Six-minute walking distance prediction model construction method, system and equipment for old people

Abstract

The invention is applicable to the technical field of medical aid decision making, and particularly relates to a method, a system and equipment for constructing a six-minute walking distance prediction model for old people, and the method comprises the steps: obtaining brain wave signals, experimental data and body data of a target user in a test process; screening the experimental data into stable data and fluctuation data based on the brain wave signal; determining correction data based on the stable data and the fluctuation data, and determining correction experiment data based on the correction data and the stable data; and taking the body data of each user target and the corrected experiment data as a group of training data in a training data set of a prediction model, and obtaining the prediction model based on multiple groups of training data. According to the old people group six-minute walking distance prediction model construction method provided by the invention, the prediction accuracy of the old people group six-minute walking distance prediction model can be improved.

Description

Technical Field

[0001] This application belongs to the field of medical auxiliary decision-making technology, and in particular relates to a method, system and device for constructing a six-minute walking distance prediction model for the elderly population. Background Technology

[0002] The six-minute walking distance can reflect the exercise tolerance of the subjects, provide a comprehensive analysis of the respiratory, cardiovascular and metabolic systems, and is simple, safe and does not require expensive equipment. This method has now become an important method for objectively assessing patients' cardiopulmonary and overall exercise function.

[0003] In related technologies, functional assessment of patients with cardiopulmonary insufficiency has always been a key focus for clinicians. The traditional approach involves asking a few questions, such as "How many floors can you climb?" or "How far can you walk?" These methods are still widely used in bedside history taking in my country. However, the patient's subjective recollection of emotions and attitudes often directly affects the assessment results. Therefore, collecting objective measurement parameters is often superior to self-description. That is, it is necessary to establish a model to predict the six-minute walking distance of the elderly. In the process of establishing the six-minute walking distance of the elderly, the target user's body data is often used as input data, and the target user's six-minute walking distance is used as output data. The body data and the six-minute walking distance are used as training data to input into the learning model for training, thereby obtaining a six-minute walking distance prediction model for the elderly population. Alternatively, a curve fitting is performed based on the body data and the six-minute walking distance to obtain a six-minute walking distance prediction model for the elderly population.

[0004] The six-minute walk distance test is conducted as follows: At the same time each day, participants walk back and forth along a straight line within a 30-meter test area. If participants experience symptoms such as shortness of breath or fatigue, they can choose to rest or terminate the test. Researchers time the participants every minute and provide consistent, steady verbal encouragement. At the end of the test, researchers immediately record the participants' 6MWD, blood pressure, heart rate, oxygen saturation, shortness of breath, and overall fatigue. The data representing the longest walk distance between two trials is taken as the experimental result.

[0005] Researchers have found that the mindset of the participants can also affect the results of the six-minute walking distance experiment, leading to inaccurate results. Consequently, the predictive model built based on these inaccurate results will also be inaccurate in predicting the six-minute walking distance, making it impossible to accurately predict the six-minute walking distance based solely on the physical data of the elderly population. Summary of the Invention

[0006] This application provides a method, system, and device for constructing a six-minute walking distance prediction model for the elderly population, which can improve the problem of inaccurate predictions in the six-minute walking distance prediction model for the elderly population.

[0007] In a first aspect, embodiments of this application provide a method for constructing a six-minute walking distance prediction model for an elderly population, including: The experiment acquires EEG signals, experimental data, and physical data of the target user during the experiment; wherein the experimental data reflects the change in walking distance of the target user over time during the experiment, and the physical data reflects the inherent physiological characteristics of the target user. The experimental data is categorized into stable data and fluctuating data based on the electroencephalogram (EEG) signals. The stable data reflects data indicating a stable mindset of the target user, while the fluctuating data reflects data indicating an unstable mindset of the target user. Based on the stable data and the fluctuation data, corrected data is determined, and based on the corrected data and the stable data, corrected experimental data is determined; wherein, the corrected data refers to the adjusted fluctuation data, and the corrected experimental data refers to the adjusted experimental data; The body data of each user target and the corrected experimental data are used as a set of training data in the training dataset of the prediction model, and the prediction model is obtained based on multiple sets of training data.

[0008] The technical solutions described in this application embodiment have at least the following technical effects: The method for constructing a six-minute walking distance prediction model for the elderly population provided in this application first acquires the target user's electroencephalogram (EEG) signals during the experiment, experimental data reflecting the change in the target user's walking distance over time, and body data reflecting the target user's inherent physiological characteristics. Then, based on the EEG signals, the experimental data is divided into stable data reflecting the target user's stable mental state and fluctuating data reflecting the target user's unstable mental state. Next, based on the stable and fluctuating data, corrected data reflecting the adjusted fluctuating data is determined. Then, based on the corrected and stable data, adjusted experimental data is determined. Finally, by using the body data and corrected experimental data of each target user as a set of training data in the training dataset of the prediction model, a prediction model capable of predicting the six-minute walking distance of the elderly population is constructed using the training data corresponding to each different target user.

[0009] This method can effectively divide experimental data into fluctuating data that needs correction and stable data that does not need correction through EEG signals. Then, the stable data is used to anchor the fluctuating data that needs correction and adjust the data accordingly. This improves the correlation between the adjusted data and the user's actual exercise level, while reducing other irrelevant parameters, such as the correlation between mindset and the user's actual exercise ability. This makes the adjusted experimental results closer to the user's actual six-minute walking distance.

[0010] In one possible implementation of the first aspect, the screening of the experimental data into stable data and fluctuating data based on the electroencephalogram (EEG) signals includes: Based on a preset interval, the EEG signal is divided into multiple signal segments, and an initial signal and a signal set are obtained based on the multiple signal segments; wherein, the initial signal refers to the first signal segment among the multiple signal segments, and the signal set refers to the set of the other signal segments excluding the first signal segment among the multiple signal segments; Based on the initial signal and the signal set, a stable time and a fluctuating time are determined; wherein, the stable time is used to reflect the period during which the target user's mentality is stable, and the fluctuating time is used to reflect the period during which the target user's mentality is unstable; The data in the experimental data that matches the stable time and the initial signal are jointly identified as stable data, and the data in the experimental data that matches the fluctuation time are identified as fluctuating data.

[0011] In one possible implementation of the first aspect, determining the settling time and fluctuation time based on the initial signal and the signal set includes: Based on the initial signal, a reference fluctuation level is determined; wherein the reference fluctuation level is used to reflect the ratio between the fluctuation level of the α wave and the fluctuation level of the β wave in the initial signal; Based on the baseline fluctuation level, the settling time and fluctuation time are determined from each signal segment in the signal set.

[0012] In one possible implementation of the first aspect, determining the settling time and fluctuation time from each signal segment in the signal set based on the reference fluctuation level includes: Based on the reference fluctuation level, a matching point is obtained from each of the signal segments; wherein, the matching point is used to reflect the data points in the signal segment that have the same reference fluctuation level; Based on the matching point, the signal segment is cumulatively statistically analyzed in reverse time order to determine the cumulative data; wherein, the cumulative data refers to the average value among the cumulatively statistically analyzed data in the signal segment; When a preset condition is met, the enclosing interval is obtained; wherein, the preset condition is used to reflect that the cumulative data is not equal to the baseline fluctuation level, and the enclosing interval is used to reflect the range of data accumulated and statistically analyzed in the signal segment; The time corresponding to the enclosed interval is identified as the stable time, and the time not corresponding to the enclosed interval is identified as the fluctuating time.

[0013] In one possible implementation of the first aspect, determining corrected data based on the stable data and the fluctuation data, and determining corrected experimental data based on the corrected data and the stable data, includes: Based on the stable data, the i-th fatigue factor is determined; wherein, the i-th fatigue factor is used to reflect the attenuation trend between the walking distance corresponding to the i-th signal segment and the walking distance corresponding to the (i+1)-th signal segment during the experiment. Based on the i-th fatigue factor, the stable data, and the fluctuation data, the i-th correction sub-data of the correction data is determined; wherein, the i-th correction sub-data refers to the fluctuation data adjusted in the (i+1)-th signal segment; The i-th corrected sub-data of the corrected data is concatenated with the corresponding stable sub-data in the stable data to obtain the i-th corrected experimental sub-data of the corrected experimental data; wherein, the i-th corrected experimental sub-data refers to the experimental data after adjustment in the (i+1)-th signal segment.

[0014] In one possible implementation of the first aspect, determining the i-th fatigue factor based on the stable data includes: Based on the time period of the stable data corresponding to the (i+1)th signal segment, first matching data is obtained by matching the stable data of the i-th signal segment; wherein, the time length of the first matching data is the same as the time length of the stable data corresponding to the i-th signal segment. The ratio between the walking rate corresponding to the matching data and the walking rate corresponding to the stable data of the (i+1)th signal segment is determined as the i-th fatigue factor.

[0015] In one possible implementation of the first aspect, determining the i-th correction sub-data of the correction data based on the i-th fatigue factor, the stable data, and the fluctuation data includes: After determining the i-th fatigue factor, based on the time period of the fluctuation data corresponding to the (i+1)-th signal segment, second matching data is obtained from the stable data of the i-th signal segment; wherein, the time length of the second matching data is the same as the time length of the fluctuation data corresponding to the i-th signal segment; Based on the second matching data and the i-th fatigue factor, the i-th correction sub-data in the correction data is determined.

[0016] In one possible implementation of the first aspect, the step of concatenating the i-th modified sub-data of the modified data with the corresponding stable sub-data in the stable data to obtain the i-th modified experimental sub-data of the modified experimental data includes: The i-th corrected sub-data is concatenated with the stable sub-data corresponding to the (i+1)-th signal segment in the stable data to obtain the i-th corrected experimental sub-data. After determining the i-th corrected experimental sub-data, the experimental data corresponding to the (i+1)-th signal segment is synchronously replaced with the i-th corrected experimental sub-data, and the i-th corrected experimental sub-data is confirmed as the stable data corresponding to the (i+1)-th signal segment.

[0017] Secondly, embodiments of this application provide a system for constructing a six-minute walking distance prediction model for the elderly population, including: The data acquisition unit is used to acquire brainwave signals, experimental data, and physical data of the target user during the experiment; wherein, the experimental data is used to reflect the change of the target user's walking distance over time during the experiment, and the physical data is used to reflect the inherent physiological characteristics of the target user. A data filtering unit is used to filter the experimental data into stable data and fluctuating data based on the electroencephalogram (EEG) signals; wherein, the stable data is used to reflect the data in the experimental data in which the target user's mentality is stable, and the fluctuating data is used to reflect the data in the experimental data in which the target user's mentality is unstable; A data correction unit is configured to determine corrected data based on the stable data and the fluctuation data, and to determine corrected experimental data based on the corrected data and the stable data; wherein the corrected data refers to the adjusted fluctuation data, and the corrected experimental data refers to the adjusted experimental data; The model building unit is used to take the body data of each user target and the corrected experimental data as a set of training data in the training dataset of the prediction model, and obtain the prediction model based on multiple sets of the training dataset.

[0018] Thirdly, embodiments of this application provide a device for constructing a six-minute walking distance prediction model for the elderly population, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the method described in any of the first aspects above.

[0019] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in any of the first aspects above.

[0020] Fifthly, embodiments of this application provide a computer program that, when run on a device for constructing a six-minute walking distance prediction model for the elderly, causes the device to execute the method for constructing a six-minute walking distance prediction model for the elderly as described in any of the first aspects.

[0021] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart illustrating a method for constructing a six-minute walking distance prediction model for the elderly population, provided in an embodiment of this application. Figure 2 This is a schematic diagram illustrating the implementation process of a method for constructing a six-minute walking distance prediction model for the elderly population, provided in an embodiment of this application. Figure 3 This is a schematic diagram of the structure of a six-minute walking distance prediction model construction system for the elderly population provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a device for constructing a six-minute walking distance prediction model for the elderly population, provided in one embodiment of this application. Detailed Implementation

[0024] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0025] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0026] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0027] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0028] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0029] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0030] In related technologies, functional assessment of patients with cardiopulmonary insufficiency has always been a key focus for clinicians. The traditional approach involves asking a few questions, such as "How many floors can you climb?" or "How far can you walk?" These methods are still widely used in bedside history taking in my country. However, the patient's subjective recollection of emotions and attitudes often directly affects the assessment results. Therefore, collecting objective measurement parameters is often superior to self-description. It is necessary to establish a model to predict the six-minute walking distance of the elderly. In the process of establishing the six-minute walking distance of the elderly, the model is often developed by using the target user's body data as input data and the target user's six-minute walking distance as output data. The body data and the six-minute walking distance are used as training data to input into the learning model for training, thereby obtaining a six-minute walking distance prediction model for the elderly population. Alternatively, a curve fitting is performed based on the body data and the six-minute walking distance to obtain a six-minute walking distance prediction model for the elderly population.

[0031] The six-minute walk distance test is conducted as follows: At the same time each day, participants walk back and forth along a straight line within a 30-meter test area. If participants experience symptoms such as shortness of breath or fatigue, they can choose to rest or terminate the test. Researchers time the participants every minute and provide consistent, steady verbal encouragement. At the end of the test, researchers immediately record the participants' 6MWD, blood pressure, heart rate, oxygen saturation, shortness of breath, and overall fatigue. The data representing the longest walk distance between two trials is taken as the experimental result.

[0032] Researchers have found that the mindset of the participants can also affect the results of the six-minute walking distance experiment, leading to inaccurate results. Consequently, the predictive model built based on these inaccurate results will also be inaccurate in predicting the six-minute walking distance, making it impossible to accurately predict the six-minute walking distance based solely on the physical data of the elderly population.

[0033] To address the aforementioned issues, this application provides a method, system, and device for constructing a six-minute walking distance prediction model for the elderly population. The method first acquires the target user's electroencephalogram (EEG) signals during the experiment, experimental data reflecting the change in the target user's walking distance over time, and body data reflecting the target user's inherent physiological characteristics. Then, based on the EEG signals, the experimental data is divided into stable data reflecting the target user's stable mental state and fluctuating data reflecting the target user's unstable mental state. Next, based on the stable and fluctuating data, corrected data reflecting the adjusted fluctuating data is determined. Then, based on the corrected and stable data, adjusted experimental data is determined. Finally, the body data and corrected experimental data for each target user are used as a set of training data in the training dataset of the prediction model.

[0034] This method can effectively divide experimental data into fluctuating data that needs correction and stable data that does not need correction through EEG signals. Then, the stable data is used to anchor the fluctuating data that needs correction and adjust the data accordingly. This improves the correlation between the adjusted data and the user's actual exercise level, while reducing other irrelevant parameters, such as the correlation between mindset and the user's actual exercise ability. This makes the adjusted experimental results closer to the user's actual six-minute walking distance.

[0035] The method for constructing a six-minute walking distance prediction model for the elderly population provided in this application embodiment can be applied to a device for constructing a six-minute walking distance prediction model for the elderly population. In this case, the device for constructing a six-minute walking distance prediction model for the elderly population is the executing entity of the method for constructing a six-minute walking distance prediction model for the elderly population provided in this application embodiment. This application embodiment does not impose any restrictions on the specific type of device for constructing a six-minute walking distance prediction model for the elderly population.

[0036] The device for building the six-minute walking distance prediction model for the elderly population can be a terminal device, which can be a mobile phone, tablet, laptop, ultra-mobile personal computer (UMPC), netbook, smart screen, smart TV, handheld device with wireless communication function, desktop computer, computer, laptop computer, handheld computing device, etc.

[0037] To better understand the method for constructing a six-minute walking distance prediction model for the elderly population provided in this application embodiment, the specific implementation process of the method for constructing a six-minute walking distance prediction model for the elderly population provided in this application embodiment will be described exemplarily below.

[0038] Figure 1 and Figure 2 A schematic flowchart illustrating the method for constructing a six-minute walking distance prediction model for the elderly population provided in this application is shown. Please refer to [link / reference]. Figure 1 and Figure 2 Methods for constructing a six-minute walking distance prediction model for the elderly population include: S100 acquires the target user's electroencephalogram (EEG) signals, experimental data, and physical data during the experiment; the experimental data reflects the change in the target user's walking distance over time during the experiment, and the physical data reflects the target user's inherent physiological characteristics.

[0039] It is understandable that electroencephalogram (EEG) signals can reflect the physiological signals of the electrical activity of neuronal populations in the target user's brain. Their waveform characteristics can indirectly correlate with the user's attention state, cognitive load, fatigue level, or emotional stability during the experiment. Experimental data refers to the distance-time curve formed by the distance walked by the target user during the six-minute walking distance experiment. The data corresponding to the last time point of this curve at the end of the six minutes represents the experimental results of the six-minute walking distance experiment. Physical data refers to the physiological characteristics of the target user collected before the six-minute walking distance experiment, such as age, height, gender, BMI, and exercise habits. The acquisition of EEG signals and experimental data needs to be synchronized; that is, the time when EEG signals are acquired should also be the time when the six-minute walking distance experiment begins.

[0040] For example, brainwave signals can be acquired using wearable EEG acquisition devices. These devices require multiple electrodes to be attached to specific locations on the user's head and a preset sampling frequency to continuously record raw brainwave data throughout the experiment, ultimately forming brainwave signals. Experimental data can also be acquired using positioning devices or motion sensing devices. These devices need to record timestamps at preset time intervals and simultaneously record the user's cumulative walking distance at the corresponding time points, ultimately forming a time-walking distance dataset. Body data such as height and weight can be obtained using standard measurement tools, and body data such as age and exercise habits can be obtained through questionnaires, and so on.

[0041] S200 categorizes experimental data into stable data and fluctuating data based on electroencephalogram (EEG) signals. Stable data reflects data indicating stable mindset of the target user, while fluctuating data reflects data indicating unstable mindset of the target user.

[0042] It's understandable that fluctuations in the six-minute walking distance could be caused by physiological or psychological factors. The six-minute walking distance is a crucial metric for assessing the cardiopulmonary function of target users; if fluctuations are due to psychological factors, the assessment of their cardiopulmonary function will be inaccurate.

[0043] For example, brainwave signals can be divided into multiple segments at preset intervals. The time periods of stable and unstable mental states for the target user can then be determined based on these segments. The time periods of stable mental states in the experimental data are considered stable data, and the time periods of unstable mental states are considered fluctuating data. Alternatively, brainwave signals and experimental data can be input into a learning model. The learning model outputs corresponding stable and fluctuating data. The training process of the learning model can involve processing the brainwave signals, experimental data, and corresponding stable and fluctuating data to create a training dataset. This training dataset is then input into the learning model for training, ultimately resulting in the learned model. And so on, but not limited to these examples.

[0044] In one possible implementation, in step S200, the experimental data is categorized into stable data and fluctuating data based on the electroencephalogram (EEG) signals, including: S210, based on a preset interval, divides the EEG signal into multiple signal segments, and obtains an initial signal and a signal set based on the multiple signal segments; wherein, the initial signal refers to the first signal segment among the multiple signal segments, and the signal set refers to the collection of other signal segments among the multiple signal segments excluding the first signal segment.

[0045] As can be understood, the preset interval refers to the time interval during a six-minute walking distance experiment in which the target user is given the same, steady verbal encouragement while being informed of the time. The EEG signal is segmented according to the preset interval, resulting in multiple data segments, which are called signal segments. The first signal segment refers to the segment where the EEG signal first appears.

[0046] For example, the preset intervals can be manually entered or obtained directly from an interval database. The interval database refers to a database containing preset intervals corresponding to the six-minute walking distance experiment. This data can be obtained through laboratory experiments, on-site measurements and monitoring, and past experience. After acquisition, the collected data is organized, classified, and archived, useful information and patterns are extracted, and the relevant data is saved into the database to form the interval database.

[0047] S220, based on the initial signal and signal set, determines the stable time and fluctuation time; wherein, the stable time is used to reflect the period when the target user's mentality is stable, and the fluctuation time is used to reflect the period when the target user's mentality is unstable.

[0048] It is understandable that during the walking distance test, the target users' mindset is easily influenced by external factors. When the target users were given time updates at preset intervals and received the same, stable verbal encouragement, their mindset fluctuated to varying degrees. These different mindset levels affect walking distance; for example, walking with a higher-than-normal mindset results in a longer distance than walking with a normal mindset, and vice versa. When the target users were never given time updates and received the same, stable verbal encouragement—that is, during the initial signal—their mindset did not fluctuate. Furthermore, during the time period corresponding to the initial signal, because the experiment had just begun, there was no sense of urgency as the experiment neared its end, further indicating that the signal segment corresponding to the initial signal was obtained when the target users were in a stable mindset.

[0049] For example, the ratio of the fluctuation levels between the α wave and the β wave in the initial test signal can be determined using the initial test signal, and then the settling time and fluctuation time can be determined from the signal set based on this ratio. Alternatively, feedback image information can be input into the learning model, and the learning model can output the corresponding sensitivity value, and so on, but it is not limited to these methods.

[0050] In one possible implementation, step S220, determining the settling time and fluctuation time based on the initial signal and signal set, includes: S221, Based on the initial signal, determine the reference fluctuation level; wherein, the reference fluctuation level is used to reflect the ratio between the fluctuation level of the α wave and the fluctuation level of the β wave in the initial signal.

[0051] It's understandable that the initial signal is a mixture of multiple brainwave frequencies. It needs to be separated into alpha and beta waves through frequency filtering. Alpha waves refer to brainwave signals with a fixed frequency of 8Hz-13Hz, while beta waves refer to brainwave signals with a fixed frequency of 14Hz-30Hz. Alpha waves correspond to a stable state of relaxation and focused attention, while beta waves correspond to a state of tension and active cognitive activity. The degree of fluctuation refers to the standard deviation of the brainwave signal amplitude. Baseline fluctuation = Alpha wave fluctuation ÷ Beta wave fluctuation. A larger baseline fluctuation indicates that alpha waves are more stable than beta waves (i.e., it indicates that the user is more relaxed and has a stable mindset), while a smaller baseline fluctuation indicates that beta waves are less stable than alpha waves (i.e., it indicates that the user is more tense and anxious).

[0052] S222, based on the baseline fluctuation level, determines the settling time and fluctuation time from each signal segment in the signal set.

[0053] It's understandable that when target users hear the time announcement and receive consistent, calm verbal encouragement, their emotional fluctuations can be categorized into five scenarios: First, an initial increase followed by a slow decrease; second, no significant change; third, a brief increase followed by a rapid decrease; fourth, a brief decrease followed by recovery; and fifth, an increase followed by maintenance. Of these five scenarios, the first three are more common, while the latter two are less so. The fourth scenario is less common because maintaining gait balance and rhythm requires a high level of concentration for target users during a physically demanding six-minute walking experiment. While an external auditory stimulus might momentarily disrupt their rhythm, this is typically more likely to occur when an elderly person's gait is extremely unstable or their cognitive processing abilities have significantly declined. The fifth scenario is less common because target users have limited physical reserves and recovery capabilities; fatigue accumulates over time, and a one-time psychological stimulus is unlikely to sustainably counteract the physiological exertion over a long, pre-set interval.

[0054] For example, a feature point can be matched from each signal segment in the signal set using a baseline volatility level. The volatility level corresponding to this feature point is the same as the baseline volatility level. Then, by processing the corresponding signal segment using this feature point, the average value among the accumulated statistical data in the signal segment is determined. When this average value is not equal to the baseline volatility level, the range of accumulated statistical data is obtained. Finally, the time within this data range is identified as the stable time, and the time outside this data range is identified as the fluctuating time. Alternatively, the baseline volatility level and the signal set can be input into the learning model, and the learning model can output the corresponding stable time and fluctuating time, and so on, but it is not limited to these methods.

[0055] With this setup, the activity level of alpha waves in brainwaves is positively correlated with a relaxed and calm state of mind, while the activity level of beta waves is positively correlated with fluctuating states of mind such as tension and anxiety. The ratio of their fluctuation levels can directly and objectively reflect the stability of the mind. By establishing the ratio of alpha waves to beta waves in the initial signal as the baseline fluctuation level, each signal segment in the subsequent signal set can be compared with this baseline to quantitatively determine the degree of deviation from the mindset, thereby qualitatively dividing the stable time and the fluctuating time, significantly improving the scientific rigor and consistency of the judgment.

[0056] In one possible implementation, in step S222, based on the reference fluctuation level, the settling time and fluctuation time are determined from each signal segment in the signal set, including: S2221, based on the reference fluctuation level, a matching point is obtained from each signal segment in the signal set; wherein, the matching point is used to reflect the data points in the signal segment that have the same reference fluctuation level.

[0057] It is understandable that the fluctuation level corresponding to the matching point is the same as the baseline fluctuation level. The calculation method for the fluctuation level corresponding to the matching point can be obtained in a similar way to the baseline fluctuation level obtained in step S221. The baseline fluctuation level is obtained from a segment of signal, while the fluctuation level of the matching point is a single point in time. Because at each preset interval, it is necessary to announce the time and provide the target user with the same and stable verbal encouragement, it is necessary to match each signal segment to obtain the matching point corresponding to that signal segment.

[0058] S2222, based on the matching point, the signal segment is accumulated and statistically analyzed in reverse time order to determine the accumulated data; where the accumulated data refers to the average value among the accumulated statistical data in the signal segment.

[0059] It is understandable that cumulative statistics in reverse chronological order means that data points are included in the statistical set one by one in reverse order. After each data point is included, the average value of all data in the current set is calculated immediately, and the cumulative data is finally generated.

[0060] For example, if a point in time is used as the anchor point, the formula for calculating cumulative data is: Where k represents the cumulative data, i corresponds to the matching point, and X... i Let Y be the sum of the amplitudes of the α wave. i This is the sum of the amplitudes of the α wave, and so on. If the preset interval is 60s and the time point corresponding to the matching point is 90s, the cumulative data obtained from the first calculation is the ratio between the amplitude of the α wave and the amplitude of the β wave when the second signal segment is at 90s. If the preset condition is not met, the cumulative data obtained from the second calculation is the ratio between the average value of the sum of the amplitudes of the α wave and the average value of the sum of the amplitudes of the β wave during the time period of 89s-90s for the second signal segment, and so on.

[0061] S2223, when the preset conditions are met, obtain the enclosing interval; wherein, the preset conditions are used to reflect that the cumulative data is not equal to the baseline fluctuation, and the enclosing interval is used to reflect the range of data accumulated in the signal segment.

[0062] It can be understood that the inclusion interval refers to the data range between the matching point and the inclusion point. The inclusion point refers to the point in the data segment that is farthest from the matching point when the cumulative data calculated after performing the reverse time cumulative statistics according to step S2222 is not equal to the baseline fluctuation level. For example, according to the example in step S2222, when the calculated cumulative data changes from equal to the baseline fluctuation level to not equal in the time period of 77s-90s, it can be said that 77s-90s is the inclusion interval, and so on.

[0063] S2224 identifies the time corresponding to the enclosed interval as the stable time and the time not corresponding to the enclosed interval as the fluctuating time.

[0064] It is understandable that the fluctuation level of EEG signals within the enclosed range is the same as the baseline fluctuation level, while the fluctuation level of EEG signals outside the enclosed range is different from the baseline fluctuation level. Since the baseline fluctuation level corresponds to a stable mindset, it can be explained that the time corresponding to the enclosed range is the time when the target user's mindset is stable, and the time not corresponding to the enclosed range is the time when the target user's mindset is fluctuating.

[0065] With this setup, in a six-minute walk experiment, elderly users' mental states may fluctuate briefly within the same signal segment. If only the entire signal segment is compared with the baseline, some stable periods might be misjudged as fluctuations. Conversely, by first locating data points within the signal segment that perfectly match the baseline fluctuation level through matching points, and then using those points as starting points to accumulate statistics in reverse order (calculating the average ratio of alpha and beta waves back towards the starting point), continuous stable periods can be dynamically captured. When the accumulated data matches the baseline fluctuation level, it indicates that the periods within the backtracking range are in a stable state. When the accumulated data does not match the baseline, the encompassing interval precisely defines the last continuous stable range. This point-by-point tracing method improves the accuracy of time division from the signal segment level to the data point level, significantly reducing misjudgments of state.

[0066] S230: Data that matches the stable time in the experimental data and the initial signal are jointly identified as stable data, and data that matches the fluctuation time in the experimental data are identified as fluctuating data.

[0067] It can be understood that stable data refers to experimental data that is only affected by the physiological properties of the target users, while fluctuating data refers to experimental data that is affected not only by the physiological properties of the target users but also by their psychological properties. In other words, stable data is experimental data that does not require correction, while fluctuating data is experimental data that requires correction.

[0068] This setup breaks down a continuous six-minute EEG signal into multiple independent signal segments at preset intervals, transforming long-sequence data into units that can be analyzed segment by segment. This reduces the analytical errors that may occur due to information overload when directly processing continuous signals. By setting the initial signal, the EEG characteristics at the beginning of the experiment (usually the most stable mental state) are used as a benchmark, providing an objective reference standard for judging the mental state of subsequent signal segments. This allows for a quantitative comparison of whether the mental state is stable or not. By time matching, the stable and fluctuating times obtained from the EEG analysis are correlated with the walking experiment data, accurately distinguishing stable data that is less affected by mental state interference and fluctuating data that is more affected by interference, thus achieving noise removal from the original experimental data.

[0069] S300 determines corrected data based on stable and fluctuating data, and determines corrected experimental data based on the corrected and stable data; where corrected data refers to the adjusted fluctuating data, and corrected experimental data refers to the adjusted experimental data.

[0070] It is understandable that the corrected data is the data after correcting the fluctuating data, and the corrected experimental data is the data after combining the stable data and the corrected data. Both the corrected experimental data and the experimental data are distance-time change curves. The data corresponding to the last time point of the six-minute test is the experimental result of the six-minute walking distance test, while the data corresponding to the last time point of the six-minute test in the corrected experimental data is the six-minute walking distance after eliminating the influence of the target user's psychological properties.

[0071] For example, stable data can be used to determine the attenuation trend between the walking distance corresponding to the i-th signal segment and the walking distance corresponding to the (i+1)-th signal segment during the experiment. Then, based on this attenuation trend, stable data, and fluctuation data, adjusted correction data for the (i+1)-th signal segment can be determined. These adjusted correction data are then concatenated with the stable data to obtain the corrected experimental data. Alternatively, stable data and fluctuation data can be input into a learning model, which outputs the corresponding corrected data, and so on, but are not limited to these methods.

[0072] In one possible implementation, in step S300, corrected data is determined based on stable data and fluctuation data, and corrected experimental data is determined based on the corrected data and stable data, including: S310, Based on stable data, determine the i-th fatigue factor; wherein, the i-th fatigue factor is used to reflect the attenuation trend between the walking distance corresponding to the i-th signal segment and the walking distance corresponding to the (i+1)-th signal segment during the experiment.

[0073] It is understandable that within the walking distance corresponding to each signal segment, the target user's physical strength gradually decreases due to continuous walking, and the fatigue factor is a characteristic value that maps the degree of gradual decrease.

[0074] For example, data corresponding to the stable data of the (i+1)th signal segment can be extracted from the stable data of the i-th signal segment based on the time period of the stable data corresponding to the (i+1)th signal segment, and then the ratio of the walking rate between this data and the stable data corresponding to the (i+1)th signal segment can be used as the i-th fatigue factor. Alternatively, the stable data can be input into the learning model, and the learning model can output the corresponding fatigue factor, and so on, but it is not limited to these methods.

[0075] In one possible implementation, in step S310, determining the i-th fatigue factor based on stable data includes: S311, based on the time period of the stable data corresponding to the (i+1)th signal segment, first matching data is obtained from the stable data of the i-th signal segment; wherein, the time length of the first matching data is the same as the time length of the stable data corresponding to the i-th signal segment.

[0076] It can be understood that the relative time between the first matched data and the stable data corresponding to the (i+1)th signal segment is the same, and the time interval between the points where the relative time between the first matched data and the stable data corresponding to the (i+1)th signal segment is the same is also the same, all of which are preset intervals. For example, if the time period of the stable data corresponding to the (i+1)th signal segment is from 1 minute 30 seconds to 2 minutes, the corresponding first matched data is the data in the i-th signal segment that is from 30 seconds to 1 minute, and so on.

[0077] S312, the ratio between the walking rate corresponding to the first matched data and the walking rate corresponding to the stable data of the (i+1)th signal segment is determined as the i-th fatigue factor.

[0078] It can be understood that the fatigue factor = walking speed corresponding to the first matched data ÷ walking speed corresponding to the stable data of the (i+1)th signal segment. The walking speed corresponding to the first matched data and the walking speed corresponding to the stable data of the (i+1)th signal segment are both average walking speeds within the corresponding time period, i.e., walking speed = (walking distance corresponding to the last time point - walking distance corresponding to the initial time point) ÷ time period, where the last time point refers to the maximum time point in the time period, and the initial time point refers to the minimum time point in the time period.

[0079] This setup ensures that the physical decline of older adults is typically gradual and exhibits significant individual differences. By comparing stable data from different signal segments before and after each individual's own experience to calculate the fatigue factor, the individual adaptability issues associated with using universal fatigue standards (such as fixed attenuation ratios) are reduced, making the quantification results more closely reflect each user's actual physical condition. Walking speed directly reflects walking ability, and the ratio of stable data rates between the i-th and (i+1)-th signal segments accurately quantifies the proportion of natural physical decline over a continuous period. This quantification method not only conforms to the physiological law that fatigue leads to a gradual decrease in walking ability but also provides a unified numerical reference standard for fatigue levels across different signal segments and users, solving the problem of difficulty in comparing fatigue levels across time periods and individuals.

[0080] S320, based on the i-th fatigue factor, stable data, and fluctuation data, determine the i-th correction sub-data of the correction data; where the i-th correction sub-data refers to the fluctuation data adjusted in the (i+1)-th signal segment.

[0081] For example, after determining the i-th fatigue factor, based on the time period of the fluctuation data corresponding to the (i+1)-th signal segment, data corresponding to that time period is extracted from the stable data of the i-th signal segment. Then, the i-th correction data is determined based on this data and the i-th fatigue factor. Alternatively, the fatigue factor, stable data, and fluctuation data can be input into the learning model, and the learning model can output the corresponding correction sub-data, and so on, but it is not limited to these methods.

[0082] In one possible implementation, in step S320, based on the i-th fatigue factor, stable data, and fluctuation data, the i-th correction sub-data of the correction data is determined, including: S321, after determining the i-th fatigue factor, based on the time period of the fluctuation data corresponding to the (i+1)-th signal segment, the second matching data is obtained from the stable data of the i-th signal segment; wherein, the time length of the second matching data is the same as the time length of the fluctuation data corresponding to the i-th signal segment.

[0083] It can be understood that the i-th fatigue factor represents the attenuation trend of the target user's walking distance within a preset interval between the experimental data corresponding to the i-th signal segment and the experimental data corresponding to the (i+1)-th signal segment. That is, after determining the i-th fatigue factor, for the fluctuating data that needs correction, a second matching data is matched from the stable data of the i-th signal segment. This matching data has the same relative time as the fluctuating data corresponding to the (i+1)-th signal segment.

[0084] S322, Based on the second matching data and the i-th fatigue factor, determine the i-th correction sub-data in the correction data.

[0085] It can be understood that the product of the walking rate corresponding to the second matched data and the i-th fatigue factor is the walking rate corresponding to the i-th modified sub-data. The i-th modified sub-data is plotted based on the walking rate corresponding to the i-th modified sub-data, the time period of the fluctuation data corresponding to the (i+1)-th signal segment, and the starting point of the fluctuation data corresponding to the (i+1)-th signal segment. The plotting process involves using the starting point of the fluctuation data corresponding to the (i+1)-th signal segment as the starting point of the i-th modified sub-data, using the walking rate corresponding to the i-th modified sub-data as the trend of walking distance change, and the time length of the plotted graph being the time period of the fluctuation data corresponding to the (i+1)-th signal segment.

[0086] This setup, through the calculation logic of the second matching data multiplied by the fatigue factor, anchors the fluctuating data to the user's own fatigue decline trend, achieving the dual effect of noise reduction and preservation of regularity. The second matching data reflects the user's stable walking ability over the same time length in the i-th signal segment. Combined with the i-th fatigue factor, the calculated corrected sub-data not only eliminates outliers caused by psychological fluctuations but also retains the fatigue decline trend that conforms to physiological laws, transforming the originally disruptive fluctuating data into effective information that reflects true physical fitness.

[0087] S330, the i-th corrected sub-data of the corrected data is concatenated with the corresponding stable sub-data in the stable data to obtain the i-th corrected experimental sub-data of the corrected experimental data; where the i-th corrected experimental sub-data refers to the adjusted experimental data in the (i+1)-th signal segment.

[0088] Data splicing can be understood as the process of concatenating a stable sub-data point from stable data with the i-th corrected sub-data point from corrected data. After obtaining the i-th corrected sub-data point, the curve shape of the i-th corrected sub-data point changes compared to the original fluctuating data within that time period, creating a gap between the i-th corrected sub-data point and the corresponding stable sub-data point. Since the stable sub-data point represents experimental data under stable mental conditions, it means that the corresponding stable sub-data point reflects the distance walked by the target user during the six-minute walking experiment within the corresponding time period. In other words, the data splicing process involves concatenating the first endpoint of the i-th corrected sub-data point with the last endpoint of the corresponding stable sub-data point in the stable data to obtain the i-th corrected experimental sub-data point, and so on. Because the i-th corrected sub-data point is constructed based on the corresponding stable data, the first endpoint of the i-th corrected sub-data point and the last endpoint of the corresponding stable sub-data point are at the same point.

[0089] For example, the i-th corrected sub-data can be obtained by concatenating the i-th corrected sub-data with the stable sub-data corresponding to the (i+1)-th signal segment in the stable data. After obtaining the i-th corrected experimental sub-data, the experimental data is updated with the i-th corrected sub-data to form the corrected experimental data.

[0090] In this setup, the fluctuation data of elderly users in the six-minute walk experiment is not entirely worthless; it still implicitly reflects the declining walking ability due to physiological fatigue, but this is masked by the noise of emotional fluctuations. By using a fatigue factor as a benchmark to correct the fluctuation data, the noise of emotional interference is removed while retaining the physiological fatigue information. This transforms fluctuation data that might otherwise be discarded into useful information, thereby improving data utilization. The six-minute walk is a continuous physiological process, and the previous segmented screening may have resulted in a temporal disconnect between stable and fluctuation data. By concatenating the corrected sub-data with the stable sub-data, the corrected fluctuation data is integrated with the stable data within the same signal segment into a continuous corrected experimental sub-data. This ensures that the data within each signal segment fully reflects the actual changes in walking ability during that period, and the data from different signal segments are smoothly connected in the time dimension, restoring the continuous physiological process of the six-minute walk and reducing data breaks caused by segmented processing.

[0091] In one possible implementation, in step S330, the i-th modified sub-data of the modified data is concatenated with the corresponding stable sub-data in the stable data to obtain the i-th modified experimental sub-data of the modified experimental data, including: S331, the i-th correction sub-data and the stable sub-data corresponding to the (i+1)-th signal segment in the stable data are concatenated end to end to obtain the i-th correction experimental sub-data.

[0092] It can be understood that data splicing refers to splicing the beginning of the corresponding stable sub-data in the stable data with the end of the i-th corrected sub-data in the corrected data. The mention in step S100 that the experimental data is a distance-time variation curve formed by the distance and time traveled by the target user indicates that the corrected sub-data, stable data, corrected experimental sub-data, and fluctuation data are all in curve form.

[0093] S332, after determining the i-th corrected experimental sub-data, the experimental data corresponding to the (i+1)-th signal segment is simultaneously replaced with the i-th corrected experimental sub-data, and the i-th corrected experimental sub-data is confirmed as the stable data corresponding to the (i+1)-th signal segment.

[0094] It can be understood that after determining the i-th corrected experimental sub-data, the experimental data corresponding to the (i+1)-th signal segment is replaced with the i-th corrected experimental sub-data to form an update operation. Upon completing another update, the i-th corrected experimental sub-data is confirmed as the stable data corresponding to the (i+1)-th signal segment. Therefore, when analyzing to obtain the (i+1)-th fatigue factor and correcting the fluctuation data corresponding to the (i+2)-th signal segment based on the (i+1)-th fatigue factor, the (i+1)-th corrected experimental sub-data can be used as stable data to determine the (i+1)-th fatigue factor. This forms an adjustment logic chain of "i-th fatigue factor → correcting the fluctuation data corresponding to the (i+1)-th signal segment → updating stable data → i+1-th fatigue factor → correcting the fluctuation data corresponding to the (i+2)-th signal segment → updating stable data," ultimately completing the adjustment of the entire experimental data and obtaining the corrected experimental data.

[0095] This setup constructs an iterative correction loop by replacing the original experimental data and updating stable data. The experimental data of the (i+1)th signal segment is replaced with the corrected experimental sub-data and confirmed as the new stable data, allowing subsequent processing of the (i+2)th signal segment to be based on the optimized baseline. This iterative logic, where the preceding correction results provide a baseline for subsequent processing, solves the problem of accumulated errors caused by fixed baseline data in segmented processing. As the signal segments progress, the stable data is continuously updated with more precise corrected data, forming a self-optimizing chain for the correction of the entire six-minute experimental data. The final output corrected experimental data more closely reflects the actual changes in the walking ability of the elderly in terms of overall trend.

[0096] S400, the body data and corrected experimental data of each user target are used as a set of training data in the training dataset of the prediction model, and the prediction model is obtained based on multiple sets of the training data.

[0097] It is understandable that the corrected experimental data is the distance-time curve of the target user within six minutes, and the endpoint of this curve represents the target user's six-minute walking distance. Table 1 is obtained by organizing the target user's six-minute walking distance and corresponding data. In Table 1, P < 0.05 indicates that the differences between the statistical data are statistically significant.

[0098] Table 1. 6MWD results of subjects with different characteristics Table 2 Correlation analysis between demographic characteristics and 6MWD The process of obtaining the six-minute walking distance prediction model for the elderly population involves first performing correlation analysis on multiple data points in the body data using the Pearson correlation coefficient. The Pearson correlation coefficient measures the strength and direction of the linear relationship between two continuous variables. A positive Pearson correlation coefficient indicates a positive correlation between the two continuous variables, while a negative Pearson correlation coefficient indicates a negative correlation. The closer the absolute value of the Pearson correlation coefficient is to 1, the stronger the linear relationship between the two continuous variables (see the P-values ​​in Table 2). In the collected data, the parameters... The experiment included 104 elderly men and 110 elderly women. Physical data was collected from these 214 (104+110) participants, including age, height, weight, and BMI. Pearson correlation analysis revealed that, in both the male and female elderly groups, all three physical data points except weight were close to 1, indicating a strong linear relationship between the six-minute walk distance (6MWD) and the participants' age, height, and BMI.

[0099] After determining the corrected experimental data for each target user and the corresponding six-minute walking distance, the six-minute walking distances were arranged in order of size. Individual data points from the target users' body data were analyzed using the Pearson correlation coefficient to extract features related to the six-minute walking distance (i.e., age, height, and BMI). When age, height, and BMI were all found to be correlated with the six-minute walking distance, age, height, and BMI were used as independent variables, and the corresponding six-minute walking distance as the dependent variable. This was then fed into SPSS software to construct a mathematical model. Finally, by validating the six-minute walking distance and the corresponding body data of the target users, the final result was: Male six-minute walking distance (m) = 373.331 - 3.907 × age (years) + 3.486 × height (cm) - 6.646 × BMI (kg / m²) 2 ), R 2 =0.263; Female 6MWD(m)=231.588-4.07×age (years)+4.651×height (cm)-7.852×BMI (kg / m²) 2 ), R 2=0.323, which can predict the six-minute walking distance (6MWD) of 26.3% of the elderly male population and 32.3% of the elderly female population, respectively. After further analysis of the target users' physical data, the significantly higher 6MWD for men compared to women may be due to men's greater height and muscle mass. Age is negatively correlated with 6MWD, as various bodily functions, such as muscle strength, mass, and VO2 max, gradually decline with age. Height is positively correlated with 6MWD, possibly because taller individuals have longer strides and higher walking efficiency. The study found no significant correlation between weight and 6MWD, but BMI was negatively correlated with 6MWD. This may be related to the selection of participants, as underweight and obese individuals were excluded upon enrollment.

[0100] This setup uses EEG signals to divide experimental data into fluctuating data that requires correction and stable data that does not. The stable data is then used to anchor the fluctuating data that requires correction, resulting in adjusted data that is more closely correlated with the user's actual motor performance. This reduces other irrelevant parameters, such as the correlation between mindset and the user's actual motor ability, making the adjusted experimental results closer to the user's actual six-minute walking distance.

[0101] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0102] Corresponding to the method for constructing a six-minute walking distance prediction model for the elderly population described in the above embodiments, this application also provides a system for constructing a six-minute walking distance prediction model for the elderly population. Each module of the system can implement each step of the method for constructing a six-minute walking distance prediction model for the elderly population. Figure 3 The diagram shows a structural block diagram of the system for constructing a six-minute walking distance prediction model for the elderly population provided in an embodiment of this application. For ease of explanation, only the parts related to the embodiments of this application are shown.

[0103] Reference Figure 3 The system for building a six-minute walking distance prediction model for the elderly population includes: The data acquisition unit is used to acquire brainwave signals, experimental data, and physical data of the target user during the experiment. The experimental data reflects the change in the target user's walking distance over time during the experiment, and the physical data reflects the inherent physiological characteristics of the target user.

[0104] The data filtering unit is used to filter experimental data into stable data and fluctuating data based on EEG signals; stable data is used to reflect data in which the target user's mentality is stable, while fluctuating data is used to reflect data in which the target user's mentality is unstable.

[0105] The data correction unit is used to determine corrected data based on stable data and fluctuating data, and to determine corrected experimental data based on the corrected data and stable data; wherein, the corrected data refers to the adjusted fluctuating data, and the corrected experimental data refers to the adjusted experimental data.

[0106] The model building unit is used to take the body data of each user target and the corrected experimental data as a set of training data in the training dataset of the prediction model, and obtain the prediction model based on multiple sets of training data.

[0107] It should be noted that the information interaction and execution process between the above systems / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0108] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0109] This application also provides a device for constructing a six-minute walking distance prediction model for the elderly population. Figure 4 This is a schematic diagram of the structure of a device 4 for constructing a six-minute walking distance prediction model for the elderly population, provided in an embodiment of this application. Figure 4 As shown, the device 4 for building a six-minute walking distance prediction model for the elderly population in this embodiment includes: at least one processor 40 ( Figure 4 Only one is shown in the image), at least one memory 41 ( Figure 4(Only one is shown in the image) and a computer program 42 stored in the at least one memory 41 and executable on the at least one processor 40. When the processor 40 executes the computer program 42, it causes the elderly population six-minute walking distance prediction model construction device 4 to implement the steps in any of the above embodiments of the elderly population six-minute walking distance prediction model construction method, or causes the elderly population six-minute walking distance prediction model construction device 4 to implement the functions of each module / unit in the above embodiments of the system.

[0110] Exemplarily, the computer program 42 may be divided into one or more modules / units, which are stored in the memory 41 and executed by the processor 40 to complete this application. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program 42 in the six-minute walking distance prediction model building device for the elderly population 4.

[0111] The device 4 for building the six-minute walking distance prediction model for the elderly population can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. This device 4 may include, but is not limited to, a processor 40 and a memory 41. Those skilled in the art will understand that... Figure 4 This is merely an example of the device 4 for building a six-minute walking distance prediction model for the elderly population, and does not constitute a limitation on the device 4. It may include more or fewer components than shown in the figure, or combine certain components, or different components, such as input / output devices, network access devices, buses, etc.

[0112] The processor 40 can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0113] In some embodiments, the memory 41 may be an internal storage unit of the six-minute walking distance prediction model building device 4 for the elderly population, such as a hard disk or memory of the device 4. In other embodiments, the memory 41 may be an external storage device of the device 4, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the device 4. Furthermore, the memory 41 may include both internal and external storage units of the device 4. The memory 41 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory 41 can also be used to temporarily store data that has been output or will be output.

[0114] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0115] This application provides a computer program product that, when run on a device for building a six-minute walking distance prediction model for the elderly, enables the device to implement the steps described in any of the above method embodiments.

[0116] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0117] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0118] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0119] In the embodiments provided in this application, it should be understood that the disclosed system for constructing a six-minute walking distance prediction model for the elderly population can be implemented in other ways. For example, the embodiments of the six-minute walking distance prediction model construction system for the elderly population described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0120] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0121] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for constructing a six-minute walk distance prediction model for an elderly population, characterized by, The method comprises the following steps: obtaining brain wave signals, experimental data and body data of a target user during an experiment; wherein the experimental data is used to reflect the data of the walking distance of the target user changing with time during the experiment, and the body data is used to reflect the inherent physiological characteristics of the target user; based on the brain wave signals, the experimental data is screened into stable data and fluctuation data; wherein the stable data is used to reflect the data of the target user's stable mentality in the experimental data, and the fluctuation data is used to reflect the data of the target user's unstable mentality in the experimental data; based on the stable data and the fluctuation data, the correction data is determined, and based on the correction data and the stable data, the corrected experimental data is determined; wherein the correction data refers to the adjusted fluctuation data, and the corrected experimental data refers to the adjusted experimental data; the body data and the corrected experimental data of each target user are used as a set of training data in the training data set of the prediction model, and based on multiple sets of training data, a prediction model is obtained.

2. The method of claim 1, wherein the six-minute walk distance prediction model for the elderly population is constructed by the steps of: The method comprises the following steps: based on a preset interval, the brain wave signals are divided into multiple signal segments, and based on multiple signal segments, an initial signal and a signal set are obtained; wherein the initial signal refers to the first signal segment in the multiple signal segments, and the signal set refers to the collection of other signal segments in the multiple signal segments except the first signal segment; based on the initial signal and the signal set, the stable time and the fluctuation time are determined; wherein the stable time is used to reflect the time period of the target user's stable mentality, and the fluctuation time is used to reflect the time period of the target user's unstable mentality; the data in the experimental data matched with the stable time and the initial signal are jointly confirmed as stable data, and the data in the experimental data matched with the fluctuation time are confirmed as fluctuation data.

3. The method for constructing a six-minute walking distance prediction model for the elderly population as described in claim 2, characterized in that, The method comprises the following steps: based on the initial signal, a reference fluctuation degree is determined; wherein the reference fluctuation degree is used to reflect the ratio between the fluctuation degree of alpha wave and the fluctuation degree of beta wave in the initial signal; based on the reference fluctuation degree, the stable time and the fluctuation time are determined from each signal segment in the signal set, respectively.

4. The method of claim 3, wherein the six-minute walk distance prediction model for the elderly population is constructed by the steps of: The method comprises the following steps: based on the reference fluctuation degree, a matching point is matched from each signal segment in the signal set; wherein the matching point is used to reflect the data point in the signal segment that is the same as the reference fluctuation degree; based on the matching point, the signal segment is accumulated and counted in reverse order of time to determine the cumulative data; wherein the cumulative data refers to the average value of the data accumulated and counted in the signal segment; When a preset condition is met, an inclusive interval is obtained; wherein, the preset condition is used to reflect that the cumulative data is not equal to the reference fluctuation degree, and the inclusive interval is used to reflect a data range in the signal segment which is accumulated and counted; A time corresponding to the inclusive interval is confirmed as a stable time, and a time not corresponding to the inclusive interval is confirmed as a fluctuation time.

5. The method of constructing a six-minute walk distance prediction model for the elderly population according to claim 2, wherein, The stable data and the fluctuation data are used to determine correction data, and the correction data and the stable data are used to determine corrected experimental data, including: Based on the stable data, an i-th fatigue factor is determined; wherein, the i-th fatigue factor is used to reflect an attenuation trend between a walking distance corresponding to an i-th signal segment and a walking distance corresponding to an i+1-th signal segment in an experimental process of the target user; Based on the i-th fatigue factor, the stable data and the fluctuation data, an i-th correction sub-data of the correction data is determined; wherein, the i-th correction sub-data refers to adjusted fluctuation data in the i+1-th signal segment; The i-th correction sub-data of the correction data and a corresponding stable sub-data in the stable data are spliced to obtain an i-th correction experimental sub-data of the corrected experimental data; wherein, the i-th correction experimental sub-data refers to adjusted experimental data in the i+1-th signal segment.

6. The method of constructing a six-minute walk distance prediction model for the elderly population according to claim 5, wherein, The stable data is used to determine the i-th fatigue factor, including: Based on a time period of the stable data corresponding to the i+1-th signal segment, a first matching data is matched from the stable data of the i-th signal segment; wherein, a time length of the first matching data is the same as a time length of the stable data corresponding to the i-th signal segment; A ratio between a walking speed corresponding to the first matching data and a walking speed corresponding to the stable data of the i+1-th signal segment is confirmed as the i-th fatigue factor.

7. The method for constructing a six-minute walking distance prediction model for the elderly population as described in claim 6, characterized in that, The i-th fatigue factor, the stable data and the fluctuation data are used to determine the i-th correction sub-data of the correction data, including: After the i-th fatigue factor is determined, based on a time period of the fluctuation data corresponding to the i+1-th signal segment, a second matching data is matched from the stable data of the i-th signal segment; wherein, a time length of the second matching data is the same as a time length of the fluctuation data corresponding to the i-th signal segment; Based on the second matching data and the i-th fatigue factor, the i-th correction sub-data in the correction data is determined.

8. The method for constructing a six-minute walking distance prediction model for the elderly population as described in claim 5, characterized in that, The i-th correction sub-data of the correction data and a corresponding stable sub-data in the stable data are spliced to obtain the i-th correction experimental sub-data of the corrected experimental data, including: The i-th correction sub-data and the stable sub-data corresponding to the i+1-th signal segment in the stable data are spliced at the beginning and the end to obtain the i-th correction experimental sub-data. After determining the i-th modified experimental sub-data, the i+1-th signal segment corresponding experimental data is replaced by the i-th modified experimental sub-data synchronously, and the i-th modified experimental sub-data is confirmed as the i+1-th signal segment corresponding stable data. 9.A system for constructing a six-minute walk distance prediction model for an elderly population, characterized by, The method comprises the steps of: a data acquisition unit configured to acquire electroencephalogram signals, experimental data, and physical data of a target user during an experiment; wherein the experimental data is used to reflect the data of the walking distance of the target user changing with time during the experiment, and the physical data is used to reflect the inherent physiological characteristics of the target user; a data screening unit configured to screen the experimental data into stable data and fluctuation data based on the electroencephalogram signals; wherein the stable data is used to reflect the data of the target user's stable mentality in the experimental data, and the fluctuation data is used to reflect the data of the target user's unstable mentality in the experimental data; a data correction unit configured to determine correction data based on the stable data and the fluctuation data, and determine modified experimental data based on the correction data and the stable data; wherein the correction data refers to the adjusted fluctuation data, and the modified experimental data refers to the adjusted experimental data; a model construction unit configured to take the physical data and the modified experimental data of each user target as a set of training data in a training data set of a prediction model, and obtain a prediction model based on a plurality of sets of training data.

10. A device for constructing a six-minute walk distance prediction model for an elderly population, characterized by The computer program product comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the method of any one of claims 1 to 8.

Images