Seizure and episode detection system and method

Abstract

A seizure and episode detection system includes at least one motion sensing device, at least one intermediary device, and a motion sensing data interpretation device. The motion sensing data interpretation device is configured with a seizure and episode detection program configured to at least one of the following: flag features, flag reference information, including at least one of flagging a seizure stage, flagging a sensing location, flagging a sensing time, and normalize processing, in response to a received piece of motion sensing data.

Description

TECHNICAL FIELD

[0001] The present invention relates to a system and method for detecting seizure and its course, in particular, to a system and method for detecting seizure and its course by biomechanical assessment. BACKGROUND

[0002] Seizure or Epilepsy Seizure is a common neurological disease. There are about 500 million people worldwide suffering from epilepsy. It is generally believed that the proportion of epilepsy patients in the total population is about 0.3-0.7%. The current number of epilepsy patients in China is about 9 million.

[0003] In order to assist physicians and professionals in detecting, diagnosing, evaluating and treating epilepsy patients, various seizure and course detection recording devices have been developed by the industry. The traditional seizure and course detection device mainly detects the brain waves of the patient. This is a direct detection method because it is generally believed that seizure is caused by abnormal discharge of a group of brain cells.

[0004] Techniques for studying and monitoring seizures generally rely on electroencephalography (EEG). When measuring brain electricity, an electrode is used to connect to the scalp or head region of a patient prone to seizure or seizure, and the microelectric signal is measured to represent the electrical activity from the neuronal tissue in the electroencephalogram.

[0005] US 9,332,939 B2 discloses a method for detecting, quantifying and / or classifying seizures using multi-modal data. The invention provides for detecting seizures from the combination of motion sensor data (e.g. accelerometer) and cardiac data (e.g. EKG). For example, a tonic-clonic seizure is associated with reactivity and loss of consciousness. If the patient is standing when the seizure occurs, he will fall to the ground. Therefore, common features of motion can be used, including the fact that an inclinometer will output a peak when the seizure occurs. The accelerometer will output a period of quiet after the seizure (tonic phase), and then a series of quasi-periodic peaks (e.g. about 3 Hz) from the accelerometer. It can be determined that the clonic phase has begun. The invention detects seizures using motion data that is not specific enough, because cardiac data can be used as a reference.

[0006] US2012 / 197092(A1) discloses a dry sensor EEG / EMG and motion sensing system for detecting and monitoring seizures. The invention measures the EEG / EMG and motion activity of a user, automatically determines a seizure and performs an action, such as triggering an alarm and / or shutting down a device that stimulates the seizure. The invention provides for continuous monitoring and storage of the EEG / EMG and motion activity of the user for evaluation by a physician. The invention also provides for mounting the system to eyeglasses and automatically covering the eyes of the user to stop reflex seizures when a seizure is detected.

[0007] WO2017 / 184772A1 discloses a system and method for characterization of seizures. The invention uses electromyography (EMG) to detect seizures. The electrodes of the EMG can be placed on top of the muscle or near the skin to detect electrical activity caused by muscle fiber activation. The invention finds that filtering out low frequency EMG signals can result in more accurate results.

[0008] WO2017 / 184772A1 discloses a system and method for characterization of seizures. The invention uses electromyography (EMG) to detect seizures. The electrodes of the EMG can be placed on top of the muscle or near the skin to detect electrical activity caused by muscle fiber activation. The invention finds that filtering out low frequency EMG signals can result in more accurate results.

[0009] US2013 / 131846A1 discloses a game device for disease treatment. A game machine is disclosed, which can generate game images and sounds and play a game with disease treatment function. The patient's movements are detected by motion sensors and returned to the game device, and the corresponding images are displayed.

[0010] US2018 / 360368A1 discloses a system and method for assessing and treating neurological impairments by assessing voluntary and involuntary neuromuscular activity. Patients are required to perform a prescribed physical and cognitive skill protocol to obtain the data required for remote assessment and treatment of patients. The system uses a variety of, large number of detection devices, including different types of motion sensors, pressure sensors, to obtain the required data required for assessment.

[0011] From the observation of the prior art, it can be found that there is a strong demand in society for the detection, monitoring, recording, judgment, and classification of epilepsy. Many manufacturers have also developed a variety of devices and systems to meet the needs of professionals and patients. However, existing products often require the use of multiple sensors and detectors. And it cannot be ruled out that the use of invasive body sensing devices.

[0012] Therefore, there is a need in the industry for a novel seizure and course detection system that can use simple sensing devices to achieve accurate epilepsy course detection and recording.

[0013] Meanwhile, there is a need for a seizure and episode detection system that can be worn for a long period of time and continuously perform detection, monitoring, recording, determination, and episode classification.

[0014] Meanwhile, there is a need for a seizure and episode detection system that can collect seizure-related information of multiple people for long-term monitoring, recording, diagnosis, and analysis. SUMMARY

[0015] The present invention aims to provide a seizure and episode detection system with simple structure and easy manufacturing, which can sense the movement of a patient's body part and provide useful detection, monitoring, determination, and episode classification data.

[0016] The present invention also aims to provide a seizure and episode detection method that can be applied to a motion sensing device with simple structure and easy manufacturing, and provide automatic detection, monitoring, recording, determination, and episode classification of epilepsy.

[0017] The seizure and episode detection system according to the present invention comprises at least one motion sensing device, at least one intermediary device, and a motion sensing data interpretation device. The at least one motion sensing device is communicatively connected to the motion sensing data interpretation device via the at least one intermediary device. In some embodiments, the motion sensing data interpretation device can be configured in the intermediary device. In particular, it is configured in the intermediary device in the form of application software. The motion sensing data interpretation device can also be configured in a server computer connected to the Internet in the form of application software.

[0018] The motion sensing device comprises at least one tri-axial inertial sensor for sensing the movement of the motion sensing device itself and outputting a sensing reading, an interface device for accepting user input to set at least one predetermined format of the sensing reading output by the motion sensing device, a wireless communication device for establishing a communication channel with the at least one intermediary device to exchange data, and a power supply for supplying electrical power to the sensor, the interface device, and the wireless communication device. In a preferred embodiment of the present invention, the motion sensing device is configured to continuously output the reading of the inertial sensor in the at least one set format via the wireless communication device for a predetermined period of time.

[0019] In a preferred embodiment of the present invention, the motion sensing device can further comprise an angular rate sensor and / or a tri-axial magnetometer.

[0020] In certain embodiments of the present application, the motion sensing device can further comprise a storage device for storing the output readings of the inertial sensor and / or angular rate sensor and / or tri-axial magnetometer. In such embodiments, the motion sensing device is configured to continuously store the readings of the inertial sensor and / or angular rate sensor and / or tri-axial magnetometer in the storage device in the at least one predetermined format for a predetermined time period.

[0021] The intermediary device is a computer device equipped with wireless communication function, preferably a smart phone, configured with necessary application program for establishing communication channel with at least one of the motion sensing devices for exchanging data. The application program can also be used for establishing communication channel with the motion sensing data interpretation device for exchanging data. The intermediary device is configured to transmit the sensing data sent by the at least one motion sensing device to the motion sensing data interpretation device.

[0022] In preferred embodiments of the present application, the interface device of the motion sensing device is configured in the intermediary device. In such embodiments, the intermediary device is preferably configured to provide a setting interface, preferably a graphical setting interface, for a user to input setting parameters to be transmitted to the motion sensing device for setting the predetermined format of the motion sensing device for outputting sensing readings. In other embodiments of the present application, the interface device of the motion sensing device is configured in the motion sensing device.

[0023] The motion sensing data interpretation device is equipped with a storage device for storing the motion sensing data generated by the at least one motion sensing device. The motion sensing data interpretation device is configured with a seizure and episode detection program configured to perform at least one of the following on a piece of received motion sensing data: feature tagging, reference information tagging, such as tagging seizure stage, tagging sensing location, tagging sensing time, and normalization.

[0024] The seizure and episode detection program, when executed by the server computer, can tag features on any piece of motion sensing data. The features include at least one of the following: onset and offset of seizure; seizure stage types and their onset, offset and transition; and description information. The seizure stage types include at least one of the following: tonic phase, clonic phase, postictal phase. The description information includes at least one default description, such as falling down, getting up.

[0025] The seizure and episode detection program, when executed on the server computer, can assign a sensing time to any of the pieces of motion sensing data. The seizure and episode detection program, when executed on the server computer, can normalize any of the pieces of motion sensing data.

[0026] The seizure and episode detection information system can further include a display device for extracting one or more pieces of motion sensing data files from the storage device of the motion sensing data interpretation device, and displaying the files in a user-specified format and manner, according to user input.

[0027] In the above embodiment, the motion sensing data interpretation device is configured to identify at least one synchronization feature in the one or more pieces of motion sensing data files, and set the start and / or end time of the content to be displayed and the display content change frequency, including the change frequency along the time axis and the change frequency of the page display, according to the synchronization feature.

[0028] The present application also provides a seizure and episode detection information platform, including a plurality of motion sensing devices, a plurality of intermediary devices, and a motion sensing data interpretation device. At least one of the plurality of motion sensing devices is communicatively connected to the motion sensing data interpretation device via at least one of the plurality of intermediary devices. In this embodiment, the motion sensing data interpretation device can be configured on the one server computer, and communicatively connected to the at least one intermediary device via the Internet.

[0029] The above and other objects and advantages of the present application can be more clearly understood from the following detailed description when taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 A system diagram showing one embodiment of the seizure and episode detection system of the present application.

[0031] Figure 2 A block diagram of one motion sensing device suitable for use in the present application.

[0032] Figure 3A A seizure episode test result reading waveform diagram showing readings of right ankle tremor amplitude measured for a particular epileptic patient, plotted along a time axis. Figure 3B A table showing a comparison of test results of Figure 3A with physician judgment values.

[0033] Figure 4A A seizure episode test result reading waveform diagram showing readings of left ankle tremor amplitude measured for a particular epileptic patient, plotted along a time axis. Figure 4B A table showing a comparison of test results of Figure 4A with physician judgment values.

[0034] Figure 5A is a seizure course test result reading waveform chart showing the reading of the right wrist tremor amplitude of a specific epileptic patient, the result represented along the time axis. Figure 5B shows Figure 5A a test result compared with the physician's judgment value table.

[0035] Figure 6A is a seizure course test result reading waveform chart showing the reading of the right wrist tremor amplitude of a specific epileptic patient, the result represented along the time axis. Figure 6B shows Figure 6A a test result compared with the physician's judgment value table.

[0036] Figures 7A-7D is another set of seizure course test result reading waveform charts showing the reading of the left ankle ( Figure 7A ), right ankle ( Figure 7B ), left wrist ( Figure 7C ), right wrist ( Figure 7D ) tremor of a specific epileptic patient, the acceleration resultant force (ACC) calculated, the result marked as a feature of the onset point of the tonic phase.

[0037] Figures 8A-8D is another set of seizure course test result reading waveform charts showing the reading of the left ankle ( Figure 8A ), right ankle ( Figure 8B ), left wrist ( Figure 8C ), right wrist ( Figure 8D ) tremor of a specific epileptic patient, the acceleration resultant force (ACC) calculated, the result marked as a feature of the onset point of the clonic phase.

[0038] Figure 9 is a flow chart of a seizure and course detection method suitable for use in the seizure and course detection system of the present invention.

[0039] Figure 10 is a block diagram showing a motion sensing data interpretation device suitable for use in the present invention.

[0040] Reference mark list

[0041] 1 Seizure and course detection system

[0042] 10 Motion sensing element

[0043] 20 Interface device

[0044] 30 Wireless communication device

[0045] 40 power supply

[0046] 50 storage device

[0047] 101-105 motion sensing device

[0048] 201-205 intermediary device

[0049] 300 motion sensing data interpreting device

[0050] 301 server computer

[0051] 302 function module

[0052] 303 human-machine interface DETAILED DESCRIPTION

[0053] The present application is described in the following with reference to the drawings. It has to be noted that the following description of the embodiments of the present application with reference to the drawings is only given by way of non-limiting example. The scope of the present application is not limited to the embodiments described hereinafter.

[0054] Although not bound or limited by any theory, the inventors have found that in the detection, monitoring, recording, judging, classification of the course of epilepsy, the experts rely on visual observation and experience-based judgment criteria to provide diagnosis or recommendations. The main disadvantages of this evaluation method include: the criteria used for evaluation are not objective, mainly based on the experience of experts. The same action may result in different evaluation results and recommendations. In addition, experts observe from a certain angle, and the results may be biased. Even if the video is played at the same time from multiple angles, it is not easy to observe the correct action. Various high-tech instruments can be used to assist professionals in observation and judgment. However, the instruments used so far are specially designed precision instruments, which are very expensive. At the same time, patients need to have probes, electrodes inserted into the body or wear a large number of detection devices in the medical field. This traditional epilepsy detection, monitoring, recording, judging, and course classification operation is fundamentally incompatible with the characteristics of no warning of epilepsy.

[0055] The inventors have found that motion sensing devices made with micro-electro-mechanical technology are small in size and light in weight, and if wireless communication capabilities can be added, they are suitable for wearing on the body to sense the above-mentioned body movements. Although the sensing results of the motion sensing device itself cannot be directly used to observe and monitor the onset of epilepsy, if a setting interface is provided to set the output format of the readings of the motion sensing device, and a device with the ability to interpret the output readings is provided, it can be converted into useful epilepsy seizure description data for expert evaluation and recommendations.

[0056] According to the above findings, the present application proposes a seizure and episode detection system, which comprises at least one motion sensing device, at least one intermediary device, and a motion sensing data interpretation device. Figure 1 That is, a system diagram showing an embodiment of the seizure and episode detection system of the present application. As shown in the diagram, the seizure and episode detection system 1 comprises a plurality of motion sensing devices 101-105, a plurality of intermediary devices 201-205, and a motion sensing data interpretation device 300. Each of the motion sensing devices 101-105 is communicatively connected to the motion sensing data interpretation device 300 via a corresponding intermediary device 201-205. In some embodiments, the motion sensing data interpretation device 300 can be configured in each of the intermediary devices 201-205. In particular, in the form of application software. But in the preferred embodiment of the present application, the motion sensing data interpretation device 300 is configured in a server computer 301 connected to the Internet, in the form of application software. In this preferred embodiment, the motion sensing data interpretation device 300 can be communicatively connected to a large number of intermediary devices 201-205, each of which corresponds to one or more motion sensing devices 101-105. Each of the intermediary devices 201-205 and its corresponding motion sensing devices 101-105 and the motion sensing data interpretation device 300 form a set of seizure and episode detection system 1. Therefore, the server computer can serve a large number of seizure and episode detection systems, becoming a seizure and episode detection information platform.

[0057] Figure 2 A block diagram showing an embodiment of a motion sensing device 101-105 suitable for use in the seizure and episode detection system 1 of the present application. As shown in the diagram, the motion sensing device 101 comprises a motion sensing element 10. The motion sensing element 10 comprises at least one of a three-axis inertial sensor, an angular rate sensor, and a three-axis magnetometer. Preferably, a three-axis inertial sensor and an angular rate sensor or a three-axis magnetometer. The inertial sensor senses the motion of the motion sensing device itself and outputs a three-axis component sensing reading. The angular rate sensor measures the angular displacement in three-dimensional space and calculates the angular velocity change. The three-axis magnetometer measures the three-axis geomagnetism and outputs a three-axis component sensing reading. In many applications, the use of only a three-axis inertial sensor is usually sufficient for the application. However, as the application of motion sensing becomes more widespread, many manufacturers have introduced integrated motion sensing elements that have a three-axis inertial sensor, an angular rate sensor, and a three-axis magnetometer. Such motion sensing elements are very suitable for use in the present application. But the motion sensing element 10 suitable for use in the present application is not limited to this.

[0058] The motion sensing device 101 further comprises an interface device 20 for accepting user input to set the predetermined format of the output of the sensing readings of the motion sensing device 101. The interface device 20 is preferably a graphical setting interface for user to input setting parameters to set the type and format of the output of the sensing readings of the motion sensing device 101. The interface device 20 is connected to the storage or temporary storage device (to be described later) for storing or temporarily storing the sensing readings of the motion sensing element 10 to provide the control parameters inputted by the user to the storage or temporary storage device to set the type and format of the sensing readings provided by the storage or temporary storage device. The type includes the axis represented by the sensing readings of a tri-axial inertial sensor, an angular rate meter and a tri-axial magnetometer, such as the X-axis reading of the tri-axial inertial sensor. The format includes the sensing frequency or time interval of the output of the sensing readings.

[0059] As will be described later, the interface device 20 does not necessarily have to be provided in the motion sensing device 101. In a preferred embodiment of the present application, the interface device 20 of the motion sensing device 101 is provided in one of the intermediary devices 201-205 in the form of an application software. In such an embodiment, the intermediary device is configured to provide a setting interface, particularly a graphical user interface, for user to set the parameters. The setting results are then provided to the motion sensing device 101-105 in a wired or wireless manner.

[0060] The motion sensing device 101 further comprises a wireless communication device 30 for establishing a communication channel with one of the intermediary devices 201-205 to exchange data. The wireless communication device 30 can be any small or miniature wireless communication device. As long as it can effectively communicate with the intermediary devices 201-205, particularly to transmit the sensing readings to the intermediary devices 201-205. As will be described later, the intermediary devices 201-205 are preferably smart phones. In such an embodiment, the motion sensing devices 101-105 only need to have short-range wireless communication capability. In various embodiments, the motion sensing devices 101-105 communicate with the intermediary devices 201-205 via Bluetooth wireless communication channel.

[0061] According to the design of the present application, the plurality of motion sensing devices 101-105 included in the seizure and episode detection system 1 send the sensing readings to the motion sensing data interpretation device 300 via the plurality of intermediary devices 201-205 for processing and interpretation at the motion sensing data interpretation device 300. Preferably, a specific plurality of motion sensing devices 101-105 is assigned to a specific intermediary device 201-205, and the intermediary device is responsible for communication and data exchange with the motion sensing data interpretation device 300. In this way, a specific intermediary device and a specific plurality of motion sensing devices 101-105 form a group. The group is available for use by a specific person, such as a specific patient, a specific physician, or a professional. The motion sensing data interpretation device 300 is usually configured in the cloud, and the sensing readings are sent to the motion sensing data interpretation device 300 for interpretation via the support of the intermediary devices 201-205. In this way, the motion sensing devices 101-105 only need to be general-purpose motion sensing devices, and do not need to be equipped with specific functions, and can be used to perform various monitoring and evaluation related to epilepsy, and provide motion description data required for diagnosis, treatment, and rehabilitation. In this embodiment, the motion sensing data interpretation device can be configured in a server computer, and is connected to at least one intermediary device via an Internet communication.

[0062] Finally, the motion sensing device 101 further includes a power supply 40. The power supply 40 supplies electric power to the motion sensing element 10, the interface device 20, and the wireless communication device 30. Any power supply device can be used in the power supply 40 of the present application. The power supply 40 can be a battery, but can also be a household power supply. The more suitable type is mainly a battery. Because the battery is not tethered by a power cord, the wearer can feel comfortable. The power supply 40 can include a power management chip to save power and prevent accidents.

[0063] In the preferred embodiment of the present application, the motion sensing device 101 is configured to continuously output the readings of the motion sensing element 10 in the set format via the wireless communication device 30 within a predetermined time. Although it is broadcast in form, it is still mainly provided to the specific intermediary device 201-205.

[0064] In other embodiments of the present application, the motion sensing device 101 can further include a storage device 50 for storing the output readings of the motion sensing element 10. In this embodiment, the motion sensing device 101 is configured to continuously store the readings of the motion sensing element 10 in the set format in the storage device 50 within a predetermined time.

[0065] The intermediary devices 201-205 are computer devices equipped with wireless communication capabilities, typically smartphones or tablet computers. The intermediary devices 201-205 can of course also be special-purpose computers equipped with the necessary wireless communication capabilities to read or collect sensor readings from a specific plurality of motion sensor devices 101-105 and to forward the motion sensor data to the motion sensor data interpreter 300. Smartphones are particularly suitable for use in the present application because they can be configured with a variety of application software in addition to the above-mentioned capabilities. However, the intermediary devices suitable for use in the present application are not limited to smartphones and tablet computers.

[0066] Each of the intermediary devices 201-205 is configured with an application program to establish a communication channel with at least one of the plurality of motion sensor devices 101-105 to exchange data. The application program is primarily configured to read, extract or receive sensor data from the motion sensor devices 101-105. The application program is also configured to provide parameter setting functionality to provide user-set control parameters to the motion sensor devices 101-105. The application program can also be configured to establish a communication channel with the motion sensor data interpreter 300 to exchange data. The application program to communicate with the motion sensor data interpreter 300 can be a generic server program. Each of the intermediary devices 201-205 is configured to supply or transmit sensor data sent by at least one of the plurality of motion sensor devices 101-105 to the motion sensor data interpreter 300.

[0067] As mentioned above, the interface device of the motion sensor devices can also be configured in the intermediary devices 201-205. This embodiment has the advantage that the human-machine interface of the motion sensor devices 101-105 can be simplified or even omitted. Other advantages include the possibility of providing a graphical setting interface on the display of the smartphone, for example, to facilitate user input of setting parameters. Since both the motion sensor devices and the intermediary devices are equipped with wireless communication capabilities, the setting parameters can be easily transmitted to the motion sensor devices 101-105 to set the type and predetermined format of the output sensor readings. The graphical human-machine interface can also provide a function to read the sensor readings. This allows the setting and display to be performed on the same interface device.

[0068] In addition, as mentioned previously, in certain embodiments of the present application, at least one motion sensing device 101-105 can also be configured in one of the certain intermediary devices 201-205. In particular, many mobile phones are equipped with motion sensing elements. The motion sensing capabilities of the motion sensing elements can already be sufficient for certain biomechanical evaluations. Such applications are within the scope of the present application, but are not the primary applications of the present application. The most effective applications of the present application are still the use of many miniaturized, light-weight, and non-interfering motion sensing devices to provide motion sensing data to the motion sensing data interpretation device 300 via the intermediary device, although in certain embodiments, the motion sensing data interpretation device 300 can also be configured in the intermediary device.

[0069] The motion sensing data interpretation device 300 is usually configured in a server computer, and thus can be equipped with powerful computing and storage capabilities. The storage device of the motion sensing data interpretation device 300 can store a large amount of motion sensing data generated by the motion sensing devices. For example, if 20,000 people are to be monitored for one year for epileptic seizure, 500 TB of storage capacity can be required. Such a scale can be configured in a small-to-medium enterprise server computer. The motion sensing data interpretation device 300 can be configured with various biomechanical information interpretation programs, each of which provides at least one interpretation function after operation, and is configured to process the received motion sensing data in various ways.

[0070] According to the epileptic seizure and course detection system 1 of the present application, the motion sensing data interpretation device 300 can automatically mark the received or stored motion sensing data, including marking features, marking reference information, such as marking the stage of epileptic seizure, marking the sensing position, marking the sensing time, and normalizing the processing. The features include at least one of the following features: the start and end of the epileptic seizure; the type of epileptic stage and its start, end, and transition; and the generation of description information. The description information includes at least one default description, such as falling down and getting up. The motion sensing data interpretation device 300 can automatically mark any piece of motion sensing data with the type of epileptic stage and its start, end, and transition. The type of epileptic stage includes at least one of the following stages: tonic phase, clonic phase, and postictal phase. In addition, the motion sensing data interpretation device 300 can also automatically normalize the motion sensing data.

[0071] The following will first describe embodiments of detecting epileptic seizures using motion sensing devices. Then, the epileptic seizure and course detection programs designed according to these embodiments will be described.

[0072] Experiment 1: Detection of the staging of epileptic seizure

[0073] Figure 3A is a test result reading waveform chart of a seizure process, showing the reading of the vibration amplitude of the right ankle of a specific epilepsy patient, represented along the time axis. The motion sensor device was attached to the outside of the patient's right ankle, with the patient's foot in contact with the ground, during the test. According to the waveform of the measurement result, a simple application program was written using pattern recognition technology, and executed by a computer to determine the starting time of the three stages of a seizure, namely the tonic, clonic, and postictal stages. The determination result starting time is shown in the upper square of the figure. The numbers in the lower square of the figure are the starting time determined by a physician. Figure 3B shows Figure 3A a comparison table of the test results and the physician's determination values.

[0074] Figure 4A is a test result reading waveform chart of a seizure process, showing the reading of the vibration amplitude of the right ankle of a specific epilepsy patient, represented along the time axis. The motion sensor device was attached to the outside of the patient's right ankle, with the patient's foot in contact with the ground, during the test. According to the waveform of the measurement result, a simple application program was written using pattern recognition technology, and executed by a computer to determine the starting time of the three stages of a seizure, namely the tonic, clonic, and postictal stages. The determination result starting time is shown in the upper square of the figure. The numbers in the lower square of the figure are the starting time determined by a physician. Figure 3A shows Figure 4B a comparison table of the test results and the physician's determination values. Figure 4A is a test result reading waveform chart of a seizure process, showing the reading of the vibration amplitude of the right wrist of the same patient, represented along the time axis, at the time point shown in Figure 5A . Figure 3A shows Figure 5B a comparison table of the test results and the physician's determination values. Figure 5A is also a test result reading waveform chart of a seizure process, showing the reading of the vibration amplitude of the right wrist of the same patient, represented along the time axis, at the time point shown in Figure 6A . Figure 3A shows Figure 6B a comparison table of the test results and the physician's determination values. Figure 6A As shown in the figure, the determination result obtained by the pattern recognition method according to the amplitude change of the waveform is similar to the physician's determination result.

[0075] The absolute value of the error between Figure 5B and Figure 6B is large, but there is no substantial difference between them. Using other recognition methods, correcting the pattern recognition method, or correcting the recognition result (e.g., adding an offset), the determination result can be corrected and the accuracy can be improved. It is proved that the values obtained by simply wearing a motion sensor device on the human body can be used to determine the starting time and ending time of the object's movement.

[0076] Experiment 2: Transition of the stages of a seizure

[0077] Figures 7A-7Dis another set of test result reading waveform graphs showing the measurement of the left ankle (a), right ankle (b), left wrist (c), and right wrist (d) of the patient with epilepsy, calculating the geometric mean value (ACC, formula as follows) and the results of the frequency spectrum analysis. The motion sensor device is placed on the patient's ankle, wrist, and the motion information of the patient during hospitalization is recorded. Figure 7A Figure 7B Figure 7C Figure 7D As shown in the figure, from the results of the frequency analysis, it can be seen that the tonic phase can be roughly considered to occur at 4.5-6 Hz, especially around 5 Hz. At this time, the maximum ACC amplitude of the right ankle is RA, 0.079. However, the frequency of the left wrist is close to noise. The ACC amplitude needs to be integrated again to obtain the swing distance.

[0078]

[0079]

[0080] The frequency of the right wrist appears close to noise. The ACC amplitude needs to be integrated again to obtain the swing distance.

[0081] Figures 8A-8D is another set of test result reading waveform graphs showing the measurement of the left ankle (a), right ankle (b), left wrist (c), and right wrist (d) of the patient with epilepsy, calculating the geometric mean value (ACC, formula as follows) and the results of the frequency spectrum analysis. The motion sensor device is placed on the patient's ankle, wrist, and the motion information of the patient during hospitalization is recorded. Figure 9 Figure 8A Figure 8B Figure 8C Figure 8D The above experiments show that the epilepsy attack and its course can be judged only according to the sensing results of the motion sensor device. The acceleration force value (ACC, formula as follows) is quite important in providing the information needed for judgment.

[0082] Through the above and other related tests, it is found that the epilepsy attack and course detection system of the present application can use simple motion sensing elements, especially very basic general-purpose motion sensing elements, worn on the patient's body, preferably at the end of the hands and feet, such as the wrist and ankle, to generate sensing values during the epilepsy attack and during the epilepsy attack. After appropriate operation, useful features, classification information, sensing information, etc. are marked, correct epilepsy attack and course records are obtained as description data, which are provided to experts as the basis for epilepsy detection, monitoring, recording, judgment, and course classification.

[0083] Through the above and other related tests, it is found that the epilepsy attack and course detection system of the present application can use simple motion sensing elements, especially very basic general-purpose motion sensing elements, worn on the patient's body, preferably at the end of the hands and feet, such as the wrist and ankle, to generate sensing values during the epilepsy attack and during the epilepsy attack. After appropriate operation, useful features, classification information, sensing information, etc. are marked, correct epilepsy attack and course records are obtained as description data, which are provided to experts as the basis for epilepsy detection, monitoring, recording, judgment, and course classification.

[0084] ​​​​​​​​Figure 9 A flow chart of a seizure detection method suitable for use in the seizure detection system of the present application is shown in FIG. 10. As shown, in step 910, the acceleration vector ACC (formula above) is calculated from the three-axis acceleration readings of the motion sensor device, and the three-axis angular velocity readings of the motion sensor device are input. Each is compared to a corresponding threshold value. If both are above the threshold value, it is determined that the limb end on which the motion sensor device is mounted is shaking. In step 920, it is determined whether the shaking is consistent in amplitude. In step 930, it is determined whether the amplitude of the shaking exceeds a predetermined length, to exclude false alarms. After the above steps, it is determined that a seizure has occurred. Next, in step 940, the clustered edges of the recorded waveform are determined, to determine the onset / transition points of the stages of the seizure. In step 950, the DC component of the sensor waveform is filtered out. In step 960, a fast Fourier transform is performed on ACC. In step 970, the peak of the frequency component is detected. In step 980, it is determined whether the seizure is ongoing. If not, it is determined to be the post-seizure stage. If so, in step 990, pulse shaping is performed, and the result is used to determine whether the seizure is in the tonic phase or the clonic phase.

[0085] The present application provides a motion sensor data interpretation device 300, which is configured to automatically label individual sensor result files. Figure 10 A block diagram of a motion sensor data interpretation device 300 suitable for use in the present application is shown in FIG. 11. As shown, according to the design of the present application, the motion sensor data interpretation device 300 is equipped with a storage device 301, which is configured to store motion sensor data generated by at least one motion sensor device 101-105. The motion sensor data interpretation device 300 further includes at least one function module 302, which is configured to implement at least one motion sensor data interpretation program. In the present embodiment, the motion sensor data interpretation program is a seizure detection program, which, when implemented in the function module 302, is configured to manually or automatically add labels to specific motion sensor data files. The automatically added labels include labels of features, labels of reference information, such as labels of seizure stages, labels of sensor locations, labels of sensor times, and normalization processing. The features include at least one of the following features: onset and end of a seizure; types of seizure stages, and their onset, end and transition; and description information. The description information includes at least one default description, such as falling down and getting up.

[0086] The motion sensor data interpretation program can also automatically label any piece of motion sensor data with the seizure stage category and its onset, offset and transition after the execution of the functional module 302. The seizure stage category includes at least one of the following stages: tonic phase, clonic phase, postictal phase. In addition, the motion sensor data interpretation program can also automatically normalize the motion sensor data.

[0087] For example, the motion sensor data interpretation device 300 can be equipped with a human-machine interface 303. The motion sensor data interpretation program can display one or more pieces of biomechanical data in a predetermined format on the human-machine interface 303 after the execution of the functional module 302, so that the user can manually mark the features and / or reference information. In the preferred embodiments of the present application, the motion sensor data interpretation program can automatically generate feature and / or reference information markers after the execution of the functional module 302. For example, the seizure onset time, course transition time, and offset time can be determined using pattern recognition techniques based on the amplitude of the tremor detected by the limb motion sensor device. The identification of each stage can be determined based on the frequency and amplitude of the ACC. In this way, the seizure and course can be accurately recorded. The motion sensor data file after marking has become a seizure and course recording file. Useful information is provided for the detection, monitoring, recording, judgment, and course classification of epilepsy. The interpreted and marked recording file is stored back to the storage device 301 and can be displayed on the human-machine interface 303.

[0088] The motion sensor data interpretation program can label the sensor time for any piece of biomechanical data after the execution of the functional module 302. The motion sensor data interpretation program can normalize any piece of biomechanical data after the execution of the functional module 302. The labeling of sensor time should be known to those skilled in the art. As for normalization techniques, they can be used to correct the bias caused by errors or differences in measurement devices or measurement principles. For example, the bias found in the aforementioned Figures 6A-6B , 7A-7D can be corrected by normalization techniques. Depending on different sensor devices, sensor items and combinations of items, and sensor purposes, experts in the field can also use statistical principles or experience to achieve this. The relevant techniques should not be elaborated. After labeling or normalization is completed, the labeling or normalization results are recorded in the biomechanical data file and stored back to the storage device 301.

[0089] In a preferred embodiment of the present application, the motion sensor data interpretation device 300 is configured to identify at least one synchronization feature in one or more motion sensor data files, and set the start and / or end time of the content of each file to be displayed and the content change frequency, including the change frequency along the time axis and the change frequency of the page display, according to the synchronization feature. In application, the synchronization feature is preferably a time feature. According to the same or corresponding reference time, the sensor results from different sensors or obtained at different times and places are displayed in the same or different formats on the same display screen. This makes it easier for professionals to interpret.

[0090] The above marking features, marking information, normalization and graphing do not have a certain processing order, nor do they necessarily have to be completed. The stage detection of the seizure course of a motion sensor data file, the judgment of the stage type, and the detailed level to which it should be judged are not necessarily regular. The most important thing is that the present application provides a novel seizure and course detection system that can provide accurate seizure and course detection and recording using only the most basic sensor devices. The present application can mark and mark the received motion sensor data, and change the sensor readings that originally have no detection, monitoring, diagnosis, treatment, and rehabilitation reference value into sensor information that can effectively describe seizures and courses. There are no invasive body sensor probes or electrodes, and there are no detection devices that are attached or hung on the body and interfere with normal activities, so they can be worn for a long time and continuously collect motion sensor data for evaluation. The present application further provides a seizure and course detection data platform that can collect a variety of types and large amounts of motion sensor data for long-term detection, monitoring, diagnosis, and analysis.

Claims

1. A seizure and episode detection system comprising at least one motion sensing device, at least one intermediary device, a motion sensing data interpretation device and a display device, wherein: the at least one motion sensing device is communicatively connected to the motion sensing data interpretation device via the at least one intermediary device, the motion sensing device comprises at least one tri-axial inertial sensor for sensing motion of the motion sensing device itself and outputting a sensing reading, an interface device for accepting user input to set at least one predetermined format of the sensing reading output by the motion sensing device, a wireless communication device for establishing a communication channel with the at least one intermediary device to exchange data, and a power supply for supplying electrical power to the tri-axial inertial sensor, the interface device and the wireless communication device, the intermediary device is a computer device equipped with wireless communication functionality and configured with an application program for establishing a communication channel with the at least one motion sensing device to exchange data, the intermediary device is configured to transmit the sensing data sent by the at least one motion sensing device to the motion sensing data interpretation device, the motion sensing data interpretation device is equipped with a storage device for storing the motion sensing data generated by the at least one motion sensing device, the motion sensing data interpretation device is configured with a seizure and episode detection program configured to at least one of the following processes on a piece of received motion sensing data: marking a feature, indicating reference information, including at least one of the following: indicating a seizure phase, indicating a sensing location, indicating a sensing time, and normalizing, the display device is configured to extract one or more pieces of motion sensing data files from the storage device of the motion sensing data interpretation device according to user input and display in a format and manner specified by the user, and the motion sensing data interpretation device is configured to identify at least one synchronization feature in the one or more pieces of motion sensing data files and set a start and / or end time and a display content change frequency of each file content to be displayed according to the synchronization feature, including a change frequency along a time axis and a change frequency of page display. the seizure and episode detection program is capable of marking a feature on any piece of motion sensing data, the feature including at least one of the following: a start and an end of a seizure, a seizure phase type and a start, an end and a transition thereof, and generating description information.

3. The seizure and episode detection system of claim 2, wherein the description information includes at least one default description.

4. The seizure and episode detection system of claim 2, wherein the seizure phase type includes at least one of the following phases: a tonic phase, a clonic phase, and a postictal phase. ​ ​ ​ ​ 2. The seizure and episode detection system of claim 1, wherein, ​ ​ ​ 5. The seizure and episode detection system according to claim 2, wherein the seizure and episode detection program, when executed by the server computer, is capable of marking the sensing time for any piece of motion sensing data.

6. The seizure and episode detection system according to claim 2, wherein the seizure and episode detection program, when executed by the server computer, is capable of normalizing any piece of motion sensing data.

7. The seizure and episode detection system of claim 1, 2, or 4, wherein, The motion sensing device further comprises an angular rate sensor and / or a tri-axial magnetometer.

8. The seizure and episode detection system of claim 7, wherein, The motion sensing device further comprises a storage device for storing the output readings of the tri-axial inertial sensor and / or angular rate sensor and / or tri-axial magnetometer.

9. The seizure and episode detection system of claim 8, wherein, The motion sensing device is configured to store the readings of the tri-axial inertial sensor and / or angular rate sensor and / or tri-axial magnetometer in the storage device in at least one set format continuously for a predetermined period of time.

10. The seizure and episode detection system of claim 1, 2, or 4, wherein, The motion sensing device is configured to output the readings of the tri-axial inertial sensor via the wireless communication device in at least one set format continuously for a predetermined period of time.

11. The seizure and episode detection system of claim 1, 2, or 4, wherein, The interface device of the motion sensing device is configured in the intermediary device.

12. The seizure and episode detection system of claim 11, wherein, The intermediary device provides a set interface for a user to input set parameters to be transmitted to the motion sensing device to set the predetermined format of the sensing readings outputted by the motion sensing device.

13. The seizure and episode detection system according to claim 12, wherein the intermediary device is a smart phone.

14. The seizure and episode detection system of claim 1, 2, or 4, wherein, The motion sensing data interpretation device is configured in the intermediary device in the form of an application software.

15. The seizure and episode detection system according to claim 14, wherein the intermediary device is a smart phone.

16. The seizure and episode detection system of claim 1, 2, or 4, wherein, The motion sensing data interpretation device is configured in a server computer connected to the Internet in the form of an application software.

17. A seizure and episode detection information platform comprising a plurality of motion sensing devices according to any one of claims 1 to 16, a plurality of intermediary devices according to any one of claims 1 to 16 and a motion sensing data interpretation device according to any one of claims 1 to 16, wherein at least one of the plurality of motion sensing devices is communicatively connected to the motion sensing data interpretation device via at least one of the plurality of intermediary devices, and the motion sensing data interpretation device is configured in a server computer and communicatively connected to the at least one intermediary device via the Internet.

Images