A rehabilitation system for hemiplegic patients after stroke

Abstract

The present invention relates to a rehabilitation system for hemiplegic patients due to stroke. The system includes: a memory; and at least one processor coupled to the memory. The at least one processor includes: a first processor for acquiring behavior data collected by a wearable device worn only on the affected upper and lower limbs of a monitored object; a second processor for determining a corresponding rehabilitation guidance level based on the historical rehabilitation data of the monitored object; and a third processor configured to adjust a preset rehabilitation guidance level when determining that the current mode is a rehabilitation guidance mode, based on a stage attribute association rule between at least one motion feature value extracted from the behavior data by the first processor and a preset rehabilitation guidance level to determine whether the behavior data meets the rehabilitation requirements, and transmit the action compliance information to at least one mobile digital device carried by the surrounding monitoring personnel when the rehabilitation requirements are met.

Description

[0001] The original basis of this divisional application is the patent application with application number CN202010214132X, application date March 24, 2020, and invention name “A bed leaving warning system for hemiplegic patients due to stroke”. Technical Field

[0002] The present invention relates to the field of intelligent rehabilitation technology, and in particular to a rehabilitation system for hemiplegic stroke patients. Background Art

[0003] Stroke, also known as cerebrovascular accident (CVA), is a type of vascular spasm, occlusion or rupture caused by lesions in the cerebral artery system, resulting in acute development of local cerebral circulation disorders and limb dysfunction mainly with hemiplegia, including ischemic stroke and hemorrhagic stroke. As a common neurological disease, stroke has become a major disease that seriously threatens the health and quality of life of residents due to its high incidence, mortality, recurrence and disability rates. Diabetes is an independent risk factor for ischemic stroke. Long-term high blood sugar increases the risk of stroke. The incidence of stroke in diabetic patients is four times that of normal people. High blood sugar damages cerebral blood vessels through various pathways, leading to atherosclerosis, affecting the elasticity and hardness of the vascular wall, and causing pathological changes such as vascular intimal plaque formation, stenosis, and occlusion. Due to complications such as dyslipidemia, hypertension, atherosclerosis, and increased blood viscosity, the heart, brain, and systemic tissues are damaged, and ischemic or hemorrhagic symptoms occur.

[0004] Stroke can cause damage to upper motor neurons, often leading to central paralysis and limb dysfunction, including positive and negative symptoms. Negative symptoms include the following aspects: decreased muscle strength, decreased motor control ability, and decreased limb coordination, while positive symptoms include hyperreflexia and increased spasticity. In addition, there are some special motor coordination disorder patterns: such as co-contraction and abnormal coordinated movement. The abnormal coordinated movement pattern after hemiplegia due to stroke has become a typical feature of upper limb motor dysfunction. These abnormal movement patterns hinder the completion of normal movement patterns and have a very serious impact on the patient's daily living activities. According to the statistics of the third national cause of death survey, the leading cause of death in China is stroke. More than 75% of survivors have sequelae of varying degrees, of which 40% are severely disabled, including limb, language, cognitive and other disorders. The most common and most influential one for patients is motor dysfunction, especially hemiplegia. More than 70% of patients lose their ability to live and work to varying degrees, which seriously affects their daily living activities (ADL). According to relevant research, the rates of falls in stroke patients during the acute treatment period, rehabilitation treatment stage, and returning to community life are 14% to 64.5%, 24% to 47%, and 37.5% to 73%, respectively. Falls have become a major complication with a very high incidence in stroke patients.

[0005] Falling refers to a sudden, involuntary, unintentional change in body position that causes one to fall to the ground or a lower plane. Falling can cause brain damage, soft tissue damage, fractures and dislocations, and even death. Especially for stroke patients, falling and its consequences can cause great physical and mental harm to patients and their families, seriously affecting the quality of life of patients and their families. The fear of falling causes some patients to reduce their activities or their families to restrict their activities, resulting in a decrease in the patient's self-care ability. At the same time, falling also increases the cost and length of hospital stay for patients, bringing a huge burden to the family and society. In order to effectively prevent the occurrence of falls in stroke patients, the commonly used method is to use a scale to evaluate patients, screen for high-risk groups prone to falls, take unified measures such as ensuring the safety of the patient's living environment, strengthening health education on fall prevention knowledge for patients and their families, and hanging warning signs at the head of the bed to prevent patients from falling. However, the above measures lack pertinence, and there is currently no existing technology that proposes a nursing device for preventing and intervening in the risk of falling out of bed for hemiplegic stroke patients. Therefore, to date, the risk of falls, especially the risk of falling out of bed for hemiplegic stroke patients, remains one of the main problems that plagues medical staff.

[0006] For example, a nursing system proposed in the patent document with the authorization announcement number CN208876548U with the authorization publication date of May 21, 2019, includes: a sleep monitor, a signal repeater, a server and a terminal device, the sleep monitor is connected to the signal repeater, the signal repeater is connected to the server, and the server is connected to the terminal device; the sleep monitor collects the user's micro-motion signal and sends the micro-motion signal to the signal repeater; the signal repeater sends the micro-motion signal to the server; the server receives the micro-motion signal and obtains the physiological characteristic signal according to the micro-motion signal, the server obtains the user's sleep information and activity information according to the physiological characteristic signal, and sends the sleep information and activity information to the terminal device, the activity information includes: the user is in bed or out of bed, the sleep information includes: the user's sleep state and the user's sleep quality. The nursing system obtains the user's sleep information and activity information according to the user's physiological characteristic signal, thereby realizing intelligent nursing and effectively increasing the intensity of nursing.

[0007] Another example is a patent document with a publication number of CN110853299A published on February 28, 2020, which proposes a child anti-falling-out-of-bed and bed leaving detection alarm device. A bed rail (1) is arranged on one side of the bed, and the device also includes a photoelectric emitting tube LED1 (2) installed on one side of the bed head, and a photoelectric receiving tube V1 (3) installed on the other side of the bed head. The photoelectric receiving tube V1 is used to receive the transmission signal of the photoelectric emitting tube LED1. A wireless transmitter (4) is installed on the other side of the bed head, and a wireless receiver corresponding to the wireless transmitter is installed on the child. An infrared sensor IRX (5) is installed under the bed, and the device also includes a peripheral circuit connected to the photoelectric emitting tube LED1, the photoelectric receiving tube V1, the wireless transmitter, and the infrared sensor IRX. The alarm device arranged therein can monitor the sleeping state of the child at all times, and can sound an alarm when the child accidentally falls out of the bed or leaves the bed.

[0008] In view of the poor early warning problem in the above patent documents, the following prior art proposes a technical solution with better early warning:

[0009] For example, the patent document with the authorization announcement number CN103824418B and the authorization announcement date of August 31, 2016 proposes an alarm system for monitoring when leaving the bed, and the alarm system for monitoring when leaving the bed includes two piezoelectric film sensors, a processor and an alarm. The two piezoelectric film sensors are laid on the bed in sequence along the extension direction of the long side of the bed, and synchronously and continuously collect pressure signals from the bed. The processor forms two periodic physiological characteristic signals of the human body on the bed according to the two pressure signals, and when the two periodic physiological characteristic signals are interrupted synchronously, it is judged that the human body falls out of the bed, and when the two periodic physiological characteristic signals are interrupted before and after, it is judged that the human body first sits up and then stands up to leave the bed by itself. The alarm issues an alarm prompt when falling out of the bed. Its alarm system can detect the human body falling out of the bed in time, so as to facilitate rapid remedial measures. The patent document also relates to a bed equipped with the alarm system for monitoring when leaving the bed, an alarm device for monitoring when leaving the bed used in conjunction with the alarm system for monitoring when leaving the bed, and an alarm method thereof.

[0010] The existing bed leaving alarm device proposed in the above patent document must realize the bed leaving alarm function on the basis that the mobility of the monitored person is very good, because the device determines whether the human body leaves the bed by itself, and first determines whether the human body sits up upright. The process of directly converting from a supine position to a sitting position requires a lot of abdominal strength from the monitored person. This action is impossible to complete for people with limited mobility, such as hemiplegic stroke patients. Therefore, the above device cannot be used for bed leaving alarm for people with limited mobility, such as hemiplegic stroke patients, and will only increase the false alarm rate.

[0011] For another example, a patent document with a publication number of CN109757928A published on May 17, 2019, proposes a fall-proof smart baby bed, which includes at least a baby bed body and a baby bed frame fixed to the baby bed body, and the fall-proof smart baby bed also includes a baby climbing detection module, a plurality of hand behavior detection modules and a plurality of foot behavior detection modules. When at least one specific event occurs, at least one of the plurality of foot behavior detection modules and at least one of the plurality of hand behavior detection modules obtain and record the pressure change information of the fall-proof smart baby bed when the specific event occurs, and based on the received pressure change information, the baby climbing prediction analysis is performed and the matching baby climbing prediction result and / or baby climbing risk level are output to the central processing module, and the central processing module issues a warning prompt to the smart terminal operated by the guardian based on the received baby climbing prediction result and / or baby climbing risk level and / or issues a first sound and light information through the sound and light prompt module.

[0012] The bed leaving alarm device proposed in the above patent document can provide a better early warning effect by utilizing the interaction between the user's hands and feet and the pressure sensors arranged in different areas on the bed body. However, it is only suitable for infants and young children who need to climb over the bed frame before leaving the bed. It is also not suitable for elderly patients with limited mobility, patients after knee or hip replacement, patients with anemia or weakness after surgery, or hemiplegic patients due to stroke, who do not listen to the advice of medical staff, get out of bed without authorization, and do not ask family members to accompany them, resulting in accidental leaving the bed or falling out of bed.

[0013] In view of the problem that the above-mentioned prior art documents have good early warning but too limited scope of application, the prior art proposes to abandon non-contact monitoring devices with limited behavior monitoring scope (such as the mattress with multiple sensors listed above) and adopt wearable monitoring devices with better behavior monitoring capabilities, for example:

[0014] A clinical early warning system proposed in the patent document with publication number CN109966088A, which was published on July 5, 2019, includes an electronic tag, a tag reading module, a ward routing terminal, a cloud platform, a monitoring terminal and a nurse station terminal; the electronic tag includes a foot ring set on the patient's foot, and an RFID electronic tag is set in the foot ring; the tag reading module is set under the bed, including a shielding cover, the shielding cover is provided with an opening toward the side of the bed, and the shielding cover is provided with a reading unit for wirelessly reading the RFID electronic tag. Compared with the prior art, in the clinical early warning system, when the tag reading module under the bed reads the electronic tag, it indicates that the patient has gotten out of bed, and the ward routing terminal transmits the information to the cloud platform, which processes the information and transmits it to the accompanying terminal and the nurse station terminal respectively, so as to monitor the distance between the foot ring and the bed in real time, and warn of the risk of falling in real time, and send warning information to the accompanying personnel and the nurse station in time.

[0015] When such technical solutions do not take actual applications into consideration, the monitored person may trigger the alarm of the device simply by extending his feet out of the bed, but in reality the monitored person has no intention of getting out of bed and has no tendency to get out of bed on his own.

[0016] In addition, on the one hand, the technical personnel understood by the applicant and the examination department are bound to be different; on the other hand, the inventor 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 at any time in accordance with the relevant provisions of the examination guidelines. Summary of the invention

[0017] The present invention relates to a rehabilitation system for hemiplegic patients due to stroke, the system comprising: a memory; and at least one processor coupled to the memory, the at least one processor comprising: a first processor for acquiring behavior data collected by a wearable device worn only on the affected upper and lower limbs of a monitored object; a second processor for determining a corresponding rehabilitation guidance level according to the historical rehabilitation data of the monitored object; and a third processor configured to determine whether the behavior data meets the rehabilitation requirements by adjusting the preset rehabilitation guidance level based on the phase attribute association rule between at least one motion feature value extracted from the behavior data by the first processor and the preset rehabilitation guidance level when the current mode is determined to be the rehabilitation guidance mode, and transmit the action reaching standard information to at least one mobile digital device carried by the surrounding monitoring personnel when the rehabilitation requirements are met. The first processor, the second processor and the third processor are all arranged in the wearable device.

[0018] According to a preferred embodiment, at least one processor is further configured to preset a motion feature change trend corresponding to at least one specific posture for indicating the motion feature change trend that may occur in the specific posture.

[0019] According to a preferred embodiment, the fourth processor determines whether the current mode is converted to the rehabilitation guidance mode and instructs the fifth processor in the system for presetting the motion characteristic change trend to increase its trigger threshold when converted to the rehabilitation guidance mode, or instructs the fifth processor in the system for presetting the motion characteristic change trend to lower its trigger threshold when converted from the rehabilitation guidance mode to the bed exit warning mode.

[0020] According to a preferred embodiment, the movement characteristic change trends include: a first mode with an extremely low risk factor, in which the limbs are slightly lifted, placed, turned, etc. and the posture of the monitored object does not change; a second mode with a medium risk factor, in which the monitored object attempts to turn over or tries to get up and the independence of the monitored object is poor; a third mode with a relatively high risk factor, in which the monitored object attempts to turn over or tries to get up and the independence of the monitored object is good; and a fourth mode with a relatively high risk factor, in which the limbs are slightly lifted, placed, turned, etc. and the posture of the monitored object changes; to be distinguished from different body positions, each body position corresponds to a predefined first mode, second mode, third mode and fourth mode. The stage attribute association rule determines the movement feature change trend corresponding to the first movement feature value and the movement feature change trend corresponding to the second movement feature value based on the degree of movement dysfunction of the patient in the current stage; when the movement dysfunction assessment result determined by at least one mobile digital device operated by the guardian is that the independence of the monitored object is poor, the preset upper limb movement amplitude, upper limb preset movement amplitude, lower limb preset movement amplitude and lower limb preset movement amplitude corresponding to this situation are determined, and based on the determined amplitude / trigger threshold, the boundaries between the first mode, the second mode, the third mode and the fourth mode are distinguished, and the system can switch between the bed leaving warning mode and the rehabilitation guidance mode by overall raising or lowering the amplitude / trigger threshold of the four modes.

[0021] According to a preferred embodiment, for a patient in the healthy side lying position without monitoring, the fourth mode corresponding to the patient's above-mentioned body position is at least preset as follows: if the wearable device worn on the affected upper limb has collected behavioral data that meets the preset motion range of the upper limb, then when it is monitored that the behavioral data collected by the wearable device on the affected lower limb meets the preset motion range of the lower limb, and when it is monitored that the patient changes from lying on the side to lying on the back or from lying on the back to lying on the side, it is determined that the current patient condition is the fourth mode with a higher risk factor of small-scale lifting, lowering, and flipping of the limbs and changes in the posture of the monitored object.

[0022] According to a preferred embodiment, for a patient in the healthy side lying position without monitoring, the third mode corresponding to the patient in the above-mentioned position is at least preset as follows: if the wearable device worn on the affected upper limb has collected behavioral data that meets the preset motion range of the upper limb, and the monitored subject has good independence, then when it is monitored that the behavioral data collected by the wearable device on the affected lower limb meets the preset motion range of the lower limb, or when it is monitored that the pressure distribution on the right side of the mattress gradually decreases and the pressure distribution on the left side gradually increases, since the monitored subject has good independence and is not difficult to turn over or get up, the risk factor is relatively high, and it is determined to be the third mode.

[0023] According to a preferred embodiment, for a patient in the healthy side lying position without monitoring, the second mode corresponding to the patient in the above-mentioned position is at least preset as follows: if the wearable device worn on the affected upper limb has collected behavioral data that meets the preset motion range of the upper limb, and the monitored subject has poor independence, then when the behavioral data collected by the wearable device on the affected lower limb is monitored to meet the preset motion range of the lower limb, or when the pressure distribution on the right side of the mattress is monitored to gradually decrease and the pressure distribution on the left side is gradually increased, the patient is usually trying to turn over or trying to get up, but because the monitored subject has poor independence and turning over is difficult, the risk factor is medium, and it is determined to be the second mode.

[0024] According to a preferred embodiment, for a patient in the healthy side lying position without monitoring, the first mode corresponding to the patient's above-mentioned position is at least preset as: when the wearable device worn on the affected upper limb collects behavioral data that meets the preset movement range of the upper limb, or when the wearable device monitoring the affected lower limb collects behavioral data that meets the preset movement range of the lower limb, and the patient's posture has not changed, the patient only needs to slightly adjust the current posture and the risk factor is low, and it is determined to be the first mode.

[0025] The present invention also proposes a bed leaving warning system for hemiplegic patients due to stroke, and is particularly suitable for a bed leaving warning system for monitored subjects who have motor dysfunction in both upper and lower limbs on the same side and who have a distinction between affected upper and lower limbs and healthy upper and lower limbs.

[0026] The bed leaving warning system of the present invention is proposed to solve the shortcomings of the bed leaving alarm device proposed in the prior art for realizing the bed leaving alarm function. The shortcomings of the prior art are: for the high incidence and high harm of falling problems, especially for the self-leaving and falling problems of patients with unilateral hemiplegia due to stroke, the bed leaving alarm device proposed in the prior art for realizing the bed leaving alarm function can be roughly divided into solutions based on non-contact monitoring equipment or wearable monitoring equipment. First, as for the solutions based on non-contact monitoring equipment such as smart mattresses proposed in the prior art, such as the patent document with authorization announcement number CN103824418B, the judgment of whether the human body leaves the bed by itself in such solutions is equivalent to judging whether the human body sits up upright. In fact, the action of sitting up upright needs to be directly converted from a supine position to a sitting position, and this process requires a lot of abdominal strength for the monitored object. Therefore, the realization of the bed leaving alarm function in such solutions is based on the very good mobility of the monitored object. For people with limited mobility, such as patients with hemiplegia due to stroke, this action is very difficult or even impossible to complete. Therefore, such solutions cannot be applied to the bed leaving alarm of people with limited mobility, such as stroke hemiplegia patients, and will only increase the false alarm rate. Secondly, for wearable monitoring devices, such as the solution based on the foot ring with an RFID electronic tag, proposed in the patent document with the publication number CN109966088A, the realization of the bed leaving alarm function in such solutions is achieved by monitoring whether the foot ring moves out of the area outside the bed body, and it is determined that as long as the foot ring moves out of the bed body, it is out of bed and an alarm is issued. However, such solutions do not take into account the actual application. The monitored object may only extend his feet out of the outer edge of the bed, but has not left the bed and has no tendency to leave the bed by himself. However, in this case, the alarm of the device will still be triggered, and the false alarm rate is high, which increases the workload of the nursing staff. At the same time, because the foot ring is set on the foot with high mobility, not only is the discomfort strong when wearing it, but the monitoring purpose is very obvious, which is easy to cause the patient to be disgusted and remove it by himself, and it is impossible to monitor whether he leaves the bed without authorization. Especially for subjects such as unilateral hemiplegia caused by stroke, who have motor dysfunction in both the upper and lower limbs on the affected side and severe limb sensory impairment, patients often need the support of the healthy side to move the affected side. They are highly dependent on manual assisted care and are very averse to restricted movement.

[0027] In view of the shortcomings of the bed leaving alarm device in the above-mentioned prior art, the present invention proposes a bed leaving warning system for hemiplegic stroke patients, the system comprising: a collection module, which is used to obtain the behavior data collected by the wearable device only worn on the affected upper and lower limbs of the monitored object; a risk control level classification module, which is used to determine the corresponding risk control level according to the historical risk control data of the monitored object, wherein the monitored object refers to a group of people with motor dysfunction in both upper and lower limbs on the same side and distinguishing between the affected upper and lower limbs and the healthy upper and lower limbs, the system also comprises: an early warning information processor, which is configured to determine whether its behavior data triggers an early warning based on at least one movement feature value extracted from the behavior data collected by the collection module from at least one wearable device worn on the affected upper and lower limbs of the monitored object and the stage attribute association rule corresponding to the motor dysfunction assessment result determined by at least one mobile digital device operated by the guardian and the preset risk control level. Preferably, when the early warning is triggered, the early warning information is transmitted to at least one mobile digital device carried by the guardian in the surrounding area.

[0028] The present application sets a wearable device that is only worn on the upper and lower limbs on the side with motor dysfunction. Since the affected limb has severe sensory impairment and low perception ability, it has a better acceptance of the wearable device and retains the good unrestricted activity ability of the healthy side. At the same time, the activity level of the affected limb is low, which greatly reduces the amount of data that needs to be processed. Based on this, the early warning information processor in the present invention is based on different risk control levels of different patients and different degrees of motor dysfunction on the side of hemiplegia, and defines several monitoring modes for indicating the early warning risk and different early warning risks from each other, so that the early warning information processor can compare the patient's current behavior data with several preset monitoring modes, and judge whether an early warning is needed based on the comparison results and the preset stage attribute association rules between the upper and lower limbs on the affected side, thereby improving the timeliness of the early warning to notify the guardian to provide timely care or assistance to the monitored object before the danger occurs.

[0029] According to a preferred embodiment, the motion feature value includes at least a first motion feature value in the behavioral data collected by a first wearable device worn on the affected upper limb of the monitored object, and a second motion feature value in the behavioral data collected by a second wearable device worn on the affected lower limb of the monitored object, wherein the acquisition module pre-stores a stage attribute association rule between the first motion feature value and the second motion feature value for indicating the motion feature change trends corresponding to the first motion feature value and the second motion feature value respectively.

[0030] The bed leaving warning system in the present invention is based on the behavioral data of the upper and lower limbs on the affected side collected by the wearable device, and the placement posture and / or posture change trend determined by the placement posture detection module, and compares it with the movement feature change trend preset in the behavior rule module, so that the system can adapt to the special requirements of the bed rest position of hemiplegic stroke patients to monitor their bed rest behavior, and the warning information processor determines whether a warning is needed based on the comparison result and the preset stage attribute association rule between the upper and lower limbs on the affected side, thereby improving the timeliness of the warning to notify the guardian to provide timely care or assistance to the monitored object before the danger occurs.

[0031] The following is a unified list of multiple modules included in the bed leaving warning system of the present invention: the system includes a collection module for collecting behavioral data of the monitored object, a risk control level classification module for determining the risk control level, an early warning information processor for judging the triggering of the early warning, a placement posture detection module for determining the current placement posture, a behavior rule module for presetting the trend of movement feature changes, a wearable device worn on the affected side of the monitored object, a mobile digital device carried by the guardian, and a mattress with a processor and a sensor. In addition, a pressure sensor array for collecting patient pressure changes and a processor connected to the pressure sensor array are arranged on the mattress. The above-mentioned devices mentioned in the present invention can all be computer processors, Figure 1 A simplified logic flow chart of a bed-leaving warning system for hemiplegic stroke patients is shown, and the operations of the logic flow chart represent a series of operations that can be implemented by hardware, computer instructions, or a combination thereof. In the context of computer instructions, the operations represent computer executable instructions stored on one or more computer-readable storage media, and the computer executable instructions perform the operations when executed by one or more computer processors. In general, computer executable instructions include routines, programs, objects, components, data structures, etc. that perform specific functions or implement specific data types. The order in which the operations are described is not intended to be interpreted as limiting, and any number of the described operations can be combined in any order and / or in parallel to implement the process. In addition, the data transmission process between modules can be performed under the control of one or more computer systems configured with executable instructions, and can be implemented as code (e.g., executable instructions, one or more computer programs, or one or more applications), which is executed uniformly on one or more computer processors, or implemented by hardware, or by a combination of the above two. The code can be stored on a computer-readable storage medium, for example in the form of a computer program, which includes multiple instructions that can be executed by one or more processors. The computer-readable storage medium can be non-temporary. In some embodiments, Figure 1The data transmission process between the modules can be stored in the memory of the computer processor and executed by the computer processor.

[0032] The features described may be implemented in digital electronic circuitry, or in computer hardware, firmware, software, or a combination thereof. The apparatus may be implemented in a computer program product executed by a programmable processor, the computer program product being tangibly embodied in an information carrier, such as in a machine-readable storage device; and the method steps may be performed by a programmable processor executing a program of instructions to perform the functions of the described implementation by operating on input data and producing output. The features described may advantageously be implemented in one or more computer programs executable on a programmable system, the programmable system comprising: at least one programmable processor connected to receive data and instructions from and transmit data and instructions to a data storage system; at least one input device; and at least one output device. A computer program is a set of instructions that can be used directly or indirectly for a computer to perform an activity or bring about a result. A computer program may be written in any form of programming language, including compiled or interpreted languages, and the computer program may be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0033] Suitable processors for executing instruction programs include, by way of example, both general-purpose and special-purpose microprocessors, and a single processor or one of multiple processors of any kind of computer. In general, the processor will receive instructions and data from a read-only memory or a random access memory or both. The essential elements of a computer are a processor for executing instructions and one or more memories for storing instructions and data. In general, a computer will also include one or more mass storage devices for storing data files, or operatively connected to communicate with them; such devices include magnetic disks, such as internal hard disks and removable disks; magneto-optical disks; and optical disks. Storage devices suitable for tangibly embodying computer program instructions and data include all forms of non-volatile memory, including, by way of example: semiconductor memory devices, such as EPROM, EEPROM, and flash memory devices; magnetic disks, such as internal hard disks and removable disks; magneto-optical disks; and CD-ROM and DVD-ROM disks. The processor and memory may be supplemented by or incorporated into an ASIC (Application Specific Integrated Circuit).

[0034] According to a preferred embodiment, the system further includes a behavior rule module, which is connected to the placement posture detection module and the risk control level classification module respectively, and is used to preset the motion feature change trend that corresponds to at least one placement posture and indicates the motion feature change trend that may occur in the placement posture. In the present application, the behavior rule module has preset different motion feature change trends corresponding to the several placement postures determined by the placement posture detection module, which can distinguish the dangerous actions or behaviors that may occur in each placement posture, and directly compares with the preset motion feature change trend to determine whether the action or behavior has potential risks, thereby avoiding the problem of long data processing feedback time caused by analyzing and processing a large amount of data and then judging in the prior art.

[0035] According to a preferred embodiment, the motion feature change trend includes at least a motion feature change trend for a monitored object that has received guidance on body position changes and a motion feature change trend for a monitored object that has not yet received guidance on body position changes. The guidance on body position changes referred to here mainly refers to the monitored object being taught the correct and appropriate posture for turning over or getting up by the monitored object after the object has been hospitalized for a period of time. Relatively speaking, not having received guidance on body position changes mainly refers to the monitored object who has just been hospitalized or has not been hospitalized for a long time and has not yet received guidance on turning over or getting up. The two motion feature change trends distinguished in this way can distinguish the actions or behaviors that are more likely to occur in the current monitored object when out of monitoring, can effectively improve the speed of data comparison processing, and further improve the accuracy and effectiveness of monitoring of the monitored object.

[0036] According to a preferred embodiment, the degree of motor dysfunction of the monitored object with the highest risk control level determined by the risk control level classification module is not completely lower than that of the monitored objects at other risk control levels. For some monitored objects with better independence, the degree of motor dysfunction on the hemiplegic side is relatively small, and it is not difficult to turn over or get up, but they often overestimate that they can get out of bed independently or turn over at will, causing the risk of falling out of bed for such people to be higher than that of people with poor independence. In this application, the risk control level classification module determines the risk control level of the monitored object based on the degree of motor dysfunction determined by the mobile digital device, and can then provide different degrees of warning for the behaviors or actions of different monitored objects, and medical staff can also immediately know the severity of the warning event when receiving the warning.

[0037] According to a preferred embodiment, the system further comprises a posture detection module, which is connected to the acquisition module and is used to determine the relative position relationship between the affected side of the monitored object and the mattress and generate the current posture of the monitored object based on the relative position relationship and at least part of the motion characteristic value. Due to the particularity of the posture of hemiplegic patients, hemiplegic patients often need to use such as Figure 2 The multiple pillows shown are used to assist the patient in maintaining a good limb position, but since some limbs are indirectly pressed on the mattress through pillows, there is a problem that the pressure distributions corresponding to the two placement postures may be very similar and different postures cannot be distinguished. In the present invention, a placement posture detection module is set up. The module can predict and determine the pressure distribution data of the patient's good limb position on the mattress based on the predetermined relative position relationship between the side where the patient's affected limb is located and the mattress. In this way, the current placement posture or posture change can be accurately analyzed by comparing the current pressure distribution data with the predicted pressure distribution data.

[0038] According to a preferred embodiment, the mattress is divided into a hip bone area and a shoulder and elbow area which are spaced apart from each other and each of which is provided with at least one sensor, and the bed body corresponding to the hip bone area has a higher resilience deformation capacity than that corresponding to the shoulder and elbow area. Since the bed body corresponding to the hip bone area has a better resilience deformation capacity, the pressure on the hip bone, sacrum and other limbs of the hemiplegic stroke patient when lying on his side or lying down can be reduced, which is beneficial to the improvement of blood circulation and the reduction of damage to the affected side or the healthy side caused by squeezing when lying on the side.

[0039] According to a preferred embodiment, the distribution density of sensors arranged on the shoulder and elbow region is at least greater than the distribution density of sensors arranged on the hip bone region. The mattress suitable for monitored subjects with motor dysfunction in both upper and lower limbs on the same side of the present application improves the effectiveness and sensitivity of data acquisition while ensuring the comfort of the monitored subjects, at least in terms of sensor distribution and the resilience and deformation capability of the bed body.

[0040] The system proposed in this application not only plays the role of the above-mentioned bed leaving warning, but also can be used as a rehabilitation system suitable for hemiplegic stroke patients by simply adjusting the parameters in the above-mentioned bed leaving warning system. The rehabilitation system at least includes a first processor (acquisition module), a second processor (risk control level division module), a third processor (warning information processor), a fourth processor, a fifth processor, a wearable device and a mobile digital device. The rehabilitation system is provided with a fourth processor for entering the rehabilitation guidance mode when the information of the patient determined by the medical staff for rehabilitation training is collected, and the fourth processor instructs the fifth processor (behavior rule module) used in the system for presetting the trend of movement characteristics to increase its trigger threshold. Several rehabilitation guidance levels correspond to at least one trigger threshold, and the rehabilitation guidance level corresponding to the degree of motor dysfunction of the upper and lower limbs on the affected side is selected accordingly, so that the system is transformed into a rehabilitation guidance mode that will only issue a notification when the patient's action reaches the standard, and the patient can clarify the rehabilitation training goals and achieve better rehabilitation effects. At the same time, due to the realization of the rehabilitation auxiliary role, the patient's subjective willingness to wear the wearable device is enhanced, and the patient is willing to wear it for a long time, and will not take it off because of the aversion to its monitoring purpose. Therefore, on the contrary, the system can also better realize the bed leaving warning monitoring function during the non-rehabilitation training period.

[0041] A rehabilitation system for hemiplegic patients after stroke, the rehabilitation system comprising: a memory; and at least one processor coupled to the memory, the at least one processor comprising: a first processor, which acquires behavior data collected by a wearable device worn only on the affected upper and lower limbs of the monitored object; a second processor, which is used to determine the corresponding rehabilitation guidance level according to the historical rehabilitation data of the monitored object; a third processor, which is configured to adjust the preset rehabilitation guidance level in a manner to determine whether the behavior data meets the rehabilitation requirements based on the stage attribute association rule between at least one motion feature value extracted from the behavior data by the first processor and the preset rehabilitation guidance level when the current mode is determined to be the rehabilitation guidance mode, and transmit the action achievement information to at least one mobile digital device carried by the surrounding monitoring personnel when the rehabilitation requirements are met, wherein the first processor, the second processor and the third processor are all arranged in the wearable device. According to a preferred embodiment, at least one processor is also configured to preset a motion feature change trend that corresponds to at least one specific posture and is used to indicate the possible occurrence of the specific posture in the specific posture. Preferably, the fourth processor determines whether the current mode is converted to the rehabilitation guidance mode and instructs the fifth processor (behavior rule module) in the system for presetting the motion characteristic change trend to increase its trigger threshold when converted to the rehabilitation guidance mode, or instructs the fifth processor (behavior rule module) in the system for presetting the motion characteristic change trend to lower its trigger threshold when converted from the rehabilitation guidance mode to the bed exit warning mode. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a simplified structural connection relationship diagram of the bed leaving warning system for hemiplegic stroke patients proposed by the present invention;

[0043] Figure 2 is a simplified schematic diagram of the body position of the monitored object mentioned in the present invention; and

[0044] Figure 3 It is a simplified side view structural schematic diagram of a mattress in the bed leaving warning system proposed by the present invention.

[0045] Reference numerals list

[0046] 1: Wearable device 2: Collection module

[0047] 3: Risk control level classification module 4: Warning information processor

[0048] 5: Behavior rules module 6: Posture detection module

[0049] 7: Mobile digital devices 8: Mattresses

[0050] 9: Hip area 10: Shoulder and elbow area

[0051] 1a: First wearable device 1b: Second wearable device DETAILED DESCRIPTION

[0052] The present invention is described in detail below with reference to the accompanying drawings.

[0053] The relevant terms and their explanations of the present invention:

[0054] Wearable device 1: Wearable device 1 refers to a portable accessory with partial computing functions that can be connected to mobile phones and various terminals. It can include watches supported by the wrist, including watches and wristbands, shoes supported by the feet, including shoes, socks or other leg-worn products, and glasses supported by the head, including glasses, helmets, headbands and other product forms. The wearable device 1 referred to in the present invention mainly refers to a device in the form of a watch product supported by the wrist. Preferably, the wearable device 1 referred to in the present invention can also refer to a device in the form of a watch product supported by the wrist and ankle. Among them, the wearing side is mainly the affected side of the monitored object.

[0055] The affected side of the monitored subject: The affected side of the monitored subject refers to the upper and lower limbs on the side of the monitored subject where motor dysfunction occurs. Motor dysfunction is the most prominent problem after a stroke. Different lesions can cause various disorders. The most typical motor disorder is hemiplegia. In severe cases, the paralyzed limbs cannot move autonomously, lose sensation, and have no muscle strength. In milder cases, the muscle strength of the paralyzed side is reduced, making it difficult to move.

[0056] Behavior data: Behavior data includes information about the posture or posture change of the monitored person determined based on pressure data collected by sensors arranged on the mattress 8. Behavior data also includes information about the posture or posture change of the monitored person determined based on motion data collected by the wearable device 1 worn by the monitored person. Behavior data also includes the side of the monitored person's affected limb determined by the monitored person's historical case data, and thus can also determine the relative position relationship between the monitored person's affected side and the mattress 8.

[0057] Relative position relationship: The relative position relationship between the affected side of the monitored subject and the mattress 8 refers to the left side and the right side of the mattress 8 on which the two sides for the monitored subject to get on and off the bed are pre-set, respectively. When the monitored subject lies on his back on the mattress 8, the affected side of the monitored subject is located on the left side of the mattress 8 or the right side of the mattress 8.

[0058] Historical risk control data of the monitored object: The historical data mentioned here refers to the basic information and medical information of the monitored object stored in the hospital information system. The historical data mentioned in the present invention also includes the scale evaluation results and historical behavior data (such as the number of times the monitored object got out of bed on his own and the number of times he fell after getting out of bed) input by the guardian.

[0059] Risk control level: Risk control level refers to the risk control level of the patient leaving bed corresponding to the patient, which is determined based on the patient's historical risk control data. It includes Level I (basically independent but overestimating their ability to move), Level II (conditionally independent or very slightly dependent), Level III (moderate or severe dependence), and Level IV (extremely severe or completely dependent). Risk control level is used to provide different degrees of restricted movement risk control for patients with different degrees of independence. A single risk control level includes the degree of restriction of unilateral limb movement in a certain posture.

[0060] Motor dysfunction in both upper and lower limbs on the same side: The monitored subjects referred to in the present invention are mainly people with motor dysfunction in both upper and lower limbs on the same side, that is, both upper and lower limbs on the same side have motor dysfunction that makes them difficult to move. Such patients spend most of their time in bed. Usually, in order to avoid complications such as hypertonia and loss of mobility, the monitored subjects need to be assisted by the monitoring staff to adjust to a suitable bed resting position. The position will be directly related to the prognosis of rehabilitation.

[0061] Motion characteristic value: Motion characteristic value refers to the posture, pressure distribution, and limb movement data (parameters such as the amplitude of movement of the upper or lower limb movements) of the monitored object in bed determined by the behavioral data. Specifically, the motion characteristic value includes the first motion characteristic value extracted by the early warning information processor from the behavioral data collected by the first wearable device 1a worn on the affected upper limb of the monitored object, and the second motion characteristic value extracted by the early warning information processor from the behavioral data collected by the second wearable device 1b worn on the affected lower limb of the monitored object. The first motion characteristic value and the second motion characteristic value mentioned here mainly refer to the content contained in the limb movement data in the motion characteristic value, that is, the motion characteristic value can also include the third motion characteristic value corresponding to the posture, the fourth... Preferably, there is a stage attribute association rule between the first motion characteristic value and the second motion characteristic value.

[0062] Mobile digital device 7: Mobile digital device 7 refers to battery-powered handheld devices, such as mobile phones, computers, portable media players, remote control devices, personal digital assistants, paging devices, GPS positioning devices, PDAs, calculators, portable medical devices, etc. In the present invention, the mobile digital device 7 is mainly carried and operated by the guardian. It mainly serves to inform the guardian of the warning information in a visual, auditory, tactile, etc. way, and can display at least one scale on its operation interface to interact with the guardian, obtain the input information entered by the guardian, and generate a scale evaluation result based on the input information.

[0063] Scale: The scale, or functional comprehensive assessment scale, is an objective, effective and accurate method to assess the patient's functional impairment. It is used by guardians to understand the functional status of the monitored person, formulate rehabilitation treatment plans and evaluate the efficacy. It can include: functional comprehensive assessment (FCA) scale, which is used to evaluate the patient's comprehensive function; mini-mental state examination (MMSE), which uses the modified Chinese version of the scale compiled by Folstein et al. in 1975, with 1 point for each correct answer and a total score range of 0 to 30 points, used to evaluate the patient's cognitive function; Barthel index (BI), which uses the scale compiled by Barthel et al. in 1965, used to evaluate the patient's daily living activities (ADL) ability; the quality of life index (QLI), which evaluates the patient's activities, daily life, health, support and prospects in five items, with a maximum score of 2 points and a minimum score of 0 points for each item, with a total score range of 0 to 10 points, used to evaluate the patient's quality of life; Functional Independent Measure (FIM) scale, which is an effective testing tool for evaluating the daily living activities ability of disabled patients. The motor dysfunction assessment results determined by at least one mobile digital device 7 operated by the guardian may include Level I (basic independence but overestimation of one's own mobility), Level II (conditional independence or extremely mild dependence), Level III (moderate or severe dependence), and Level IV (extremely severe or complete dependence). The motor dysfunction assessment results may be expressed as one of the two results of better or worse independence of the monitored subject (mainly refers to the degree of help required for the patient to complete a certain prescribed action or project). The independence of the monitored subject may be the degree of help required for the patient to complete the action of half-standing up or getting out of bed. The better independence of the monitored subject includes Level I (basic independence but overestimation of one's own mobility) and Level II (conditional independence or extremely mild dependence), and the poor independence of the monitored subject includes Level III (moderate or severe dependence) and Level IV (extremely severe or complete dependence).

[0064] Posture: also known as body posture or good limb position, refers to a therapeutic posture designed to prevent or counter the occurrence of spastic postures, maintain the shoulder joint and induce separation movements in the early stage. This point is the special requirement for the bed rest position of patients with hemiplegia due to stroke. The typical posture of patients with hemiplegia due to stroke is characterized by the sinking and retraction of the shoulders, flexion of the elbows, pronation of the forearms, palmar flexion of the wrists, and flexion of the fingers; external rotation of the lower limbs, extension of the hips and knees, and foot drop and inversion. Paying attention to and maintaining the correct posture in bed in the early stage can help prevent or reduce the occurrence and aggravation of the above-mentioned spastic postures. Such as Figure 2 As shown, the following positions are usually selected: lying on the affected side, lying on the healthy side, and lying on the back. "Lying on the affected side" means that the affected side is at the bottom and the healthy side is at the top. The head is comfortably supported with a pillow, the affected upper limb is extended forward, the shoulder is forward, and the inner edge of the scapula is made to rest flat against the chest wall. The upper arm is extended forward to avoid compression and retraction of the shoulder joint. The elbow joint is extended, the forearm is supinated, the fingers are open, and the palm is facing up. The healthy upper limb cannot be placed in front of the body, because it will drive the entire trunk forward and cause the scapula to retract. The affected lower limb is at the back, the affected hip joint is slightly extended, the knee joint is slightly flexed, and the sole of the foot is pushed against a support. "Lying on the healthy side" means that the healthy side is at the bottom and the affected side is at the top, and the pillow for the head should not be too high. Place a pillow under the affected upper limb, flex the shoulder 90° to 130°, extend the elbow and wrist, pronate the forearm, dorsiflex the wrist joint, pronate the affected pelvis, and place the hip and knee joints in a natural semi-flexed position on the pillow; keep the affected foot and calf as vertical as possible, and be careful not to turn the foot inward and hang on the edge of the pillow; place a pillow behind the body for support, which is conducive to body relaxation. Place the healthy lower limb flat on the bed, slightly extend the hip, and slightly bend the knee. "Supine position" means placing a pillow under the head, but not too high, with the face facing the affected side; place a pillow slightly higher than the trunk on the affected shoulder, and place the extended upper limb on the pillow to prevent the scapula from retracting; supinate the forearm, with the palm facing up, and the fingers extended and open; place a pillow under the affected hip and thigh to prevent the affected pelvis from retracting.

[0065] Movement characteristic change trend: The movement characteristic change trend that may occur under this placement posture refers to the behavior that the monitored object may exhibit without supervision after the guardian adjusts the posture of the monitored object to a suitable placement posture. It includes the first mode with extremely low risk factor, such as small lifting and turning of the limbs, and the placement posture of the monitored object does not change, the second mode with medium risk factor, such as trying to turn over or trying to get up, and the independence of the monitored object is poor, the third mode with a high risk factor, such as trying to turn over or trying to get up, and the independence of the monitored object is good, and the fourth mode with a high risk factor, such as small lifting and turning of the limbs, and the placement posture of the monitored object changes. Different from different body placement postures, each body placement posture corresponds to a predefined first mode, second mode, third mode and fourth mode.

[0066] Phase attribute association rule: The phase attribute association rule corresponds to the motor dysfunction assessment result determined by at least one mobile digital device 7 operated by the guardian. The phase attribute association rule mentioned here refers to the motor feature change trend corresponding to the first motion feature value and the motor feature change trend corresponding to the second motion feature value determined according to the motor dysfunction degree determined by the current motor dysfunction assessment result of the user, which is different from the motor dysfunction degree of the patient at the current stage. The motion feature change trend mentioned here mainly refers to the behavior that may occur when the monitored object is kept in a certain posture without supervision, which is mainly used to distinguish the boundaries / trigger thresholds between the first mode, the second mode, the third mode and the fourth mode. Further preferably, in the "warning information processor 4 determines whether its behavior data triggers a warning based on the preset risk control level and the phase attribute association rule corresponding to the motor dysfunction assessment result determined by at least one mobile digital device 7 operated by the guardian" mentioned in the present invention, the phase attribute association rule and the risk control level are only part of the data to be processed for the warning information processor 4 to determine whether the behavior data triggers a warning.

[0067] The following is an example of the stage attribute association rule: when the motor dysfunction assessment result determined by at least one mobile digital device 7 operated by the guardian is that the independence of the monitored object is poor, the preset upper limb movement amplitude (upper limb movement amplitude trigger threshold), upper limb preset movement amplitude (upper limb movement amplitude trigger threshold), lower limb preset movement amplitude (lower limb movement amplitude trigger threshold) and lower limb preset movement amplitude (lower limb movement amplitude trigger threshold) corresponding to this situation are determined, and based on the determined above amplitude / trigger threshold, the boundaries between the first mode, the second mode, the third mode and the fourth mode can be distinguished, and the system can be switched between the bed leaving warning mode and the rehabilitation guidance mode by raising or lowering the above amplitude / trigger threshold of the four modes as a whole. Different from the corresponding reminder methods in the bed leaving warning mode and the rehabilitation guidance mode, the bed leaving warning mode can be a phone ringing-like continuous ringing that must be turned off manually by the medical staff, and the rehabilitation guidance mode can be a short reminder tone like a text message that only serves to remind the medical staff and the monitored object.

[0068] In order to clarify the correlation between the trend of motion characteristics and body position, the trend of motion characteristics is further illustrated by an example: when the medical staff assists the patient to adjust to the position of the healthy side lying position (i.e. Figure 2The last schematic diagram in the figure, in which the patient lies on his side and his affected side is located above the healthy side), because a pillow is provided on the right side of the mattress 8 and the affected side is located on the right side of the mattress 8 at this time, that is, in the absence of monitoring, the patient wants to get out of bed, and needs to use the healthy side to support, first turn the affected side toward the left side of the mattress 8, turn it to the supine position and then continue to use the healthy side to get up. The patient cannot turn over to the right side of the mattress by himself without monitoring. In this process, the patient needs to first retract the affected upper limb placed on the pillow, retract it to the vicinity of the affected waist, and use the power of the healthy side to gradually turn the affected lower limb to the left side of the mattress. In this process, since the patient is in the healthy side lying position, his healthy side is closer to the left edge of the mattress 8, that is, if the patient turns over toward the left side of the mattress 8, it is very likely to directly turn over the mattress 8 or be in a dangerous position very close to the edge of the bed. In this process, the patient may only want to adjust the position of the affected limb in the healthy side lying position to alleviate the possible poor blood circulation caused by long-term compression, that is, it is possible to use the healthy side to move the affected upper limb up and down or move the affected lower limb slightly.

[0069] Through the above analysis of the possible behaviors of patients in the healthy side lying position without monitoring, the first mode, second mode, third mode and fourth mode corresponding to the body position are preset respectively as follows: if the wearable device 1 worn on the affected upper limb has collected behavioral data that meets the preset movement range of the upper limb, then when it is monitored that the behavioral data collected by the wearable device 1 on the affected lower limb meets the preset movement range of the lower limb, and when it is monitored that the patient changes from lying on the side to lying on the back or from lying on the back to lying on the side, it is determined that the current patient condition is the fourth mode with a large risk factor of small-scale lifting, flipping of the limbs, etc., and the posture of the monitored object changes.

[0070] If the wearable device 1 worn on the affected upper limb has collected behavioral data that meets the preset range of motion of the upper limb, and the monitored subject has good independence, then when it is monitored that the behavioral data collected by the wearable device 1 on the affected lower limb meets the preset range of motion of the lower limb, or when it is monitored that the pressure distribution on the right side of the mattress 8 gradually decreases and the pressure distribution on the left side gradually increases, since the monitored subject has good independence and it is not difficult to turn over or get up, the risk factor is relatively high, and it is determined to be the third mode.

[0071] If the wearable device 1 worn on the affected upper limb has collected behavioral data that meets the preset range of motion of the upper limb, and the monitored subject has poor independence, then when the behavioral data collected by the wearable device 1 on the affected lower limb is monitored to meet the preset range of motion of the lower limb, or when it is monitored that the pressure distribution on the right side of the mattress 8 gradually decreases and the pressure distribution on the left side gradually increases, the patient is usually trying to turn over or trying to get up, but because the monitored subject has poor independence and it is difficult to turn over, the risk factor is medium and it is determined to be the second mode.

[0072] When the wearable device 1 worn on the affected upper limb collects behavioral data that meets the preset movement range of the upper limb, or when the wearable device 1 monitoring the affected lower limb collects behavioral data that meets the preset movement range of the lower limb, and the patient's posture has not changed, the patient may only need to adjust the current posture slightly and the risk factor is low, and it is determined to be the first mode.

[0073] Regarding “the trend of changes in movement characteristics of monitored subjects who have received guidance on body position changes and the trend of changes in movement characteristics of monitored subjects who have not yet received guidance on body position changes”: the guidance on body position changes referred to here mainly refers to the monitored subjects being taught the correct and appropriate posture for turning over or getting up after they have been hospitalized for a period of time. Relatively speaking, not having received guidance on body position changes mainly refers to monitored subjects who have just been hospitalized or have not been hospitalized for long and have not yet received guidance on turning over or getting up. The two types of movement characteristic change trends distinguished in this way can distinguish the movements or behaviors that are more likely to occur in the current monitored subject when out of monitoring, which can effectively improve the speed of data comparison processing, and further improve the accuracy and effectiveness of monitoring of the monitored subject.

[0074] like Figure 1 What is shown is a simplified structural connection relationship diagram of the bed leaving warning system for hemiplegic stroke patients proposed by the present invention.

[0075] The system includes a wearable device 1. The wearable device 1 referred to in the present invention mainly refers to a device in the form of a watch product supported by the wrist and ankle. As the acquisition module 2 proposed in the patent document with the announcement number CN104524760B with the authorization announcement date of August 29, 2017, it uses the smart bracelet worn by the monitored object during exercise, which is turned on and has a three-axis gyroscope sensor built in that can obtain acceleration in three directions, so as to capture the real-time acceleration value and vibration amplitude of the smart bracelet / affected limb in real time. Among them, the built-in sensor in the wearable device 1 is not limited to the three-axis gyroscope sensor, and any sensor that can detect the upward acceleration and forward acceleration and vibration amplitude of the basketball action can be used, such as gravity acceleration. The three-axis gyroscope sensor can simultaneously measure the position, movement trajectory, and acceleration in six directions. The wearable device 1 has a built-in setting module for presetting the position, movement trajectory, acceleration range / amplitude captured by the sensor corresponding to the limb movements such as lifting, flipping, and putting down.

[0076] The system includes a collection module 2 for collecting behavioral data of the monitored object, a risk control level classification module 3 for determining the risk control level, an early warning information processor 4 for determining the triggering of an early warning, a posture detection module 6 for determining the current posture, a behavior rule module 5 for presetting the trend of motion feature changes, a wearable device 1 worn on the affected side of the monitored object, a mobile digital device 7 carried by the monitoring personnel, and a mattress 8 with a processor and a sensor. In addition, a pressure sensor array for collecting patient pressure changes and a processor connected to the pressure sensor array are arranged on the mattress 8. The above-mentioned devices mentioned in the present invention can all be computer processors, Figure 1 A simplified logic flow chart of a bed-leaving warning system for hemiplegic stroke patients is shown, and the operations of the logic flow chart can represent a series of operations implemented by hardware, computer instructions or a combination thereof. In the context of computer instructions, the operations represent computer executable instructions stored on one or more computer-readable storage media, and the computer executable instructions perform the operations when executed by one or more computer processors.

[0077] When the guardian adjusts the monitored object to a certain posture, the guardian records the adjusted posture through the mobile digital device 7 operated by the guardian or the posture detection module 6 determines the adjusted posture. When the monitored object moves his or her limbs on the mattress 8 or adjusts the posture by himself or herself, the processor arranged on the mattress 8 obtains the pressure change data collected by multiple sensors and generates pressure distribution data and pressure change data about the monitored object based on the pressure change data. The processor transmits the pressure change related information (including pressure distribution data and pressure change data) on the mattress 8 for analyzing the behavior trend, posture or posture change trend of the monitored object to the acquisition module 2.

[0078] When the monitored subject moves his / her limbs on the mattress 8 or adjusts his / her posture by himself / herself, the wearable device 1 obtains the behavioral data of the monitored subject, especially the behavioral data corresponding to the upper limb and the lower limb on the affected side, respectively, and is used to extract the first motion characteristic change trend and the second motion characteristic change trend. The wearable device 1 transmits the collected behavioral data to the acquisition module 2 at a predetermined time interval (which can be 1 minute or 30 minutes) preset for real-time monitoring of the patient's behavior.

[0079] The acquisition module 2 transmits the mattress 8 pressure-related information and behavior data it obtains to the posture detection module 6, and when the acquisition module 2 obtains the behavior data of the monitored object determined by the wearable device 1, it generates a motion feature value corresponding to each behavior data based on the behavior data, including at least the upper limb preset movement amplitude, the upper limb preset movement amplitude, the lower limb preset movement amplitude, and the lower limb preset movement amplitude, and transmits the generated motion feature value to the behavior rule module 5.

[0080] The posture detection module 6 processes the received data (the information related to the pressure of the mattress 8 and / or the behavior data) and determines whether there is an obvious upper and lower fault in the pressure distribution on the mattress 8 obtained by the processing and / or whether the pressure distribution on one side is diffuse relative to the other side, and then determines the current posture information of the monitored object, and the posture detection module 6 transmits the posture information to the behavior rule module 5 and the risk control level classification module 3. As a preferred embodiment, the determination of the posture information here is mainly based on the pressure related information set on the mattress 8 or is determined by the nursing staff manually entering it after adjusting the posture.

[0081] Further explanation of the posture detection module 6 is as follows: since a pressure sensor is provided on the mattress 8, the current pressure distribution on the mattress 8 can be obtained by processing the data collected by the pressure sensor. Figure 2 As shown, when there is no obvious upper and lower fault in the obtained pressure distribution, and the pressure distribution on one side is diffuse relative to the other side, based on the known relative position relationship between the affected side of the patient and the mattress 8 (i.e., the position relationship of the affected side closer to the left or right side of the mattress 8 when the patient lies on his back), the posture detection module 6 generates the posture information of the current patient in the supine position. Figure 2 As shown, when there is an obvious upper and lower fault in the obtained pressure distribution, and the pressure distribution on one side is diffuse relative to the other side, based on the known relative position relationship between the patient's affected side and the mattress 8, the posture detection module 6 generates the posture information of the current patient's position in the healthy side lying position or the affected side lying position. Among them, "obvious upper and lower faults appear in the pressure distribution" means: because when the patient lies on his side, one upper limb and one lower limb need to be placed on the pillow respectively, especially the patient's affected upper limb must be fully spread out and placed on the pillow, which is different from normal people who can arbitrarily place both upper limbs at a position close to their head / body. Therefore, the mattress 8 placed between the upper limb and one lower limb on the upper and lower sides of the mattress 8 is not subjected to pressure, and there is an obvious fault between the upper and lower sides where no pressure is monitored as observed from the pressure distribution diagram. And this obvious fault only exists in the healthy side lying position or the affected side lying position. "The pressure distribution on one side is diffuse relative to the other side" means that the side with a higher relative diffusion degree can be observed from the pressure distribution diagram, according to Figure 2As shown, the higher relative diffusion degree is caused by the stretched upper and lower limbs on one side being placed on the pillow, so it can be determined that the side with the higher relative diffusion degree is the side on the mattress 8 where the patient's stretched upper and lower limbs are located. For example, when it is determined that the relative diffusion degree is higher on the left side of the mattress 8, and it is known that the patient has right upper and lower limb paralysis, the placement posture detection module 6 generates the placement posture information of the current patient's body placement posture in the affected side lying position. For example, when it is determined that the relative diffusion degree is higher on the right side of the mattress 8, and it is known that the patient has right upper and lower limb paralysis, the placement posture detection module 6 generates the placement posture information of the current patient's body placement posture in the healthy side lying position.

[0082] The behavior rule module 5 processes the posture information determined by the posture detection module 6 and the multiple motion feature values ​​including at least the first motion feature value and the second motion feature value determined by the acquisition module 2 based on the preset motion feature change trend of the behavior that may occur when the monitored object maintains a certain posture without supervision, and generates affected limb motion data including at least the motion feature value and used to indicate the behavior of the affected upper limb or the affected lower limb, and transmits it to the early warning information processor 4.

[0083] The risk control level classification module 3 is configured to output the preset risk control level data corresponding to the placement posture and used to indicate the degree of movement restriction of the affected limb based on the placement posture information when the placement posture information is input by the guardian through the mobile digital device 7 operated by the guardian or determined by the placement posture detection module 6 after the guardian adjusts the placement posture of the monitored object and / or when the placement posture detection module 6 determines the self-adjusted placement posture information after the monitored object in the unmonitored state moves the limbs on the mattress 8 or adjusts the placement posture by itself, and transmit the data to the early warning information processor 4. A single risk control level includes the degree of movement restriction of a unilateral limb under a certain placement posture. The risk control levels include level I (basically independent but overestimating one's own activity ability), level II (conditionally independent or extremely mildly dependent), level III (moderate or severe dependence) and level IV (extremely severe or completely dependent). Further, at least two risk control levels are divided into evaluations of the independence of the monitored object. Level I (basic independence but overestimation of one's own ability to move) and Level II (conditional independence or very mild dependence) are classified as the evaluation of better independence of the subjects under supervision, and Level III (moderate or severe dependence) and Level IV (extremely severe or complete dependence) are classified as the evaluation of poor independence of the subjects under supervision.

[0084] The warning information processor 4 determines whether it is necessary to issue a warning to the monitored object who is currently moving the limbs on the mattress 8 or adjusting the posture by himself based on the received affected limb movement data determined by the behavior rule module 5 and the risk control level data determined by the risk control level classification module 3 and / or the independence evaluation of the monitored object. When it is determined that the affected limb movement data satisfies at least one pattern in at least one motion characteristic change trend, the warning condition is met. When the trigger condition is met, the bed number, name, warning content, etc. of the monitored object are sent to the surrounding guardians.

[0085] Preferably, if Figure 3 As shown, the present application proposes a mattress suitable for a monitored subject with motor dysfunction in both the upper and lower limbs on the same side, which ensures the comfort of the monitored subject while improving the effectiveness and sensitivity of data acquisition, at least from the two aspects of sensor distribution and bed body resilience and deformation capacity. In the present invention, the mattress 8 is divided into a hip bone area 9 and a shoulder and elbow area 10, which are arranged at intervals from each other and each of which is provided with at least one sensor. The hip bone area 9 corresponds to the area where the buttocks of the monitored subject are located when lying down, and the area extends radially perpendicular to the height direction of the human body and is in the shape of a long strip. The shoulder and elbow area 10 corresponds to the area between the shoulder and elbow of the monitored subject, and the area extends radially perpendicular to the height direction of the human body and is in the shape of a long strip. The bed body resilience and deformation capacity corresponding to the hip bone area 9 is at least higher than the bed body resilience and deformation capacity corresponding to the shoulder and elbow area 10.

[0086] Since the bed corresponding to the hip bone area 9 has better resilience and deformation ability, it can reduce the pressure on the hip bone, sacrum and other limbs of the hemiplegic stroke patient when lying on his side or lying down, which is beneficial to the improvement of blood circulation and the reduction of damage to the affected side or the healthy side caused by side lying. In addition, the sensor used in the present invention is a flexible thin film grid tactile pressure sensor, which is only 0.1mm thick and has good flexibility, and can better improve the detection accuracy and speed.

[0087] Further preferably, the distribution density of sensors arranged on the shoulder and elbow region 10 is at least greater than the distribution density of sensors arranged on the hip region 9. Since the possible contact area between the monitored object and the shoulder and elbow region 10 is much smaller than the possible contact area between the monitored object and the hip region 9, the distribution density of sensors is relatively increased to facilitate the improvement of detection accuracy. The sensor arranged on the mattress 8 used in the present invention is preferably a Teksc an tactile and pressure distribution sensor produced by Beijing Xinbaite Technology Co., Ltd., whose hardware includes a PC-based A / D conversion circuit and a reusable sensor, which is combined with a pressure display and analysis software based on MS Windows to form a pressure monitoring system, which can perform static and dynamic measurements on the pressure distribution of any contact surface, and display the pressure profile and various data in real time with intuitive and vivid two-dimensional and three-dimensional color images, and record and / or store the entire measurement process, so that the monitoring personnel can view and analyze the measurement records at any time.

[0088] It should be noted that the above 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 description and its drawings are illustrative and do not constitute a limitation of the claims. The protection scope of the present invention is defined by the claims and their equivalents.

Claims

1. A rehabilitation system for hemiplegic patients after stroke, the system include: Memory; and at least one processor coupled to the memory, Characterized in that at least one processor comprises: A first processor acquires behavioral data collected by a wearable device (1) worn only on the affected upper and lower limbs of the monitored subject; A second processor is used to determine a corresponding rehabilitation guidance level according to the historical rehabilitation data of the monitored object; a third processor configured to, when determining that the current mode is the rehabilitation guidance mode, adjust the preset rehabilitation guidance level based on the stage attribute association rule between at least one motion feature value extracted from the behavior data by the first processor and the preset rehabilitation guidance level to determine whether the behavior data meets the rehabilitation requirements, and transmit the action achievement information to at least one mobile digital device (7) carried by the surrounding guardian when the rehabilitation requirements are met, The first processor, the second processor and the third processor are all arranged in the wearable device. The fourth processor determines whether the current mode is converted to the rehabilitation guidance mode and instructs the fifth processor in the system for presetting the motion characteristic change trend to increase its trigger threshold when the mode is converted to the rehabilitation guidance mode, or instructs the fifth processor in the system for presetting the motion characteristic change trend to lower its trigger threshold when the mode is converted from the rehabilitation guidance mode to the bed leaving warning mode.

2. The rehabilitation system for hemiplegic stroke patients according to claim 1, It is characterized in that At least one processor is further configured to preset a motion feature change trend corresponding to at least one specific posture and used to indicate the motion feature change trend that may occur in the specific posture.

3. The rehabilitation system for hemiplegic stroke patients according to claim 2, It is characterized in that The motion characteristic change trends include: A first mode with a very low risk factor, in which the limbs are slightly lifted, placed, turned, etc., and the posture of the monitored object does not change; a second mode with a medium risk factor, in which an attempt is made to turn over or try to get up, and the independence of the monitored object is poor; a third mode with a high risk factor, in which an attempt is made to turn over or try to get up, and the independence of the monitored object is good; and a fourth mode with a high risk factor, in which a limb is slightly lifted, placed, turned, etc., and the posture of the monitored object changes; different from different body positions, each body position corresponds to a predefined first mode, second mode, third mode and fourth mode respectively; The stage attribute association rule is distinguished from the degree of motor dysfunction of the patient at the current stage to determine the movement feature change trend corresponding to the first movement feature value and the movement feature change trend corresponding to the second movement feature value; When the motor dysfunction assessment result determined by at least one mobile digital device (7) operated by the monitoring personnel is that the independence of the monitored object is poor, the preset upper limb movement amplitude, upper limb preset movement amplitude, lower limb preset movement amplitude and lower limb preset movement amplitude corresponding to this situation are determined, and based on the determined amplitude / trigger threshold, the boundaries between the first mode, the second mode, the third mode and the fourth mode are distinguished, and the system can switch between the bed leaving warning mode and the rehabilitation guidance mode by raising or lowering the amplitude / trigger threshold of the four modes as a whole.

4. The rehabilitation system for hemiplegic stroke patients according to claim 3, It is characterized in that For a patient in the lying position on the healthy side without monitoring, the fourth mode corresponding to the patient in the above-mentioned position is at least preset as follows: If the wearable device (1) worn on the affected upper limb has collected behavioral data that meets the preset range of motion of the upper limb, then when the behavioral data collected by the wearable device (1) on the affected lower limb is monitored to meet the preset range of motion of the lower limb, and when it is monitored that the patient changes from lying on the side to lying on the back or from lying on the back to lying on the side, it is determined that the current patient condition is the fourth mode with a higher risk factor of small-scale lifting, lowering, and flipping of the limbs and the change of the posture of the monitored object.

5. The rehabilitation system for hemiplegic stroke patients according to claim 3, It is characterized in that For a patient in the lying position on the healthy side without monitoring, the third mode corresponding to the patient in the above-mentioned position is at least preset as: If the wearable device (1) worn on the affected upper limb has collected behavioral data that meets the preset range of motion of the upper limb, and the monitored subject has good independence, then when the behavioral data collected by the wearable device (1) on the affected lower limb is monitored to meet the preset range of motion of the lower limb, or when the pressure distribution on the right side of the mattress (8) is monitored to gradually decrease and the pressure distribution on the left side is monitored to gradually increase, since the monitored subject has good independence and is not difficult to turn over or get up, the risk factor is relatively high and it is determined to be the third mode.

6. The rehabilitation system for hemiplegic stroke patients according to claim 3, It is characterized in that When a patient is in the lying position on the healthy side without monitoring, the second mode corresponding to the patient's position is at least preset as follows: If the wearable device (1) worn on the affected upper limb has collected behavioral data that meets the preset range of motion of the upper limb, and the monitored subject has poor independence, then when the behavioral data collected by the wearable device (1) on the affected lower limb is monitored to meet the preset range of motion of the lower limb, or when the pressure distribution on the right side of the mattress (8) is monitored to gradually decrease and the pressure distribution on the left side is monitored to gradually increase, the patient is usually trying to turn over or trying to get up, but because the monitored subject has poor independence and it is difficult to turn over, the risk factor is medium, and it is determined to be the second mode.

7. The rehabilitation system for hemiplegic stroke patients according to claim 3, It is characterized in that For a patient in the lying position on the healthy side without monitoring, the first mode corresponding to the patient in the above-mentioned position is at least preset as: When the wearable device (1) worn on the affected upper limb collects behavioral data that meets the preset range of motion of the upper limb, or when the wearable device (1) monitoring the affected lower limb collects behavioral data that meets the preset range of motion of the lower limb, and the patient's posture does not change, the patient only needs to slightly adjust the current posture and the risk factor is low, and it is determined to be the first mode.

Images