A probe assembly as a feeding tool for a patient

Abstract

A probe assembly as an eating tool for a patient, the probe assembly is used to carry food and detect eating data, the probe assembly is configured as a structure on which at least a part can carry food and can be operated by the patient to send the food into the mouth, a visual detector is arranged thereon, and the visual detector is configured to acquire the kind data of the food carried on the probe assembly itself and / or the patient eating image data in another direction different from the food position based on at least two-way visual detection.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of medical care monitoring, in particular to a detection assembly as a patient eating tool. BACKGROUND

[0002] The elderly population or the population with eating disability due to some diseases needs to be cared for when eating, otherwise they may have eating difficulties and cannot eat. The current common solution is for nursing staff to monitor the entire eating process of the relevant population at all times, but this solution will cause discomfort to some patients on the one hand, and will greatly increase the labor of the nursing staff on the other hand.

[0003] CN110119107B discloses a food intake detection method. Some diet habit monitoring technologies assisted by technical devices, such as cameras and RFID markers, are implemented in laboratory conditions and are harmful to the human body. The present application is a food intake detection device, which comprises a neck strap, a detection module and a control module. The control module comprises a power module, a voltage amplification module, a filter module and a main control module. The detection module comprises a curved shell, an upper protective sponge, a piezoelectric sensor and a lower protective sponge. The upper protective sponge, the piezoelectric sensor and the lower protective sponge are sequentially stacked. The piezoelectric sensor is on the neck strap, and the user can wear it comfortably. Compared with invasive sensors, the present application is safe and healthy, and is convenient to carry. The present application combines a decision tree intelligent recognition algorithm to judge the food type, and then determines the corresponding food type according to the vibration signal detected by the piezoelectric sensor.

[0004] However, the prior art lacks research on the impact of the detection device on the patient's perception. Some patients, especially some patients with abnormal mental state, will have a strong repulsion to the observation of their eating process, and even refuse to eat in the presence of nursing staff or obvious monitoring devices. In addition, without manual observation, it is very difficult to determine the type of food eaten by the patient, especially when the patient chooses the food himself. This problem is particularly prominent, and without clear information about the type of food eaten by the patient, it will be very difficult to guide the patient's eating, because the processing process of the food in the mouth is closely related to the type of food.

[0005] And the prior art lacks a scheme for judging whether an individual is eating according to preset rules according to the changes in each part of the individual in sequence and the change parameters generated when eating, resulting in that the situation of the individual when eating is unknown, and the monitoring of the elderly or disabled persons with difficulty in swallowing cannot be well achieved. In addition, the prior art lacks detection and application of the resting potential change signal of the muscles related to the individual when eating. On the other hand, the prior art usually adopts a scheme of continuous detection by a machine to monitor the eating situation of an individual, but since the time for eating of a person accounts for a very low proportion of the total time in a day, and the eating time cannot be definitely determined to a certain period, the way of long-time detection will not only produce a large amount of detection data and calculation tasks, but also cause a large amount of energy consumption of the detection device, especially for a wearable detection device that can follow the individual, the working endurance will be greatly affected. The prior art usually adopts a sound sensor as a detection component, but such a sensor will not only produce a large amount of sound recognition calculation difficulty, but also cause the individual's concern about the leakage of personal privacy to a certain extent.

[0006] In addition, on the one hand, there are differences in the understanding of those skilled in the art; on the other hand, a large number of literatures and patents have been studied by the applicant when making the present application, but due to the limitation of space, all the details and contents have not been listed in detail, which does not mean that the present application does not have these characteristics of the prior art. On the contrary, the present application already has all the characteristics of the prior art, and the applicant reserves the right to add relevant prior art in the background art. SUMMARY

[0007] In view of the deficiencies of the prior art, the present application provides an intelligent eating reminding device, which at least comprises a detection component for carrying food and detecting eating data, the detection component is configured as a structure that at least part of it can carry food and can be operated by the patient to send the food into the mouth, and a visual detector is arranged thereon, the visual detector is configured to acquire the kind data of the food carried on the detection component and / or the patient eating image data in another direction different from the food position based on at least two-way visual detection.

[0008] The present application provides a detection device with detection function and auxiliary eating tool function to the patient, to a certain extent, the detection is integrated into the process of the patient eating. The patient, especially the older patient with mental confusion or certain paranoid personality, usually has great resistance to the detection of his eating process. The detection component is configured as a tool for the patient to eat, such as a spoon. This method is less likely to stimulate the patient, and the visual detection method can conveniently identify the food type through machine vision when the patient is eating. Compared with manual input or indirect detection of food type by sound, vibration, etc., this method is more rapid, accurate and simple.

[0009] Preferably, the part of the detection assembly contacted by the body part of the patient when operating the detection assembly is provided with a touch detector, which selectively generates a trigger signal in a manner capable of detecting whether the body part of the patient is in contact.

[0010] Preferably, based on the generation of the trigger signal, a visual detector on the detection assembly is controlled to be turned on for detection at this time.

[0011] Preferably, in the case of selecting to turn on the continuous eating mode, based on the continuity of the trigger signal of the touch detector and the abnormal eating signal of the patient detected by the visual detector, an alarm information is selectively sent to the outside world.

[0012] Preferably, when the trigger signal is interrupted, an alarm information is sent to the outside world, the visual detector continuously detects the eating process of the patient based on the preset normal eating standard, obtains an abnormal eating signal by comparing the detection data and the preset eating standard, and sends an alarm information to the outside world after the abnormal eating signal is generated.

[0013] An intelligent eating reminding device, comprising at least a detection assembly for detecting at least one action signal in the eating process of a patient, and a calculation assembly for processing the action signal, wherein the detection assembly detects and sends corresponding action signals to the calculation assembly based on the action sequence of mouth opening, chewing and swallowing of the patient, the calculation assembly determines the legality of the action signal based on a preset diet rule, and alarms the outside world when an illegal action signal is generated, and the action signal is obtained based on the reversal signal of the resting potential of muscle cells of the orbicularis oris muscle, the masseter muscle and the laryngeal muscle.

[0014] The present application divides the eating actions of an individual into a time sequence according to the muscle actions of each part of the eating system, describes each eating parameter of the individual eating in the machine processing based on the muscle excitation action represented by the reversal of the resting potential of each muscle, and automatically determines the non-standard events in the individual eating by using the preset diet rule conforming to the correct or nursing needs. Through automatic judgment and automatic alarm function, the caregiver does not need to observe the daily life of the individual at all times, and the individual can be reminded when the individual has abnormal eating, which greatly reduces the working difficulty of the caregiver and improves the working efficiency.

[0015] Preferably, the detection assembly is composed of a vibration or electromyographic sensor, wherein the calculation assembly determines whether a false event affecting the judgment of the eating time is generated by only judging whether the vocal cord vibration signal is generated after detecting the action signal of the orbicularis oris muscle, controls itself and the detection assembly to enter the accurate detection mode when no false event is generated, and the accurate detection mode is more than the continuous monitoring mode in at least the type, data, detection position and data amount of the detection assembly.

[0016] Preferably, the computing component controls to enter the accurate detection mode when the lip muscle movement signal is detected.

[0017] Preferably, in the accurate detection mode, the computing component generates an alarm to the outside world when no movement signal of the throat muscle is received within a preset time after the last time the masseter muscle movement signal is received based on the preset eating rule.

[0018] Preferably, the computing component controls the stimulation component arranged on the throat part of the patient to generate stimulation to the part of the patient to promote the generation of swallowing when the alarm to the outside world is generated. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 is a unit relationship diagram of a preferred embodiment provided by the present application;

[0020] Figure 2 is a schematic diagram of the oral muscle part described in the present application.

[0021] In the figure: 100, detection component; 200, computing component; 300, reminding component; 400, stimulation component; 500, central terminal; DETAILED DESCRIPTION

[0022] The present application will be described in detail below with reference to the accompanying drawings. Figure 1 and 2 will be described in detail.

[0023] The present application provides an intelligent eating reminding device, which is used at least for detecting the eating process of a user or a patient. The eating process is a necessary behavior for every physiologically human being, which provides energy, material, water and the like required for maintaining physiological activities. However, irregular diet and eating disorder phenomenon occurs in a large number of people. The lower degree of eating problems usually occur in the eating process of normal people, for example, people usually generate a set of their own eating habits when eating. Although the eating habits of every person generally follow a certain mainstream rule, there are still differences in details. The more serious problem is the eating disorder problem. The people with limited ability or ability defects, such as the old, the mentally ill, the disabled, the slightly paralyzed, and the people with nervous system problems, have great differences in the eating process compared with normal people. For the above-mentioned cases of lower or higher degree of eating problems, it is necessary to detect or monitor the eating process of these people. For the convenience of description, the detection object in the present application is referred to as a patient.

[0024] In this embodiment, the eating result of a person is analyzed by examining the physiological parameters of eating. The physiological parameters of eating in this embodiment can vary according to the purpose of the test and the data composition of the detection module 100. In one embodiment, the physiological parameters of eating include at least the vibration / throat displacement generated by the throat of a person when eating, and the detection module 100 can be selected as a pressure sensor or displacement sensor that detects the throat area. When a person is eating or chewing, the neck, especially the throat, will generate vibration or displacement. By detecting the vibration signals along the radial direction of the neck, the change in the movement of the laryngeal prominence along the axis of the neck is formed into one or more vibration data sets. The vibration data set can contain the amplitude, direction, frequency, etc. of the vibration. By analyzing the vibration data set, information such as the type of food, chewing frequency, chewing difficulty, swallowing frequency, swallowing difficulty, single eating time, total eating time, etc. can be obtained. The number of times the vibration with the same vibration amplitude changes can indirectly represent the number of times the patient chews while eating, and in addition to the cumulative count, the time interval between the counts can also be processed to obtain the frequency. By considering the frequency and amplitude as a whole, the type of food eaten by the patient can be obtained. For food with sufficient juice, the characteristic of eating is that the vibration generated by chewing occurs repeatedly combined with the action of the larynx rising once. For liquid food, the characteristic of eating is that there is no chewing, and the larynx moves up and down in an orderly swallowing action for a period of time. The identification of the vibration or displacement of the chewing and swallowing action can represent the physiological changes in the eating process, especially the changes in the neck, which can be achieved by identifying a series of specific values of chewing or swallowing through experiments. By screening a large amount of vibration data, a series of values can be found to describe the chewing or swallowing action, and processed in time sequence.

[0025] In another embodiment, the physiological parameters of eating include at least the sound generated by the throat or mouth when eating, and the detection module 100 can be selected as a microphone, voice sensor or other measuring device, which can be attached or fixed to the neck or mouth or nearby of the person to detect the sound generated by eating. Different types of food will produce different eating characteristics when eating, that is, the sound will be different. For hard solid food, sharp sounds such as tearing or a lot of friction sounds are often produced when chewing. For example, when eating potato chips, biscuits, nuts, etc., the sound emitted may be shorter, higher and sharper. These characteristics can be used to correspond to eating hard solid food. For liquid food, when the liquid food and air mix in the throat when drinking water and swallowing; for food rich in water, a lot of juice noise is generated when chewing. By recording the pitch, intensity, frequency, duration and other parameters of the sound generated by each food component when eaten, a database of corresponding sounds can be obtained. When an unknown eating sound is detected next time, the exact type of food eaten can be obtained by matching the characteristics of the food in the database.

[0026] In another embodiment, the person's eating is tested by vibration and sound detection. By superimposing the two tests, more accurate eating characteristics can be obtained. By analyzing these eating characteristics, data about the person's eating can be obtained, including the content of the eating, the length of the eating, the number of chews, the time / length of the swallowing in the eating, etc. Preferably, a standard value is set for each food, based on existing medical or experimental results, which data constitute the normal eating data for a normal person when eating the food. By comparing the eating data with the standard value, a difference between the two can be obtained, and the difference can be output as a selective output to the outside, either to the person or to the caregiver. In addition, when the eating data of one or more tests exceeds the standard value, an alarm message can be set for the caregiver or the person himself.

[0027] In another embodiment, continuous monitoring of eating disorders is provided. By the continuous monitoring of the eating process by the detection assembly 100, it is known that the eating behavior is occurring at this time, and when the corresponding standard value is used as a matching standard, the current eating stage can be known from the monitored eating data, and by comparing the current eating characteristic data with the standard value, it can be roughly understood that the elderly person is in the chewing stage or has a chewing physiological disorder due to injury, such as difficulty swallowing, significantly less swallowing times, no chewing, etc. Since the eating has the above serious problems, it can cause the patient to choke, be unable to eat, suffocate, etc. Continuous monitoring of food intake is very important for monitoring the incapacitated population, especially the elderly. On this basis, if a sudden stop of chewing or swallowing is found in the continuous food intake test, it is preferable to send a message to the person or the caregiver to remind the person to chew or swallow, or to remind the caregiver to pay attention to the on-site care.

[0028] Preferably, these functions can be performed by a dedicated detection assembly, which generally comprises a circuit assembly consisting of a large number of electronic components. The more important components include a computing assembly 200 for controlling, calculating, sending and receiving data, a detection assembly 100 for detecting physiological parameters of a patient during eating, and a signal processing assembly for amplifying analog signals and converting them into digital signals for power supply, and a power supply assembly. Preferably, in order to improve the portability and timeliness of the detection assembly, it can be constructed as a device that can be attached to the throat of an individual, which does not bring great difficulty to the daily life of the individual. For example, the detection module 100 for vibration detection can be set as a piezoelectric film sensor with thinner and more flexible thickness, which can be set behind a layer with buffering effect, and the buffering layer can be made of sponge, rubber and the like. Preferably, the computing assembly 200 involved in bulk computing can be set on other devices near the patient, or other devices participate in the function implementation of the whole device as the computing assembly 200, and these devices at least have the functions of data receiving and processing calculation, for example, a computer, a mobile intelligent device or the like can be used to realize the functions.

[0029] Preferably, the present application provides a scheme or process for judging the eating behavior of an individual. First, the eating process of an individual should be divided into the conceptual stage, the preparation stage, the chewing stage, the swallowing stage and the esophageal stage in chronological order. The conceptual stage refers to the stage in which a person is stimulated by the outside world or his own thoughts to eat a certain food, such as a person seeing a biscuit on the table and generating the need for food. Due to the interaction of neural information, the general physiological performance of this period is generated in the brain of the individual. The preparation stage refers to the stage in which a person moves the food he wants to eat to the mouth using the limbs after the idea of producing the food is generated, for example, a person takes a biscuit out of the packaging box and sends it to the mouth. The chewing stage refers to the process in which a person opens his mouth to bite off a part of the food and transforms it into food, for example, biting a part of the biscuit and chewing it into biscuit crumbs, the food after chewing forms food blocks, food lumps or food residues, or food lumps after one or more times of chewing are combined with saliva. The swallowing stage is the stage in which the pharynx closes the nasal cavity and the vocal tract, opens the esophagus and sends the esophagus into the esophagus after the esophagus is formed. For example, the process of swallowing the esophagus into the esophagus involves a large number of muscles in the digestive system participating in the blockage and opening of the passage. The esophageal stage refers to the process in which the food moves in the single esophagus and finally enters the stomach. This process is mainly achieved by the peristalsis of the esophagus, which promotes the esophagus to move the food into the stomach.

[0030] The detection assembly 100 described in this embodiment is used to detect the corresponding muscle or physiological characteristic trigger during at least the mastication period to the esophageal period in a direct or indirect manner, thereby obtaining the eating characteristics, eating abnormalities and / or non-eating false data generated during the removal process of the individual during eating. Specifically, the detection assembly 100 for detecting different positions, different muscle groups, and different physiological parameters transmits data to the computing assembly 200 in real time, and the computing assembly 200 realizes the above functions through logical operation, judgment, statistics, etc. The optional implementation of the detection assembly 100 is an electromyographic sensor, a vibration sensor, and a sound sensor.

[0031] In another embodiment, the detection assembly 100 involved in the present device can detect whether the patient is eating and the eating condition by visual recognition. Specifically, the present device can be configured as a structure similar to a common tool used to assist the patient in eating, at least one part of which can carry food. For example, the present embodiment adopts a structure similar to a spoon as a preferred structure of the present device. Because in the care of the elderly, the food provided is usually a liquid food that is easy for patients without teeth or with damaged teeth to eat smoothly, and the patient needs to use a spoon to eat. Additional options can be similar in shape to chopsticks, knives and forks, and other tableware. The detection assembly 100 described in the present device has at least a visual detector arranged on the detection assembly and used to detect patient eating image data and / or food type data carried on the detection assembly. These two data are at least two directions and two positions in machine vision. The image data of the patient's opening can be realized by machine vision recognition based on artificial intelligence. Through the analysis of a large number of dynamic actions of human mouth opening and the edge image of the opening of the human mouth along the axis direction of the human mouth from approximately straight to elliptical expansion, the visual detector can obtain the data of the patient's opening. By further analyzing and summarizing the number, frequency and amplitude of the opening data, the eating image data including at least the number, amplitude and frequency of the patient's eating can be obtained. Preferably, the visual detector can also be configured as a panoramic camera structure that can perform visual detection in a wide range or at least 360°. The detection assembly is configured as a structure similar to a spoon. Therefore, at least one part of the detection assembly is used to hold food when the patient is eating. Through machine recognition of the food type by the visual detector, the type of food can be obtained. Machine recognition is based on a large number of pre-set food image features, which at least include color, reflectivity, edge profile and other parameters. Preferably, the detection assembly 100 can also be provided with a touch detector for detecting whether the patient holds the detection assembly. It can be arranged at the part of the detection assembly that is contacted by the patient's hand, especially the patient's fingers, when the patient is eating. The touch detector can be realized by using a pressure sensor, a physiological potential sensor and other sensors. When the touch detector detects the trigger signal of the patient's hand contact, the above-mentioned visual detector is controlled to start working. This design can effectively reduce the working energy consumption of the detection assembly.

[0032] Preferably, in the case of selecting the open continuous eating mode, the alarm information is selectively sent to the outside world based on the continuity of the trigger signal of the touch sensor and the abnormal eating signal of the patient detected by the visual sensor. Specifically, the continuous eating mode can be selectively opened by the nursing staff, in which mode the signal of the pressure sensor continuously indicates that the patient consistently holds the detection assembly 100, and if the trigger signal disappears, it means that the detection assembly 100 is removed from the patient's hand, at which time the alarm information is sent to the outside world to remind the patient to pick up the detection assembly 100 to continue eating; on the other hand, the visual sensor continuously detects the patient's eating process based on the pre-set normal eating standard, and obtains abnormal eating signals by comparing the detection data and the pre-set eating standard, such as the patient not swallowing for a long time, the patient not chewing, the patient not correctly sending food into the mouth, etc. and sends an alarm signal to the outside world when these abnormal eating signals occur, to remind the patient himself or the nursing staff.

[0033] Unlike the above embodiments, the basic data detected in this embodiment can be based on the electromyographic action signal generated by the individual during eating. The muscle is in a resting state when it is not in action, and there is a potential difference between the inside and outside of the cell membrane. In the resting state, the membrane potential, i.e. the membrane potential, is more negative inside the membrane than outside the membrane. If the membrane potential outside the membrane is specified as 0, the membrane potential inside the membrane is generally negative, and under normal circumstances it is -90mV. At this time, the membrane potential can be referred to as the transmembrane resting potential. When the muscle produces an excitation or stimulation signal, the membrane potential of the excitation site changes rapidly, the difference between the membrane potential inside and outside the membrane rapidly decreases, and at a certain moment, the membrane potential inside the membrane becomes more positive than the membrane potential outside the membrane, and then changes rapidly again to the resting potential. The detection of the reversal of the membrane potential, i.e. the process of the resting potential transforming into the action potential and returning to the resting potential, can be determined as a characteristic of the muscle producing an action once. The electromyographic detection can be achieved by using sensors including but not limited to single-channel, double-channel to multi-channel muscle electrical modules, electromyographic sensors, electrical sensors, muscle stimulation sensors, nerve stimulation sensors, etc. These sensors can all be considered as the detection assembly 100 described in this embodiment. The detection assembly 100 transmits the electromyographic signal detected, i.e. the membrane potential change information of the muscle cells of the corresponding muscle part of the individual to the computing assembly 200.

[0034] When the computing component 200 receives the orbicularis oris muscle action signal, it is determined that the individual has produced an opening action at this time. The computing component 200 receives and processes the action signal of the masseter muscle in the next predetermined time, and processes the change in amplitude between the two as the heterogeneity value. By determining whether the heterogeneity value is greater than the first preset difference standard value, it is determined whether the individual is eating or speaking. Generally speaking, when the individual is eating, the resting membrane potential reversal frequency of the masseter muscle is relatively stable, or has a clear change trend of gradually increasing or decreasing. Accordingly, the calculation of the heterogeneity value is based on the analysis of the change amplitude of the frequency. The heterogeneity value can be obtained by using a scoring system, for example, when the frequency change exceeds a certain preset amplitude value, a score can be obtained. When the score is higher than the normal chewing range, it is considered that the heterogeneity value is high, that is, the change amplitude of the frequency is relatively chaotic. At this time, it can be considered that the individual is not producing a chewing action but a speaking action. Preferably, in order to determine whether the individual is in the chewing or speaking mode, other detection components 100 can be used to assist in confirming the judgment result, for example, by recognizing the individual's voice through voice recognition, and by detecting whether the individual has a swallowing action by detecting the swallowing-related laryngeal muscle action. Common dietary muscle electrical signal features can be summarized as at least the following muscle or muscle group excitation process, the orbicularis oris muscle performs at least one action, the masseter muscle performs at least one action, the laryngeal muscle performs at least one action, and the resting potential of the three muscles or muscle groups is changed in the order of time sequence to produce at least one excitation in the order of reversal. The computing component 200 summarizes the time, frequency, number, amplitude, etc. of the action signal of each muscle as a dietary parameter, which mainly reflects information that cannot be expressed by the action signal itself, such as chewing frequency, etc.

[0035] Preferably, the present embodiment also establishes a configuration scheme for the start time of diet detection. In the continuous monitoring mode during the individual's normal life wearing the detection device provided with the detection assembly 100, the vibration or electromyographic sensor is used to detect the opening signal of the orbicularis oris muscle and the accompanying simultaneous or delayed vocal cord vibration signal to determine the start or switching time of the accurate detection mode. The individual will inevitably control the opening of the orbicularis oris muscle when eating, and this step is a necessary step to be taken when eating. However, the action of opening the orbicularis oris muscle does not only occur during eating. The individual may speak, yawn, burp, sneeze, or even snore during sleep, which may cause the orbicularis oris muscle to open. Therefore, the present embodiment determines the accompanying simultaneous or delayed vocal cord vibration detection based on a certain preset delay time to remove these unnecessary false events when the orbicularis oris muscle opens. Among the above-mentioned false events that cause the orbicularis oris muscle to open, almost all of them will produce a certain sound, such as the sound of speaking, yawning, burping, sneezing, etc. When the computing assembly 200 detects the orbicularis oris muscle movement or after detecting the accompanying vocal cord vibration signal, it is determined that the individual is experiencing a false event, and the accurate detection mode is not started or switched. On the contrary, if no vocal cord vibration is detected, the accurate detection mode is switched. The accurate detection mode refers to a mode in which the computing assembly 200 controls the full or most of the detection assembly 100 to open and improve the detection sampling accuracy during the individual's eating process. For example, the detection assembly 100 detects the masseter muscle, laryngeal muscle, tongue muscle, etc. and improves the sampling accuracy of these detection assemblies 100, so that the sampling interval time is shortened. It can be learned that the data generated by the accurate detection mode is larger than that generated by the continuous monitoring mode, the data related to eating is more, and the individual's various situations during eating can be better reflected.The embodiment does not use a sound detector as a scheme for detecting the case where an individual emits a sound, mainly considering that using a sound sensor first requires a large amount of sound to be pre-input, and a dedicated program is needed to perform a large number of complex processes such as sound collection, impurity removal, waveform adjustment, sound matching search, etc., and 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 calculation amount undoubtedly increases the calculation burden of the calculation component 200, increases energy consumption, occupies a large part of the calculation power of the calculation component 200, and also increases the use cost. On the other hand, with the improvement of people's awareness and demand for personal privacy protection, the application of a detection component 100 with sound collection and analysis to the detection of the sound of an individual at every moment of life undoubtedly deepens the individual's concern about the detection device. Most users with ordinary knowledge cannot confirm whether the device is really intended to judge the sound patterns of speaking, yawning, sneezing, etc. rather than collecting the user's speech content and the sound around the user at all times. Especially when using the Internet to implement an artificial intelligence-based sound judgment scheme, users are worried that their speech content may be transmitted to an insecure Internet and cause information leakage. The embodiment only uses a vibration sensor to achieve the exclusion of the above-mentioned false events, without involving a microphone device, and is more easily accepted by users than a sound collection unit, without causing psychological burden to users. In addition, the detection process of the user eating involves a complex muscle excitation process, muscle coordination process, and neural signal excitation process, and involves a large number of detection components 100 and a large number of types. Therefore, the scheme of detecting only the orbicularis oris muscle and the accompanying vocal cord vibration signal in the continuous monitoring mode to determine that the individual is eating from the false events to start or switch the accurate detection mode can effectively reduce the processing workload of the calculation component 200 during non-eating, greatly reducing the amount of data generated, throughput, processing, and storage during this period, thereby reducing the energy loss of the detection device. This is very important for a mobile detection device that can be worn on the throat of an individual at any time, as it can reduce the capacity selection of the related power supply, reduce the size and weight of the power supply, and significantly improve the comfort of wearing.

[0036] Preferably, for the case that some individuals may produce meaningless openings but not sound, sound from the nasal cavity during eating, etc. may cause misjudgment, the misjudgment can be removed by the subsequent detection of the masseter muscle and the specific vocal cord vibration frequency condition of the sound from the nasal cavity.

[0037] Preferably, after switching to the accurate detection mode, the detection assembly 100 detects one or several continuous signals of the resting potential of the masseter muscle cell being reversed, which can determine that the individual is currently performing the chewing action. Within a preset time after the last detection of the above-mentioned masseter signal, if the resting potential of the throat muscle responsible for swallowing is not reversed, it is determined that the individual does not perform the swallowing action, and the calculation assembly 200 can accordingly issue an alarm instruction to the reminding assembly 300. The reminding assembly 300 can be configured as a sound, light, vibration or mixed alarm, or can be configured to a mobile or fixed device near the caregiver, which will alarm the caregiver or the individual after receiving the alarm instruction.

[0038] Preferably, the condition of determining whether the individual performs swallowing after chewing can be summarized in the diet rule, which is configured as a condition or a set of conditions for determining whether each step of the individual in the diet process is correct or meets the nursing needs. The condition set can be arbitrarily edited and updated by relevant personnel, for example, the number of chewing times that meet the nursing needs is at least 3 times, so when the individual chews only once or less than 3 times, the calculation assembly 200 will determine that it does not meet the diet rule, and then generate an alarm instruction.

[0039] Preferably, in the above-mentioned case, the calculation assembly 200 can also send a stimulation instruction to the excitation assembly 400 arranged near the throat of the individual, and the excitation assembly 400 stimulates the individual's throat in a way that stimulates the individual to produce a swallowing action after receiving the instruction. Such stimulation can be a tactile stimulation that produces pressure to the individual's throat or a micro-current stimulation that produces a small voltage to the individual's throat, which is safe to the human body. After receiving the stimulation, the individual will produce an involuntary swallowing action to achieve the effect of prompting the individual to swallow the food.

[0040] Preferably, the calculation assembly 200 can also upload the collected action signals and the processed diet parameters to the central terminal 500, which can be configured as a large server, a supervision facility of a care center, a nurse station, a medical terminal or the like. The central terminal 500 displays these data to professional caregivers to obtain a diet evaluation of the individual from manual work.

[0041] It should be noted that the above-mentioned embodiments illustrate rather than limit the application, and that those skilled in the art will be able to devise modifications which, though perhaps not specifically enumerated herein, fall within the scope of the application. Those skilled in the art will understand that the drawings, described above, and the claims, described below, are intended to be illustrative, and not restrictive, of the scope of the application. The true scope of the application is defined by the claims, and equivalents thereof. The specification and drawings are, accordingly, to be regarded as illustrative rather than restrictive. Several inventive concepts are described herein, such as "preferably," "according to one embodiment," or "optionally," each of which indicates that the corresponding paragraph discloses an independent concept, and the applicant reserves the right to file a divisional application based on each of these concepts.

Claims

1. A probe assembly as a patient eating tool, the probe assembly (100) is used to carry food and detect eating data, characterized in that, the probe assembly (100) is configured as a structure on which at least a part can carry food and can be operated by a patient to send food into the mouth, and a visual detector capable of at least 360° visual detection is arranged thereon, the visual detector is configured to obtain the kind data of the food itself carried on the probe assembly (100) and / or the patient eating image data in another direction different from the position of the food based on at least two-way visual detection, and the kind data of the food itself and the eating image data are two directions and two positions in machine vision; When the patient operates the probe assembly (100), the part of the probe assembly (100) contacted by the body part of the patient is provided with a tactile detector for detecting whether the patient holds up the probe assembly (100), the tactile detector selectively generates a trigger signal in a manner capable of detecting whether the body part of the patient contacts, and the visual detector is controlled to start working after the tactile detector detects the trigger signal of the patient's hand contact, In the case of selecting to start the continuous eating mode, the alarm information is selectively sent to the outside world based on the continuity of the trigger signal of the tactile detector and the abnormal eating signal of the patient detected by the visual detector, the continuous eating mode is selectively started by the nursing staff, the signal of the tactile detector continuously indicates that the patient always holds the probe assembly (100) in hand, if the trigger signal disappears, it means that the probe assembly (100) is separated from the patient's hand, at this time, the alarm information is sent to the outside world to remind the patient to pick up the probe assembly (100) to continue eating; the visual detector continuously detects the patient's eating process based on the preset normal eating standard, and obtains the abnormal eating signal by comparing the detection data with the preset eating standard; When the trigger signal is interrupted, the alarm information is sent to the outside world, the visual detector continuously detects the patient's eating process based on the preset normal eating standard, and obtains the abnormal eating signal by comparing the detection data with the preset eating standard, and sends the alarm information to the outside world after the abnormal eating signal is generated; The probe assembly (100) is configured as a vibration or electromyographic sensor, wherein the calculation assembly only determines whether a false event affecting the judgment of eating time is generated by judging whether a vocal cord vibration signal is generated after detecting a orbicularis oris muscle action signal, and when no false event is generated, the probe assembly and the calculation assembly are controlled to enter an accurate detection mode, which is more than the continuous monitoring mode in at least the kind, data, detection position and data amount of the probe assembly.

2. The probe assembly of claim 1, wherein, The tactile detector is arranged at the position where the patient's finger contacts the probe assembly during eating.

3. The probe assembly of claim 2, wherein, The tactile detector is configured using a pressure sensor or a physiological potential sensor.

Images