A mental fatigue monitoring method and system based on alar skin temperature

Abstract

The application discloses a mental fatigue monitoring method and system based on alar skin temperature, comprising the steps of acquiring alar skin temperature data and environmental temperature data, preprocessing the alar skin temperature data, constructing a low-temperature compensation model, performing dynamic calibration, and obtaining an alar skin temperature threshold; calculating an alar skin temperature change amount, extracting time domain features of the alar skin temperature, and inputting the alar skin temperature change amount and the time domain features into a pre-trained evaluation model to output a mental fatigue grade score; comparing the mental fatigue grade score with a preset fatigue threshold, and outputting a fatigue state judgment result according to the comparison result. The application can accurately output a fatigue grade, ensure accurate and reliable judgment, and provide scientific and convenient mental fatigue monitoring.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of health monitoring, in particular to a mental fatigue monitoring method and system based on alar skin temperature. BACKGROUND

[0002] Mental fatigue, as a common sub-health problem of contemporary office workers, is particularly prominent in modern office environments dominated by brain work. Employees working in a mode of multitasking, high-intensity decision-making, and information overload for a long time are prone to inattention, emotional burnout, and decreased work efficiency, and may also increase the risk of chronic cardiovascular and cerebrovascular diseases, anxiety, and other health risks. Therefore, developing a real-time mental fatigue monitoring system suitable for daily office environments is of great significance for providing accurate early warnings and scientific interventions for mental fatigue.

[0003] Mental fatigue is a neurophysiological phenomenon induced by long-term cognitive load, mainly manifested as inattention, emotional depression, and decreased work efficiency, and is a common manifestation of sub-health status.

[0004] In the prior art, mental fatigue detection methods mainly rely on electroencephalogram (EEG), electrocardiogram (heart rate variability, HRV), electrooculogram (EOG), facial micro-expression, and other technologies to monitor relevant physiological signals or features to determine the mental fatigue state. However, it has been found in research that mental fatigue systems relying on EEG and ECG technologies generally have the defects of discomfort, limited use scenarios, and susceptibility to environmental stimuli, and this method is more limited to laboratory research. Although the currently marketed EOG and facial micro-expression technologies have good reliability, they are still affected by the use environment, such as changes in light, facial obstruction, and other factors, affecting the accuracy of monitoring. SUMMARY

[0005] To overcome the deficiencies of the prior art, the present application provides a mental fatigue monitoring method and system based on alar skin temperature, which reflects the user's mental fatigue level by real-time monitoring and analyzing the alar skin temperature change rate, and provides real-time reports of mental fatigue for users.

[0006] To achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0007] In a first aspect, the present application provides a mental fatigue monitoring method based on alar skin temperature, comprising:

[0008] Obtaining alar skin temperature data and environmental temperature data, and obtaining an alar skin initial temperature average value from the alar skin temperature data;

[0009] Pretreating the alar skin temperature data, constructing a low-temperature compensation model based on the environmental temperature data and a preset environmental temperature threshold, and performing dynamic calibration to obtain an alar skin temperature threshold;

[0010] a change amount of the alar skin temperature is calculated, time domain features of the alar skin temperature are extracted, and the change amount of the alar skin temperature and the time domain features are input into a pre-trained evaluation model to output a mental fatigue grade score;

[0011] The mental fatigue grade score is compared with a preset fatigue threshold, and a fatigue state judgment result is output according to a comparison result.

[0012] As a further technical solution, when the environmental temperature data is less than a preset environmental temperature threshold, a low-temperature compensation model is started, and the low-temperature compensation model is constructed and represented as: ; wherein, represents an alar skin temperature threshold, represents an alar skin initial temperature average, represents a compensation coefficient, represents environmental temperature data;

[0013] When the environmental temperature data is greater than and equal to the environmental temperature threshold, the low-temperature compensation model is not started.

[0014] As a further technical solution, based on real-time acquired alar skin temperature data and an alar skin temperature threshold, a change amount of the alar skin temperature is calculated, and is specifically represented as:

[0015] ; wherein, represents a change amount of the alar skin temperature, represents real-time acquired alar skin temperature data, represents an alar skin temperature threshold.

[0016] As a further technical solution, the extracted time domain features of the alar skin temperature include an alar skin temperature change rate and a fluctuation amplitude, the alar skin temperature change rate is a rate of decrease of the alar skin temperature per unit time, and the fluctuation amplitude is a standard deviation of the alar skin temperature in a set time window.

[0017] As a further technical solution, the pre-trained evaluation model is represented as:

[0018] ; wherein, represents an age correction coefficient, represents a change amount of the alar skin temperature, represents a time interval, represents an alar skin temperature change rate, represents a fluctuation amplitude, represents a gender correction coefficient.

[0019] As a further technical solution, the fatigue threshold is divided into a first fatigue threshold, a second fatigue threshold and a third fatigue threshold; when the mental fatigue level score is less than the first fatigue threshold, it is determined to be a non-mental fatigue state; when the mental fatigue level score is greater than or equal to the first fatigue threshold and less than the second fatigue threshold, it is determined to be a mild mental fatigue state; when the mental fatigue level score is greater than or equal to the second fatigue threshold and less than the third fatigue threshold, it is determined to be a moderate mental fatigue state; and when the mental fatigue level score is greater than the third fatigue threshold, it is determined to be a severe mental fatigue state.

[0020] As a further technical solution, after outputting the fatigue state judgment result, the nose wing skin temperature change, the time domain feature and the fatigue state judgment result are transmitted to a mobile terminal through Bluetooth for display.

[0021] In a second aspect, the present application provides a mental fatigue monitoring system based on nose wing skin temperature, comprising the following modules:

[0022] The temperature acquisition module is configured to acquire nose wing skin temperature data and environmental temperature data, and obtain a nose wing skin initial temperature average value from the nose wing skin temperature data;

[0023] The data processing and calibration module is configured to preprocess the nose wing skin temperature data, and construct a low-temperature compensation model based on the environmental temperature data and a preset environmental temperature threshold, and perform dynamic calibration to obtain a nose wing skin temperature threshold;

[0024] The mental fatigue level score output module is configured to calculate a nose wing skin temperature change, extract a time domain feature of the nose wing skin temperature, and input the nose wing skin temperature change and the time domain feature into a pre-trained evaluation model to output a mental fatigue level score;

[0025] The fatigue state determination module is configured to compare the mental fatigue level score with a preset fatigue threshold, and output a fatigue state judgment result according to the comparison result.

[0026] The one or more technical solutions of the present application have the following beneficial effects:

[0027] The present application can exclude interference such as illumination level, head and face shielding by collecting nose wing skin temperature data, and is less affected by the environment compared to electro-oculography and facial micro-expression technology. In combination with the dynamic calibration (error ±0.1℃) of the environmental temperature sensor, the nose temperature change rate, fluctuation amplitude and other time domain features are extracted, and input into a specific evaluation model, which can accurately output the fatigue level and ensure accurate and reliable determination.

[0028] The mental fatigue monitoring system provided by the application can be directly integrated into glasses, a daily necessity, and there is no discomfort in the process of obtaining the temperature data of the nasal alar skin and the environmental temperature data, and the mental fatigue monitoring system is suitable for various scenes such as daily office work, and solves the problems of discomfort in wearing and limited use scene of electroencephalogram, heart rate variability and other technologies.

[0029] The application sends data to a terminal such as a mobile phone through wireless transmission, displays a nasal temperature change curve and a fatigue level (color mark) in real time, triggers vibration and pop-up reminders when moderate fatigue or above occurs, helps users to take intervention measures in time, improves brain efficiency, and reduces the incidence of chronic fatigue and related health risks. BRIEF DESCRIPTION OF DRAWINGS

[0030] The drawings constituting a part of the specification of the application are used to provide a further understanding of the application, and the illustrative embodiments of the application and the description thereof are used to explain the application, and do not constitute an improper limitation on the application.

[0031] Figure 1 A flowchart of a mental fatigue monitoring method based on the temperature of the nasal alar skin in the application. DETAILED DESCRIPTION

[0032] It should be pointed out that the following detailed description is exemplary and is intended to provide further explanation of the application. Unless otherwise specified, all technical and scientific terms used in the application have the same meaning as generally understood by those skilled in the art to which the application belongs.

[0033] Example 1

[0034] The literature "Physiological indicators of task demand, fatigue, and cognition in future digital manufacturing environments" published in the International Journal of Human-Computer Studies in 2021 pointed out that compared with low mental activity, when the human body is in high-intensity mental activity, the nasal temperature will rapidly decrease by about 0.25℃ (p<0.05); the literature "The mental nose and the Pinocchio effect: thermography, planning, anxiety, and lies" even found that when people face real high mental load and stress, the nasal temperature decreases by about 2.4℃, and when they face simulated mental work, the average decrease is 0.9℃ (p<0.05). The literature "Facial temperature as a measure of mental workload" found that the nasal temperature of people under high mental load decreased by 1.5℃. The mainstream explanation for this phenomenon is that when the brain energy consumption increases, the blood is preferentially supplied to the central nervous system, causing the peripheral blood vessels on the surface of the head to contract, and the temperature of the nose to decrease. Therefore, the embodiment provides a mental fatigue monitoring method based on the temperature of the alar skin, which reflects the mental fatigue level of the user by monitoring and analyzing the change rate of the temperature of the alar skin in real time, and provides scientific guarantee for the mental efficiency of the user. As shown in Figure 1 , the specific method steps are as follows:

[0035] S1: Obtain alar skin temperature data and environmental temperature data, and obtain the average value of the initial temperature of the alar skin according to the alar skin temperature data.

[0036] In step S1, a high-precision temperature sensor (precision 0.1℃ or higher) is embedded in the inside of the nose pad of the glasses, with a sampling frequency of 10Hz (collecting once every 100 milliseconds), which can quickly generate a continuous data sequence when the temperature changes. The high-precision temperature sensor collects the alar skin temperature data (in the following real-time obtained alar skin temperature data is represented as ), and at the same time, the environmental temperature sensor on the nosepiece of the frame collects the environmental temperature , the environmental temperature is used to compensate for the interference of the environment.

[0037] After the user wears the glasses, the high-precision temperature sensor collects the alar skin temperature data in the stationary state for 5 times continuously, and takes the average value to obtain the average value of the initial temperature of the alar The ambient temperature sensor collects real-time ambient temperature as an initial ambient temperature , and as a reference value.

[0038] S2: Preprocess the alar skin temperature data, construct a low-temperature compensation model based on the ambient temperature data and a preset ambient temperature threshold, perform dynamic calibration, and obtain an alar skin temperature threshold.

[0039] In step S2, the alar skin temperature data is preprocessed by using a low-power chip to denoise and filter the alar skin temperature data.

[0040] Based on the ambient temperature data and a preset ambient temperature threshold, a low-temperature compensation model is constructed, dynamic calibration is performed, and an alar skin temperature threshold is obtained for subsequent feature extraction, which is specifically represented as: in the present embodiment, the preset ambient temperature threshold is 22℃, and when the ambient temperature data is less than the preset ambient temperature threshold (22℃), the low-temperature compensation model is started, the alar skin temperature threshold is automatically adjusted, and the constructed low-temperature compensation model is represented as:

[0041] ; wherein, represents the alar skin temperature threshold, represents the average value of the initial alar skin temperature, represents the compensation coefficient, represents the ambient temperature data;

[0042] When the ambient temperature data is greater than and equal to the ambient temperature threshold (22℃), the low-temperature compensation model is not started, and the average value of the initial alar skin temperature is directly used for subsequent feature extraction.

[0043] S3: Calculate the alar skin temperature change, extract the time domain features of the alar skin temperature, and input the alar skin temperature change and the time domain features into a pre-trained evaluation model to output a mental fatigue level score.

[0044] In step S3, first, based on the alar skin temperature data and the alar skin temperature threshold obtained in the above steps, the alar skin temperature change is calculated, which is specifically represented as:

[0045] ; wherein, represents the alar skin temperature change, represents the real-time alar skin temperature data, represents a nasal ala skin temperature threshold value.

[0046] The time domain features of the nasal ala skin temperature are extracted again, including the nasal ala skin temperature change rate and the fluctuation amplitude, wherein the nasal ala skin temperature change rate (℃ / min) is the rate of decrease of the nasal ala skin temperature per unit time, and is represented as: The fluctuation amplitude is the standard deviation of the nasal ala skin temperature in a set time window , used to reflect the stability of the nasal ala skin temperature.

[0047] The nasal ala skin temperature change amount (T) and the time domain features (T , ) are input into a pre-trained evaluation model, which is represented as:

[0048] ; wherein, represents an age correction coefficient, in this embodiment, takes 0.9-1.1, represents the nasal ala skin temperature change amount, represents the time interval, represents the nasal ala skin temperature change rate, represents the fluctuation amplitude, represents a gender correction coefficient, the female value takes 1.92, and the male value takes 0.92. The evaluation model outputs a mental fatigue level score .

[0049] S4: Compare the mental fatigue level score with a preset fatigue threshold value, and output a fatigue state judgment result according to the comparison result.

[0050] In step S4, the fatigue threshold value is divided into a first fatigue threshold value , a second fatigue threshold value , and a third fatigue threshold value When the mental fatigue level score is less than the first fatigue threshold value , it is determined to be a non-mental fatigue state; when the mental fatigue level score is greater than or equal to the first fatigue threshold value and less than the second fatigue threshold value , it is determined to be a mild mental fatigue state; when the mental fatigue level score is greater than or equal to the second fatigue threshold value and less than the third fatigue threshold value , it is determined to be a moderate mental fatigue state; when the mental fatigue level score is greater than the third fatigue threshold value At that time, it was determined to be a state of severe mental fatigue.

[0051] After outputting the fatigue state assessment result, the change in nasal skin temperature will be... Temporal characteristics ( , The fatigue status assessment results are transmitted via Bluetooth to a mobile terminal (such as a mobile phone or a smart band app) for display. The display includes: a curve showing changes in nasal temperature; fatigue level (color markings: green = non-fatigue, yellow = mild, orange = moderate, red = severe) and warning prompts (vibration and pop-up reminders are triggered when fatigue is moderate or above).

[0052] In this embodiment, taking a woman as an example, she wears glasses with an embedded temperature sensor (a high-precision miniature temperature sensor integrated in the nose pad, with an accuracy of 0.1℃), and her mental fatigue state is determined through the following process:

[0053] During user operation, the temperature sensor built into the nose pad collects real-time data on the skin temperature of the nose wings. The average baseline value obtained from five consecutive measurements in the initial static state is... 33.5℃, ambient temperature 25℃ (If the temperature is above 22℃, use the average initial nasal skin temperature directly).

[0054] During the work, the real-time measured skin temperature data of the nasal ala were sequentially... The temperatures were 33.2℃, 32.8℃, 32.5℃, and 32.2℃. Then, the feature data were extracted and calculated to obtain the change in nasal skin temperature. The temperatures are -0.3℃, -0.7℃, -1.0℃, and -1.3℃, respectively; the set time interval is... Calculate the rate of change in skin temperature on the nasal ala within 1 minute. The temperature ranges were -0.3℃ / min, -0.7℃ / min, -1.0℃ / min, and -1.3℃ / min, respectively; the standard deviation of temperature within the time window was set to σ = 0.1, 0.2, 0.3, and 0.4 (to reflect the steady decrease in the temperature of the skin on the nasal ala).

[0055] The change in skin temperature on the nasal wing ( ) and temporal characteristics ( , The input is fed into the pre-trained evaluation model. In this embodiment, α=1.0 and β=1.92. (fatigue) represents the calculated level of mental fatigue score. The mental fatigue score is then calculated based on this score: when... At 33.2℃, =1.0×[-200×(-0.3)+100×(-0.3)]+500×0.1×1.92=1.0×(60 - 30)+96 = 126. Similarly, the fatigue scores at 32.8℃, 32.5℃, and 32.2℃ are respectively f 2 = 262, f 3 = 388, f 4 = 514. Then, fatigue level determination is carried out: Set the thresholds F1, F2, and F3 to 100 (mild fatigue), 300 (moderate fatigue), and 500 (severe fatigue) respectively. According to the obtained mental fatigue level scores above, the fatigue levels are respectively: =126>F1—Mild mental fatigue state; F1 = 100 ≤ =262<F2 = 300—Mild mental fatigue state; F2 = 300 ≤ =388 < F3 = 500—Moderate mental fatigue state (remind to rest); =514 > F3 = 500—Severe fatigue state (force to remind to stop working).

[0056] Finally, the change amount of the nasal wing skin temperature , time domain features ( , ), and the obtained fatigue state judgment results are transmitted to the mobile terminal via Bluetooth, and the change curve of the nasal wing skin temperature (from 33.5℃ to 32.2℃) and the fatigue level color marking (green: non-mental fatigue; yellow: mild mental fatigue; orange: moderate mental fatigue; red: severe mental fatigue) are displayed in real time; at the same time, different warning prompts are issued according to the fatigue level. There is no warning for non-mental fatigue and mild mental fatigue. When it is moderate fatigue, a vibration pop-up window ("It is recommended to rest") is triggered, and when it is severe mental fatigue, a pop-up window appears with "Stop working immediately".

[0057] Embodiment 2

[0058] In this embodiment, a mental fatigue monitoring system based on the nasal wing skin temperature is provided, including the following modules:

[0059] Temperature acquisition module, configured to: acquire the nasal wing skin temperature data and the ambient temperature data, and obtain the average value of the initial nasal wing skin temperature according to the nasal wing skin temperature data;

[0060] Data processing and calibration module, configured to: preprocess the nasal wing skin temperature data, and construct a low-temperature compensation model based on the ambient temperature data and a preset ambient temperature threshold for dynamic calibration to obtain the nasal wing skin temperature threshold;

[0061] The mental fatigue level score output module is configured to: calculate the change in nasal skin temperature, extract the temporal features of nasal skin temperature, and input the change in nasal skin temperature and temporal features into a pre-trained evaluation model to output a mental fatigue level score.

[0062] The fatigue state determination module is configured to compare the mental fatigue level score with a preset fatigue threshold and output the fatigue state determination result based on the comparison result.

[0063] Various modifications and variations of this invention will be apparent to those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A mental fatigue monitoring method based on alar skin temperature, characterized by, The method comprises the following steps: obtaining alar skin temperature data and environmental temperature data, and obtaining an initial alar skin temperature average value from the alar skin temperature data; preprocessing the alar skin temperature data, constructing a low-temperature compensation model based on the environmental temperature data and a preset environmental temperature threshold, and performing dynamic calibration to obtain an alar skin temperature threshold; when the environmental temperature data is less than the preset environmental temperature threshold, the low-temperature compensation model is started, and the constructed low-temperature compensation model is expressed as: ; wherein, represents a nasal ala skin temperature threshold, represents a nasal ala skin initial temperature average, represents a compensation coefficient, represents ambient temperature data; when the ambient temperature data is greater than and equal to an ambient temperature threshold, the hypothermia compensation model is not activated; based on the real-time alar skin temperature data and the alar skin temperature threshold, calculating an alar skin temperature change amount, extracting time-domain features of the alar skin temperature, and inputting the alar skin temperature change amount and the time-domain features into a pre-trained evaluation model to output a mental fatigue grade score; comparing the mental fatigue grade score with a preset fatigue threshold, and outputting a fatigue state judgment result according to the comparison result.

2. The mental fatigue monitoring method based on the alar skin temperature according to claim 1, wherein The calculated alar skin temperature change amount is expressed as: ; wherein, represents a change in temperature of the alar skin, represents real-time acquired alar skin temperature data, represents an alar skin temperature threshold value.

3. The mental fatigue monitoring method based on the alar skin temperature according to claim 1, wherein The extracted alar skin temperature time-domain features include an alar skin temperature change rate and a fluctuation amplitude, the alar skin temperature change rate is a rate of decrease of the alar skin temperature per unit time, and the fluctuation amplitude is a standard deviation of the alar skin temperature in a set time window.

4. The mental fatigue monitoring method based on the alar skin temperature according to claim 1, wherein The pre-trained evaluation model is expressed as: ; wherein, represents an age correction coefficient, represents a change in temperature of the skin of the alae nasi, represents a time interval, represents a rate of change of the temperature of the skin of the alae nasi, represents a fluctuation amplitude, represents a gender correction coefficient.

5. A mental fatigue monitoring method based on alar skin temperature as claimed in claim 1, wherein, The fatigue threshold is divided into a first fatigue threshold, a second fatigue threshold, and a third fatigue threshold; when the mental fatigue grade score is less than the first fatigue threshold, it is determined that the mental fatigue state is non-fatigue; when the mental fatigue grade score is greater than or equal to the first fatigue threshold and less than the second fatigue threshold, it is determined that the mental fatigue state is mild fatigue; when the mental fatigue grade score is greater than or equal to the second fatigue threshold and less than the third fatigue threshold, it is determined that the mental fatigue state is moderate fatigue; when the mental fatigue grade score is greater than the third fatigue threshold, it is determined that the mental fatigue state is severe fatigue.

6. The mental fatigue monitoring method based on alar skin temperature according to claim 1, wherein, After outputting the fatigue state judgment result, the alar skin temperature change amount, the time-domain features, and the fatigue state judgment result are transmitted to a mobile terminal through Bluetooth for display.

7. A mental fatigue monitoring system based on alar skin temperature, characterized by, The method comprises the following modules: a temperature acquisition module configured to acquire alar skin temperature data and environmental temperature data, and obtain an initial alar skin temperature average value from the alar skin temperature data; a data processing and calibration module configured to preprocess the alar skin temperature data, construct a low-temperature compensation model based on the environmental temperature data and a preset environmental temperature threshold, and perform dynamic calibration to obtain an alar skin temperature threshold; when the environmental temperature data is less than the preset environmental temperature threshold, the low-temperature compensation model is started, and the constructed low-temperature compensation model is expressed as: ; wherein, represents a nasal ala skin temperature threshold, represents a nasal ala skin initial temperature average, represents a compensation coefficient, represents ambient temperature data; when the ambient temperature data is greater than and equal to an ambient temperature threshold, the hypothermia compensation model is not activated; a mental fatigue grade score output module configured to calculate an alar skin temperature change amount based on real-time alar skin temperature data and the alar skin temperature threshold, extract time-domain features of the alar skin temperature, and input the alar skin temperature change amount and the time-domain features into a pre-trained evaluation model to output a mental fatigue grade score; a fatigue state judgment module configured to compare the mental fatigue grade score with a preset fatigue threshold, and output a fatigue state judgment result according to the comparison result.

8. A computer-readable storage medium having stored thereon a program, characterized in that, The program, when executed by the processor, implements the steps of a mental fatigue monitoring method based on temperature of the skin of the nasal wings according to any one of claims 1-6.

9. An electronic device comprising a memory, a processor, and a program stored on the memory and executable on the processor, characterized in that, The processor, when executing the program, implements the steps of a mental fatigue monitoring method based on temperature of the skin of the nasal wings according to any one of claims 1-6.

Images