A device for detecting and analyzing feeding behavior

Abstract

A device for detecting and analyzing feeding behavior, comprising at least a detection unit for detecting action signals of an individual feeding system, and a processing unit for processing the action signals, wherein the processing unit checks the individual mouth opening, chewing and swallowing action signals acquired by the detection unit in a time sequence based on preset dietary rules, and sends an alarm message to the outside world when the action signal itself or the dietary parameters formed by the processing between the action signals do not conform to the dietary rules, wherein the action signals of mouth opening, chewing and swallowing are realized based on detecting the resting potential reversal signals of the muscle cells involved in the relevant actions.

Description

Technical Field

[0001] The invention relates to the field of medical nursing monitoring, and in particular to a device for detecting and analyzing feeding behavior. Background Art

[0002] At present, there is life care for some patients or disabled people, especially for those who cannot control their own eating behavior due to illness or old age. They often do not swallow or have abnormal swallowing movements when eating due to pathological or physiological reasons. Usually, these patients cannot eat normally or their eating process and eating cycle are irregular. In order to ensure that such patients maintain normal dietary nutrient intake, it is necessary to detect and analyze their eating behavior for application in nursing.

[0003] However, the prior art lacks a solution for judging whether an individual is eating according to preset rules based on the changes in the movements of various parts and the change parameters generated in sequence when an individual is eating, which leads to the unclear situation of the individual when eating, and the monitoring of the elderly or disabled persons with dysphagia cannot be well realized. In addition, the prior art lacks the detection and application of the resting potential change signal of the muscles involved in the individual when eating. On the other hand, for individuals who need to be monitored for a long time, due to the busy work of the caregivers, they are usually in a state without manual care. The prior art usually adopts a solution of continuous machine detection to monitor the individual's eating situation, but because the proportion of the time for people to eat in the total time of the day is very low, and the eating time cannot be clearly determined to a certain period of time, the long-term detection method will not only generate a large amount of detection data and calculation tasks, but also cause a large amount of energy loss of the detection equipment, especially for wearable detection equipment that can move with the individual, its working endurance will be greatly affected, and most of the prior art uses sound sensors as a detection unit, but this type of sensor will not only generate a large amount of sound recognition calculation difficulty, but also cause individuals to worry about the leakage of personal privacy to a certain extent.

[0004] In addition, on the one hand, there are differences in understanding among those skilled in the art; on the other hand, the applicant studied a large number of documents and patents when making the present invention, but due to space limitations, not all details and contents are listed in detail. However, this does not mean that the present invention does not have the characteristics of these prior arts. On the contrary, the present invention already has all the characteristics of the prior art, and the applicant reserves the right to add relevant prior art to the background technology. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a device for detecting and analyzing eating behavior, which at least includes a detection unit for detecting action signals of an individual eating system, and a processing unit for processing the action signals. The processing unit checks the individual mouth opening, chewing and swallowing action signals obtained by the detection unit in a time sequence based on preset dietary rules. When the action signal itself or the dietary parameters formed by processing the action signals do not conform to the dietary rules, an alarm message is sent to the outside world, wherein the action signals of mouth opening, chewing and swallowing are realized based on detecting the resting potential reversal signals of the relevant muscle cells involved in the action.

[0006] The present invention divides the individual's eating actions according to the muscle actions of various parts of the dietary system in a time sequence, and describes the various dietary parameters of the individual's diet in machine processing based on the muscle excitation actions represented by the reversal of the resting potential of each muscle. It also uses preset dietary rules that meet correct or nursing needs to automatically determine irregular events in the individual's diet. The automatic judgment and automatic alarm functions eliminate the need for caregivers to manually observe the individual's daily life all the time, that is, they can remind the individual when the individual has an abnormal diet, which greatly reduces the difficulty of the caregiver's work and improves work efficiency.

[0007] Preferably, the action signals of mouth opening, chewing and swallowing are obtained by detecting the resting potential reversal signals of the muscle cells of the orbicularis oris, masseter and pharyngeal muscles respectively, and each action signal represents an excitation to activate the muscle.

[0008] Preferably, during the continuous detection mode, a detection unit constructed with a vibration or electromyography sensor is used to detect the orbicularis oris muscle action signal and the accompanying simultaneous or delayed vocal cord vibration signal to determine the start or switching time of the precise detection mode, wherein the continuous detection mode is smaller than the precise detection mode in at least the number and type of detection units, the amount of detection data, and the amount of data processing.

[0009] Preferably, when the orbicularis oris muscle action signal is detected and the accompanying vocal cord vibration signal is detected simultaneously or within a preset delay time, the control device remains in the continuous detection mode, and when the vocal cord vibration signal is not detected, the control device starts or switches to the precise detection mode.

[0010] Preferably, in the precise detection mode, the processing unit receives one or several consecutive masseter muscle action signals, and based on the dietary rules, chooses whether to send an alarm command to the outside world according to whether the pharyngeal muscle action signal is detected within a preset time after the last detection of the masseter muscle action signal.

[0011] Preferably, while generating the alarm instruction, the processing unit sends a stimulation instruction to a stimulation unit disposed near the throat of the individual to control the stimulation unit to stimulate the individual to promote the individual to produce a swallowing action.

[0012] Preferably, the stimulation is constituted as at least one or more of tactile stimulation for producing pressure on the throat of the individual and electrical stimulation.

[0013] Preferably, the alarm instruction is received by an alarm unit, which can be configured as an audible, optical, vibrating or mixed alarm, or as a mobile or fixed device near the nursing staff.

[0014] Preferably, the processing unit combines the received action signal and the dietary parameters obtained after processing the action signal and sends them to the information terminal, and the information terminal forms an evaluation of the individual's dietary process.

[0015] Preferably, the detection unit and the processing unit are constructed as a mobile eating behavior detection and analysis device in a manner that they fit the throat of an individual and move with the individual in daily life. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a unit relationship diagram of a preferred embodiment provided by the present invention;

[0017] Figure 2 It is a schematic diagram of the oral muscle position described in the present invention.

[0018] In the figure: 100, detection unit; 200, processing unit; 300, alarm unit; 400, stimulation unit; 500, information terminal; DETAILED DESCRIPTION

[0019] The following is combined with Figure 1 and 2 Provide detailed explanation.

[0020] The present invention provides a device for detecting and analyzing eating behavior, which is used to detect the eating behavior of users or patients and form certain analysis results. For people who need to analyze eating behavior, they are usually normal people with irregular eating habits, abnormal physiological states caused by eating, or patients with eating function defects caused by illness or physiological disability, such as elderly people, congenital or acquired dementia patients, paralyzed patients, patients with muscle incoordination caused by neurological problems, etc. The physiological movement laws, physiological states, eating habits and other parameters displayed by these patients during eating are different from the conventional data of normal people during similar eating processes. Therefore, it is necessary to perform dietary detection based on correction or monitoring nature on such users to achieve dietary management of such patients and ensure their healthy diet. For the convenience of description, the data source object of the eating behavior detection and analysis described in the present invention is called individual.

[0021] This embodiment mainly analyzes and obtains the result data of individual eating by detecting the eating physiological parameters generated by the individual when eating. In this embodiment, the eating physiological parameters have different data structures according to the purpose of the detection and the different detection units 100 used for the detection. In one embodiment, the eating physiological parameters involved at least include the vibration / displacement generated in the throat when the human body eats, and the selected detection unit 100 is generally a pressure sensor or a displacement sensor. For example, a thinner piezoelectric film sensor or a micro displacement sensor can be selected. Accordingly, the above-mentioned detection unit 100 can be set in the neck area of ​​the individual in a fitting manner, especially in the throat area of ​​the individual. When the human body is eating or chewing, its neck, especially the throat area, will vibrate or displace. By detecting the change signal of the vibration along the radial direction of the individual's neck and detecting the change signal of the Adam's apple along the axial movement of the individual's neck, one or more data groups that change over time can be obtained. This data group can be called a vibration data group. The vibration data set may include information such as vibration amplitude, vibration frequency, vibration direction, etc. By analyzing the vibration data set, we can more stereotypically analyze the types of food consumed by individuals, the frequency of chewing during eating, the difficulty of chewing, the frequency of swallowing, the difficulty of swallowing, the time of a single meal, the total time of eating, and other conclusive parameters. Among them, the frequency of chewing is related to the frequency of vibration. The frequency of the vibration can be determined by recording the number of times a value with the same amplitude of vibration changes occurs within a certain period of time. The frequency of swallowing actions can be obtained in the same way. The type of food can be inferred from the frequency and amplitude of chewing / swallowing. For dry, hard foods or foods containing hard solid ingredients, such as biscuits, meat and other ingredients, the dietary characteristics are chewing. The frequency is high but the swallowing frequency is low, for example, only one or several swallowing actions are generated after dozens of chewing; for foods containing liquid objects, such as porridge, drinking water, medicine, etc., the dietary characteristics are that the chewing frequency is low or even zero, and the swallowing action is long; for solid ingredients containing more water, such as meatballs, apples and other fruits, the dietary characteristics are that the chewing frequency is low and a swallowing action will be generated after several chewings, and the overall characteristics show a low-frequency chewing signal accompanied by a swallowing signal that appears single or repeatedly; for solid-liquid mixed ingredients that contain both liquid objects and some solid objects, such as pearl milk tea, tremella soup, etc., the dietary characteristics are long-term swallowing actions accompanied by chewing action signals. Signal recognition of chewing and swallowing actions in vibration or displacement can be achieved through physiological changes in the human body during eating, especially in the neck, combined with certain experiments to obtain a specific chewing or swallowing numerical amplitude range. By identifying a specific range of values ​​in a large amount of vibration data, data that can describe the chewing or swallowing action can be found, and the data is arranged and processed based on time series.

[0022] In another embodiment, the physiological parameters of eating involved at least include the sound produced in the throat or mouth when the human body eats. The detection unit 100 that may be used can be selected as a detection device such as a microphone or a sound sensor, which can generally be attached to or fixed on or near the neck or mouth of an individual to detect the sound produced when the individual eats. Different types of food will produce different dietary characteristics when eating. For hard solid food, sharp bite cracking sounds or a large amount of friction sounds are often produced when chewing. For example, when eating foods such as potato chips, biscuits, and nuts, a large click sound will be produced due to the bite. This sound has a short single duration, a high frequency, and a sharp tone. These characteristics can be used to reflect that the individual is eating hard solid food; for liquid food, the gurgling sound produced when the liquid food and air mix into the throat when drinking and swallowing; for food rich in water, a large amount of juice will be produced when chewing. Judging the type of food by sound can also be done through a large number of prior tests to obtain the unique sound corresponding to each ingredient. By recording the tone, intensity, frequency, duration and other parameters of the sound produced by each ingredient when being eaten, a database of sounds corresponding to the ingredients can be obtained. The next time an unknown food sound is detected, the type of ingredient eaten that time can be obtained by matching the characteristics in the database, and it can even be accurate to the exact variety of the ingredient.

[0023] In another embodiment, the detection of the diet of the individual is implemented by the simultaneous detection of vibration and sound. Through the complementarity of the two detection schemes, more accurate dietary characteristics can be obtained. By analyzing these dietary characteristics, dietary data about the individual during the diet process can be obtained. The dietary data can include the ingredients of the individual's diet, the duration of the diet, the number of chewing, frequency, duration during the diet process / the number of swallowing, frequency, duration during the diet process, and other parameters. Preferably, a standard value can be set for each ingredient according to the pre-medical or experimental results. The standard value is constituted as the dietary data of a normal person eating this kind of ingredient under normal circumstances. By comparing the current dietary data with the standard value, the difference between the two can be obtained, and then the difference is output to the outside world as an optional output item. The outside world here can refer to the individual itself, or it can be a related caregiver or dietary researcher, which is used to evaluate whether the dietary data of the individual when eating this food meets the normal level or the extent of deviation from the normal level. In addition, an alarm message can be set. When one or more items of the dietary data detected this time exceed the standard value, an alarm message is sent to the outside world to remind the caregiver or the individual itself.

[0024] In another embodiment, a function of detecting the continuity of eating is also provided. By continuously detecting the eating process by the detection unit 100, it can be known that the individual is eating at this time. After matching the corresponding standard value, the standard value and the detected eating data are compared based on the time series to roughly know what stage the current eating is in. Accordingly, when the eating data that is higher than the threshold and does not meet the standard value change for a long time is detected, it can be determined as an abnormal diet, that is, the continuity of eating is destroyed. One of the main reasons for this situation is that the individual is physically or mentally ill. The inability to chew / swallow or the low chewing / swallowing ability level caused by the influence of psychology, for example, the elderly, during the process of eating, are more likely to stop chewing or swallowing due to senile functional disorders such as Alzheimer's disease. At this time, the food in the elderly's mouth cannot be turned into a food bolus after chewing or cannot be swallowed into the stomach by the elderly. This situation generally causes the elderly to be obstructed from eating this time. The lighter consequence is that the total amount of energy obtained from the food is reduced. The more serious consequence is that the unprocessed food in the elderly's mouth slides into the elderly's esophagus, causing the elderly to choke and suffocate. Therefore, the monitoring of disabled people, especially the elderly, for the continuity of eating is very important. Preferably, on this basis, when the eating continuity detection process detects a sudden stop of chewing or swallowing, a prompt message is sent to the individual itself or the caregiver to remind the individual to chew or swallow or to remind the caregiver to perform on-site care.

[0025] Preferably, the above functions can be realized by a dedicated detection component, which is usually composed of a circuit combination of a large number of electronic components, among which the more important components are the processing unit 200 for controlling, calculating, and sending and receiving data, the detection unit 100 for detecting the individual's eating physiological parameters in the vicinity of the individual, the signal processing unit 200 for amplifying and converting analog signals into digital signals, and the power supply unit for powering. Preferably, in order to improve the overall portability and timeliness of detection of the detection component, it can be constructed to fit the throat of an individual and not cause great difficulties to the individual's daily life. For example, the detection unit 100 for detecting vibration can be set as a piezoelectric film sensor, which is thin and flexible. It can be set behind an elastic layer, and the elastic layer is set thereon by surrounding or semi-surrounding the neck of the individual. The power source can be selected as a small battery. In order to reduce the weight of the device set on the neck of the individual, preferably, a large amount of calculation and data storage can be set separately outside the individual, for example, an independent computing device is selected as a functional carrier. The selectable computing device is a smart phone, tablet, computer and other devices. The data detected by the detection unit 100 can be transmitted to the processing unit 200 through a wireless transmission unit after signal processing by the processing unit 200. The processing unit 200 can be selected as a processor of various smart devices or computers.

[0026] Preferably, the present invention provides a scheme or process for judging an individual's eating behavior. First, the individual's eating process needs to be divided into several representative stages according to the time sequence, namely, the concept stage, the preparatory stage, the chewing stage, the swallowing stage and the esophageal stage. The concept stage refers to the period of time when an individual is stimulated by external stimuli or his own ideas and wants to eat a certain food. For example, an individual sees an apple on the table and wants to eat it. The general physiological manifestation of this period is generated in the individual's brain and is the result of neural information interaction. The preparatory stage refers to the period of time when an individual drives his limbs or other parts of the body except the digestive system to move the food to the mouth after he wants to eat a certain food. For example, an individual uses his hands to bring the apple on the table to his mouth. The chewing stage refers to the process in which an individual opens his mouth to bite off part of the food and processes it in the oral cavity to turn it into a food bolus. For example, an individual bites off part of the apple and chews it into apple crumbs. This kind of food is chewed once or multiple times and mixed with saliva to form a mass, block, or shredded form. This step in the chewing stage is to process the food into a food bolus that is convenient for subsequent smooth swallowing. The swallowing phase refers to the period of time when the individual controls the throat to block the nasal cavity, vocal tract and other passages after the food bolus is produced in the oral cavity, and the esophagus is opened and the food bolus is sent into the esophagus. For example, the instantaneous process of an individual swallowing the food bolus in the oral cavity into the esophagus will mobilize a large number of muscles in the digestive system to participate in the closure and opening of the passage and the function of actuation. The esophageal phase refers to the process in which food moves in the individual's esophagus and eventually enters the stomach. This process is mainly achieved by the peristalsis of the esophagus to drive the food bolus into the stomach.

[0027] The detection unit 100 described in this embodiment is used to detect the triggering of the corresponding muscles or physiological characteristics at least in the process from the chewing stage to the esophageal stage in a direct or indirect manner, thereby obtaining the individual's dietary characteristics, dietary abnormalities, and / or non-diet pseudo data generated during the removal process. Specifically, the detection unit 100 used to detect different positions, different muscle groups, and different physiological parameters transmits data to the processing unit 200 in real time. The processing unit 200 implements the above functions through logical operations, judgments, statistics, and other schemes. The detection unit 100 can be implemented in the form of electromyographic sensors, vibration sensors, and sound sensors.

[0028] The basic data source detected in this embodiment can be based on the electromyographic action signal generated by the individual when eating. The muscle is in a resting state when it does not produce an action. There is a potential difference between the inside and outside of the cell membrane. In the resting state, the potential inside and outside the membrane, that is, the membrane potential is characterized as being more negative inside the membrane than outside the membrane. If the outside membrane potential is set to 0, the inside membrane potential is generally negative, and is -90mV under normal circumstances. The membrane potential at this time can be called the transmembrane resting potential. When the muscle generates excitement or an excitation signal, the membrane potential of the excited part changes rapidly, and the difference between the inside membrane potential and the outside membrane potential decreases rapidly. At a certain moment, the inside membrane potential becomes more positive than the outside membrane potential, and then quickly changes to the resting potential again. By detecting the potential reversal of the membrane potential, that is, the process of converting the resting potential into an action potential and then returning to the resting potential, it can be judged as a feature of the muscle generating an action. The electromyographic detection can be realized by using sensors including but not limited to: single-lead, dual-lead or multi-lead electromyographic modules, electromyographic sensors, electrical sensors, muscle excitation sensors, neural excitation sensors, etc., all of which can be regarded as the detection unit 100 described in this embodiment. The detection unit 100 transmits the detected electromyographic signal, that is, the change information of the membrane potential of the muscle cells of the corresponding muscle part of the individual, to the processing unit 200.

[0029] When the processing unit 200 receives the action signal of the orbicularis oris muscle, it determines that the individual has made an opening action at this time. The processing unit 200 receives and processes the action signal of the masseter muscle in the next preset period of time, processes the change in the amplitude of its action into a heterogeneity value, and determines whether the individual is eating or talking by judging whether the heterogeneity value is greater than the preset first differential standard value. Generally speaking, when an individual is eating, the reversal frequency of the resting membrane potential of the masseter muscle is relatively stable, or has a clear trend of gradually increasing or decreasing. Accordingly, the calculation of the heterogeneity value is determined based on the amplitude of the analysis frequency change. The heterogeneity value can be obtained using a scoring system. For example, a score can be obtained when the frequency change exceeds a preset amplitude value. When the score is higher than the range of normal chewing, it can be considered that the heterogeneity value is high, that is, the amplitude of the frequency change is relatively chaotic. At this time, it can be considered that the individual is not chewing but talking. Preferably, when obtaining the dissimilarity value to determine whether the individual is in a chewing or speaking mode, other detection units 100 can also be used to assist in confirming the judgment result, for example, by recognizing the sound of the individual's speech through voice recognition, and by detecting the swallowing-related throat muscle movements to detect whether the individual has a swallowing movement. Common dietary electromyographic signal characteristics can be summarized as a sequential excitation process that includes at least the following muscles or muscle groups: the orbicularis oris muscle performs at least one movement, the masseter muscle performs at least one movement, and the throat muscle performs at least one movement. The three muscles or muscle groups change their resting potentials in a reversed manner in the order of time to generate at least one excitation. The processing unit 200 summarizes the time, frequency, number, amplitude and other rules of the action signals of each muscle into dietary parameters, which mainly reflect information that the action signal itself cannot express, such as the number of chewing times, frequency, etc.

[0030] Preferably, this embodiment also establishes a configuration scheme for the start time of dietary detection. When an individual wears a detection device provided with the detection unit 100 and lives a normal life, during the continuous detection mode, a vibration or electromyography sensor is used to detect the orbicularis oris muscle opening signal and the accompanying or delayed vocal cord vibration signal to determine the start or switching time of the precise detection mode. When eating, an individual will inevitably control the orbicularis oris muscle to open first. This step is an inevitable step to be taken in eating. However, the action of opening the orbicularis oris muscle is not only produced during the eating process. The individual may also cause the orbicularis oris muscle to open when talking, yawning, hiccuping, sneezing, or even snoring during sleep. Therefore, this embodiment removes these pseudo events that do not need to be detected by determining the accompanying vocal cord vibration detection when the orbicularis oris muscle opens or based on a certain preset delay time. Because among the above-mentioned pseudo events that generate the orbicularis oris muscle action stimulus, almost all events will produce certain sounds, such as speaking sounds, yawning sounds, hiccuping sounds, sneezing sounds, etc. When the processing unit 200 detects the orbicularis oris muscle action or if the accompanying vocal cord vibration signal is detected afterwards, it is determined that the individual is experiencing a pseudo event at this time, and the precise detection mode is not turned on or switched. On the contrary, when the vocal cord vibration is not detected, the precise detection mode is controlled to be switched. The precise detection mode refers to a mode in which the processing unit 200 controls all or a large number of detection units 100 that detect the individual's eating process to open and improve the detection sampling accuracy, such as the detection units 100 that detect the masseter muscles, pharyngeal muscles, tongue muscles and other muscles, and improve the sampling accuracy of these detection units 100, so that the sampling interval time is shortened. It can be seen that the precise detection mode generates a larger amount of data, more types of diet-related data, and can better reflect the various situations of individuals when eating compared to the continuous detection mode.This embodiment does not use a sound detector as a solution for detecting the sound emitted by an individual. This is mainly because the use of a sound sensor first requires a large amount of sound to be input in advance, and a dedicated program is required to perform a large number of complicated processes such as sound collection, impurity removal, waveform adjustment, sound matching and retrieval. The addition of these contents will cause a large amount of calculation data to be generated, especially for a detection device that is small in size and can be worn on the throat of an individual. Such a large amount of calculation undoubtedly increases the calculation burden of the processing unit 200, increases energy consumption, and occupies a large part of the computing power of the processing unit 200, and also increases the cost of use. On the other hand, as people's awareness and demand for personal privacy protection increase, a detection unit 100 with sound collection and analysis is applied to sound detection at every moment in an individual's life, which undoubtedly deepens the individual's concerns about the detection device. Most users with only ordinary knowledge cannot confirm whether the device is really intended only to judge sound patterns such as speaking, yawning, sneezing, etc. instead of collecting the content of the user's speech and the sounds around the user's life at any time. Especially when using the Internet to perform an artificial intelligence-based sound judgment solution, users are worried about whether their speech content is transmitted to an unsafe Internet and causes information leakage. However, this embodiment can eliminate the above-mentioned pseudo events by only using a vibration sensor, without involving a microphone device, which is easier to be accepted by users than a sound collection unit, and will not cause psychological burden on users. In addition, the detection process of the user eating involves a relatively complex muscle excitation process, a muscle coordination process, and a nerve signal excitation process, and involves a large number and type of detection units 100. Therefore, the solution of judging the signal that the individual is eating from the pseudo event by only detecting the orbicularis oris muscle and the accompanying vocal cord vibration signal in the continuous detection mode to start or switch the accurate detection mode can effectively reduce the processing workload of the processing unit 200 when not eating, greatly reduce the amount of data generation, throughput, processing and storage in this time period, and then reduce the energy consumption loss of the detection device, which is very important for a mobile detection device that can be worn on the throat of an individual at any time, and can reduce the capacity selection of the relevant power supply, reduce the volume and weight of the power supply, and significantly improve the comfort of wearing.

[0031] Preferably, for situations that may cause misjudgment in some individuals, such as meaningless opening of the mouth without making any sound, and making sounds from the nasal cavity when eating, the misjudgment can be eliminated by the subsequent activation of the masseter muscle detection and the specific vocal cord vibration frequency conditions of the nasal cavity sound.

[0032] Preferably, after switching to the precise detection mode, the detection unit 100 detects one or several consecutive signals of reversal of the resting potential of the masseter muscle cells, and it can be determined that the individual is chewing at this time. Within a preset period of time after the last detection of the masseter signal, if the throat muscles responsible for swallowing do not generate a resting potential reversal signal, it is determined that the individual is not swallowing, and the processing unit 200 can issue an alarm instruction to the alarm unit 300 accordingly. The alarm unit 300 can be configured as an audible, optical, vibration or mixed alarm, or as a mobile or fixed device next to the caregiver, which will issue an alarm prompt to the caregiver or individual after receiving the alarm instruction.

[0033] Preferably, the above-mentioned conditions for judging whether an individual swallows after chewing can be summarized in the dietary rules. The dietary rules constitute a condition or a set of several conditions, which are used to judge whether each step of the individual in the dietary process is correct or meets the care needs. The condition set can be edited and updated arbitrarily by relevant personnel. For example, the number of chewings that meet the care needs is at least 3 times. Then when the individual chews only once or less than 3 times, the processing unit 200 will determine that it does not comply with the dietary rules, and then generate an alarm instruction.

[0034] Preferably, in the above case, the processing unit 200 may also send a stimulation instruction to the stimulation unit 400 disposed near the throat of the individual, and the stimulation unit 400, after receiving the instruction, stimulates the throat of the individual to promote the swallowing action of the individual. This stimulation may be selected as a tactile stimulation that generates pressure on the throat of the individual or a micro-current stimulation that is safe for the human body and has a very low voltage on the throat of the individual. After receiving the stimulation, the individual will have an unconscious swallowing action, so as to achieve the effect of prompting the individual to swallow the food bolus.

[0035] Preferably, the processing unit 200 can also upload all the collected motion signals and processed dietary parameters to the information terminal 500. The information terminal 500 can be constructed as a large server, a nursing center supervision facility, a nurse's station, a medical terminal and other equipment. The information terminal 500 displays these data to professional nursing staff to obtain manual dietary evaluation of the individual.

[0036] It should be noted that the above-mentioned specific embodiments are exemplary, and those skilled in the art can come up with various solutions inspired by the disclosure of the present invention, and these solutions also belong to the disclosure scope of the present invention and fall within the protection scope of the present invention. Those skilled in the art should understand that the present invention specification and its drawings are illustrative and do not constitute a limitation on the claims. The scope of protection of the present invention is defined by the claims and their equivalents. The present invention specification contains multiple inventive concepts, such as "preferably", "according to a preferred embodiment" or "optionally", all of which indicate that the corresponding paragraph discloses an independent concept, and the applicant reserves the right to file a divisional application based on each inventive concept.

Claims

1. A device for detecting and analyzing feeding behavior, comprising at least: A detection unit (100) is used to detect an action signal of an individual eating system, A processing unit (200) is used to process the action signal. It is characterized in that The processing unit (200) checks the individual mouth opening, chewing and swallowing action signals acquired by the detection unit (100) in a time sequence based on preset dietary rules, and sends an alarm message to the outside world when the action signal itself or the dietary parameters formed by processing the action signals do not conform to the dietary rules, wherein the mouth opening, chewing and swallowing action signals are realized based on respectively detecting the resting potential reversal signals of the muscle cells of the orbicularis oris, masseter and pharyngeal muscles involved in the action, and the three muscles or muscle groups change their resting potentials in a sequentially reversed manner in the order of the time sequence to generate at least one excitation, and by detecting only the orbicularis oris and the accompanying vocal cord vibration signals in the continuous detection mode, the signal that the individual is eating is judged from the pseudo-event to start or switch the precise detection mode, When the processing unit (200) detects the orbicularis oris muscle action signal and simultaneously or within a preset delay time detects the accompanying vocal cord vibration signal, it is determined that the individual is experiencing a pseudo-event, and the control device maintains the continuous detection mode. When the vocal cord vibration signal is not detected, the control device turns on or switches to the precise detection mode.

2. The device according to claim 1, characterized in that During the continuous detection mode, the detection unit (100) constructed using a vibration or electromyographic sensor is used to detect the orbicularis oris muscle action signal and the accompanying simultaneous or delayed vocal cord vibration signal to determine the start or switching time of the precise detection mode, wherein the continuous detection mode is smaller than the precise detection mode in terms of at least the number, type, detection data volume, and data processing volume of the detection units (100).

3. The device according to claim 2, characterized in that In the precise detection mode, the processing unit (200) receives one or several consecutive masseter muscle action signals, and based on the dietary rules, chooses whether to send an alarm instruction to the outside world according to whether the pharyngeal muscle action signal is detected within a preset time after the last detection of the masseter muscle action signal.

4. The device according to claim 3, characterized in that While generating the alarm instruction, the processing unit (200) sends a stimulation instruction to a stimulation unit (400) disposed near the throat of the individual to control the stimulation unit (400) to stimulate the individual to promote the individual to produce a swallowing action.

5. The device according to claim 4, characterized in that The stimulation is at least constituted as one or more of tactile stimulation and electric current stimulation that generates pressure on the throat of the individual.

6. The device according to claim 5, characterized in that The alarm instruction is received by an alarm unit (300), and the alarm unit (300) can be configured as an audible, optical, vibration or mixed alarm, or as a mobile or fixed device around a nursing staff.

7. The device according to claim 1, characterized in that The processing unit (200) combines the received action signal and the dietary parameters obtained after processing the action signal and sends them to the information terminal (500), and the information terminal (500) forms an evaluation of the individual's dietary process.

8. The device according to claim 1, characterized in that The detection unit (100) and the processing unit (200) are both constructed as a mobile eating behavior detection and analysis device in a manner that they fit the throat of an individual and move with the individual in daily life.

Images