient in whom a concussion has been confirmed, the steps include: attaching the sensor device to the patient's head so that the sensor is in contact with the head, in order to record the frequency of natural skull movement in at least one axis at the position of the first sensor in a maximum frequency range of approximately 25 Hz, a selected number of days after the onset of the concussion; The steps include: after the patient wears the sensor device, which can be attached to the head, and engages in physical activity, the computer receives and records signal data from the accelerometer sensor representing the natural movement of the head; Using the aforementioned computer, in order to determine the occurrence and magnitude of any increase in the frequency components of skull movement measured by the accelerometer sensor in a frequency range exceeding approximately the fourth harmonic of the patient's heartbeat indicating non-recovery from a concussion, the signal data is compared with data corresponding to non-concussion, while at the same time, the patient's physical activity prior to the time of the data is examined as a factor in any detected increase in the frequency components. If the comparison by the processor indicates non-recovery from a concussion, the steps include restricting the patient's physical activity. Methods that include...

5. The method according to claim 4, further comprising the steps of: continuing to monitor signal data from the accelerometer sensor when non-recovery from a concussion is indicated; determining whether a concussion is still indicated when the patient returns to physical activity; and repeating this procedure until physical activity no longer indicates non-recovery from a concussion.

6. A device for detecting the recovery state from a concussion in a human patient by detecting and measuring the natural head pulse movement of the patient's head due to blood flow in the brain, and the resulting movement of the brain tissue, A head-mountable device including a first accelerometer sensor that responds to the force of the head pulse in the head, A second accelerometer sensor device, which can be attached to the patient and responds to larger movements associated with shaking or impact during physical activity, A head-mountable device having a first accelerometer sensor and a second accelerometer sensor device, which include an electronic device connected to the sensor for receiving acceleration signals and generating digital acceleration signal data, A processor connected to the electronic device has a comparison means for comparing the signal data with data corresponding to non-concussion, for determining the occurrence and magnitude of any increase in the frequency components of the head pulse motion measured by the first sensor in a frequency range exceeding approximately the fourth harmonic of the patient's heartbeat, indicating non-recovery from a concussion, and relating any such increase in the frequency components to greater motion due to shaking or impact on the patient. Equipped with, The apparatus allows the second accelerometer device sensor to verify, in the event that non-recovery is indicated in the patient several weeks after the confirmation of a concussion, that the patient's physical activity prior to the time of the signal data may be a factor in the detected increase in the frequency component.

7. The apparatus according to claim 6, wherein the second accelerometer sensor device is included in the device which can be attached to the head so as to detect the shaking motion of the head.

8. The apparatus according to claim 7, wherein the processor is located remotely from the head-mountable device and is wirelessly connected to it.

9. The apparatus according to claim 6, wherein the processor is located remotely from the head-mountable device and is wirelessly connected to it.

10. The apparatus according to claim 7, wherein the processor is incorporated into the device which can be attached to the head.

11. The apparatus according to claim 10, wherein the device that can be attached to the head comprises a wearable headset.

12. The apparatus according to claim 6, wherein the device that can be attached to the head comprises a wearable headset.

13. The apparatus according to claim 7, wherein the apparatus and the smartphone have wireless communication capabilities so that, in combination with a smartphone, the smartphone displays and reports any increase in the frequency components.

Images