A human blood oxygen saturation monitoring device

Abstract

The present invention relates to a human blood oxygen saturation monitoring device, comprising: a blood oxygen detection module for detecting a patient's blood oxygen saturation to obtain first blood oxygen data; a skin feature detection module for detecting the patient's skin condition and generating skin data; and a processing module for transmitting an alarm instruction to the processing module when the first blood oxygen data detected by the blood oxygen detection module falls below a threshold. The processing module, in response to the alarm instruction, receives skin data from the skin feature detection module and performs a first round of correction on the alarm instruction based on the skin data. The skin data is generated by the skin feature detection module by collecting images of the monitored person's skin condition changes at different times, and then based on the color depth of the skin condition change images.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and in particular to a human blood oxygen saturation monitoring device. Background Art

[0002] The color of the skin and mucous membranes changes with the color of the bloodstream. The red color of blood is due to the presence of hemoglobin within red blood cells. When hemoglobin is fully oxygenated, becoming oxyhemoglobin, it is bright red. When it releases oxygen, becoming deoxyhemoglobin, it becomes dark red. Because skin is thick and contains pigment, it appears white with a reddish tint or slightly brown with a reddish tint. Venous blood, containing more deoxyhemoglobin and less oxyhemoglobin, appears dark red, but when it passes through the skin, it appears bluish-purple. Therefore, cyanosis occurs when the amount of oxyhemoglobin in the mucous membranes, nails, and capillaries and arterioles in the skin decreases, while deoxyhemoglobin increases or denatured hemoglobin is present. Therefore, changes in skin color can provide valuable information for blood oxygen level measurements.

[0003] The perfusion index (PI), hereinafter referred to as the PI value, reflects the pulsating blood flow and, therefore, the blood perfusion capacity. In the paper "Research on a Pulse Oximeter Low Perfusion Algorithm," Ai Zhiguang notes that when the perfusion index is low and the pulse is very weak, the pulse wave signal detected by the pulse oximeter is extremely weak. Interference from environmental and circuit noise can drown out the useful signal, making it impossible to calculate various parameters.

[0004] Liu Fan conducted a study titled "Comparative Observation of the Dropout Rate of Two Types of Oxygen Saturation Probes in the ICU." The observation group used adhesive oximetry probes and received 619 rounds of inspections; the control group used overlay oximetry probes and received 579 rounds of inspections. Patients in both groups placed the probes at the designated fingertips. Clinical data and oximetry probe dropout rates were compared between the two groups. Results showed a statistically significant dropout rate of 0.81% in the observation group and 12.95% in the control group. The conclusion shows that adhesive oximetry probes in the ICU can effectively reduce dropout rates compared to overlay oximetry probes, are safer and more effective in clinical use, and are worthy of promotion.

[0005] Prior art, such as CN104887246B, provides a blood oxygen measurement method and blood oxygen measurement device, wherein the method includes the following steps: obtaining the distance between the emitting device of the blood oxygen measurement device and the fingertip of the measured finger during the measurement process, the red light peak-to-valley difference, the red light baseline value, the infrared light peak-to-valley difference, and the infrared light baseline value; calculating the blood oxygen calculation parameters based on the distance between the emitting device and the fingertip of the measured finger; calculating the ratio of the red light peak-to-valley difference to the red light baseline value and the ratio of the infrared light peak-to-valley difference to the infrared light baseline value; and calculating the corresponding blood oxygen value based on the ratio of the red light peak-to-valley difference to the red light baseline value, the ratio of the infrared light peak-to-valley difference to the infrared light baseline value, and the blood oxygen calculation parameters. The above method determines the position of the measured finger illuminated by the incident light source, then calculates the blood oxygen calculation parameters; and finally calculates the corresponding blood oxygen value, so that the blood oxygen value at the corresponding position can be measured more accurately according to the different measurement positions of the measured finger. The device obtains blood oxygen data by irradiating the fingers, but when used clinically, the finger cuffs may come off if the patient is in an agitated state. In addition, critically ill patients have poor peripheral circulation, and the detection data of the fingers is inaccurate. At the same time, when the patient's body parts are in a weak perfusion state, the detection data errors are large.

[0006] In addition, on the one hand, there are differences in understanding among those skilled in the art; on the other hand, the inventor studied a large number of documents and patents when making the present invention, but due to space limitations, not all details and contents are listed in detail. However, this does not mean that the present invention does not have the characteristics of these prior arts. On the contrary, the present invention already has all the characteristics of the prior art, and the applicant reserves the right to add relevant prior art to the background art. Summary of the Invention

[0007] Although pulse oximetry is widely used, its accuracy is often affected by external factors due to technical and engineering limitations. For example, different sensor structures can significantly affect the light propagation path, resulting in discrepancies in measurement results. Low temperatures, poor perfusion, different skin tones, and nail polish can cause the transmitted light intensity collected by the receiving unit to deviate from normal values. Strong ambient light, body motion, and high-frequency electrosurgery interference can introduce significant noise, drowning out the useful signal. Existing instruments are designed to measure oxygenated and deoxygenated hemoglobin but lack error correction for pathological hemoglobin levels (such as carboxyhemoglobin and methemoglobin). To eliminate external interference, users must remain stationary, pay attention to the operating environment, and select a sensor probe that fits the body well and is compatible with the main unit. Traditional pulse oximeter sensors are typically used in areas with dense vascular tissue, such as fingers, toes, and earlobes. Due to the complexity of the human body, traditional oximeters cannot promptly reflect physiological changes in tissues that do not rely on periodic arterial pulsation.

[0008] In response to the shortcomings of the devices proposed in the prior art, the present application proposes a human blood oxygen saturation monitoring device, including an adhesive attachment for being attached to the patient's skin surface. The adhesive attachment is made of anti-allergic material, which can prevent allergic patients from experiencing worsening of their condition due to allergies, and the adhesive attachment can be reused after cleaning and disinfection. A blood oxygen detection module is used to monitor the blood oxygen saturation of the monitored person and issue an alarm prompt instruction. The surface of the adhesive attachment is provided with a sticky substance, and the blood oxygen detection module is integrated into the surface of the adhesive attachment. It can be attached to the skin surface of the monitored person to avoid inaccurate monitoring due to poor peripheral circulation of the patient. While providing a more accurate measurement method, the adhesive blood oxygen saturation probe can effectively reduce the incidence of shedding compared to the insertable blood oxygen saturation probe, making it safer and more effective in clinical use.

[0009] According to a preferred embodiment, the monitoring device includes: a blood oxygen detection module for detecting a patient's blood oxygen saturation to obtain first blood oxygen data; a skin feature detection module for detecting the patient's skin condition and generating skin data; and a processing module for sending an alarm prompt instruction to the processing module when the first blood oxygen data detected by the blood oxygen detection module falls below a threshold value. Preferably, the threshold value is set to 90%; an alarm prompt instruction is issued when the blood oxygen saturation falls below 90%. Furthermore, because the monitored person's blood oxygen saturation may be at different levels of danger, the alarm prompt is divided into multiple levels. The first level of the alarm prompt condition is: blood oxygen saturation is below 55%; the second level of the alarm prompt condition is: blood oxygen saturation is between 55% and 65%; the third level of the alarm prompt condition is: blood oxygen saturation is between 65% and 80%; and the fourth level of the alarm prompt condition is: blood oxygen saturation is between 80% and 90%. The four levels of blood oxygen saturation can provide timely prompts to medical personnel in actual medical treatment, especially when rescuing the monitored person, and medical personnel can directly determine the degree of hypoxia of the monitored person through the prompt information.

[0010] According to a preferred embodiment, the adhesive attachment is provided with a water-absorbing layer, positioned adjacent to the adhesive material, for absorbing sweat to prevent moisture on the skin surface from reducing the adhesiveness of the adhesive material. In response to the alarm prompt instruction, the processing module receives skin data from the skin feature detection module and performs a first round of correction on the alarm prompt instruction based on the skin data. Preferably, the skin data is obtained by the skin feature detection module by collecting images of the monitored subject's skin condition changes at different times, and then converting the images into RGB color values ​​based on the color depth of the skin condition change images to create a curve fit. The detection steps are as follows: s1. Determine that the skin image of the same area at the first moment is the first skin image, and the skin image at the second moment is the second skin image; s2. Select the same area in the first and second skin images; s3. Divide the same area into numerous pixel blocks and collect color information for each pixel block; s4. Compare the color information of each pixel block in the collected second skin image with the color information of the pixel blocks in the first skin image and generate a color change curve based on the color depth. The color change curve can be updated in real time. The first moment and the second moment can be any moment, and there can be information from multiple moments for comparison, and the detection cycle can be to collect image information once per second and analyze it. Preferably, the RGB value refers to brightness and is represented by an integer. Normally, RGB each has 256 levels of brightness, represented by numbers from 0, 1, 2... to 255. According to calculations, 256 levels of RGB color can be combined to produce a total of about 16.78 million colors, that is, 256×256×256=16777216. By drawing a color change curve, the degree of darkening of the skin color of the monitored person is measured. The degree of darkening of the skin color can reflect the degree of deterioration of the monitored person's condition to a certain extent. At the same time, since what is measured is the degree of darkening of the skin color, compared to directly measuring the skin color, drawing a color change curve can exclude the situation where the monitored person's skin color is darker and has not deteriorated.

[0011] According to a preferred embodiment, the first round of correction is: when the blood oxygen detection module sends an alarm prompt instruction, the processor issues a first mode alarm instruction when the color of the skin state change image deepens based on the color deepening trend of the skin state change image in the skin data. When the blood oxygen saturation of the monitored person is low, the deoxyhemoglobin of the monitored person increases, or denatured hemoglobin appears, cyanosis will appear, and the skin will appear bluish purple. Therefore, when the skin color of the monitored person darkens, it indicates that the monitored person may be in a state of low blood oxygen or the condition continues to worsen. Based on this situation, after the processing module receives the alarm prompt instruction issued by the blood oxygen detection module, it makes a judgment based on the result of the first round of correction. If the color change curve shows a trend of extending in the dark direction, the processing module issues a first mode alarm instruction.

[0012] According to a preferred embodiment, it also includes a temperature detection module. The temperature detection module can monitor the epidermis of the monitored person by collecting epidermal temperature data at different times when the processor issues a first mode alarm instruction, and the temperature detection module can fit a temperature curve based on the high and low changes in the collected epidermal temperature data and send the temperature curve to the processing module. The temperature detection module can detect the epidermal temperature of the monitored person, and the detection steps are: s1. Determine the skin temperature of the same area of ​​the monitored epidermis at the first moment as the first skin temperature, and the skin temperature at the second moment as the second skin temperature; s2. Compare the collected second skin temperature with the first skin temperature and generate a temperature change curve chart based on the high and low temperatures. The temperature change curve chart can be updated in real time. The first moment and the second moment can be any moment, and there can be temperatures at multiple moments for comparison, and the detection cycle can be to collect temperature data once per second and perform analysis. By detecting the temperature, the blood oxygen saturation status of the monitored person can be further determined to determine whether the monitored person is in a weak perfusion state. When the skin color of the monitored person further darkens, by measuring the body temperature change curve of the monitored person, the phenomenon of short-term darkening of skin color caused by trauma and other conditions in a short period of time can be further ruled out. Therefore, the weak perfusion state can be determined more accurately, and the accuracy of the detection results is improved.

[0013] According to a preferred embodiment, the processing module is capable of receiving a temperature curve and performing a second round of correction on the first blood oxygen data based on the temperature curve. The second round of correction determines that the first blood oxygen data is inaccurate when the temperature curve approaches a low temperature, and determines that the first blood oxygen data is accurate when the temperature curve approaches a non-low temperature. The processor is capable of selecting a preset algorithm corresponding to the correction result based on the result of the second round of correction to calculate the second blood oxygen data. The preset algorithm is to use a weak perfusion algorithm when the processing module determines that the first blood oxygen data is inaccurate, and to use a normal state algorithm when the processing module determines that the first blood oxygen data is accurate. When the monitored person's blood oxygen saturation is low, deoxyhemoglobin increases, or denatured hemoglobin appears, the monitored person's body temperature will decrease. Therefore, when the monitored person's skin temperature decreases, it indicates that the monitored person may be in a state of hypoxemia or his condition continues to worsen. Based on this situation, after the processing module issues a first-mode alarm instruction, it makes a judgment based on the results of the first round of calibration. If the temperature change curve shows a trend extending toward a low temperature, the processing module determines that the monitored person is in a weak perfusion state, and the calculation of the first blood oxygen data is inaccurate. The processing module can calculate the second blood oxygen data according to the corresponding weak perfusion algorithm, and issue the second-mode alarm instruction based on the second blood oxygen data that is lower than the threshold. When the second blood oxygen data is higher than the threshold, the alarm instruction is not triggered. When the human body is in a normal state, a normal state algorithm is used. Preferably, the normal state algorithm can be to use red light with a wavelength of 660nm and infrared light with a wavelength of 960nm to alternately pass through human tissue. Based on the different absorption rates of oxygenated hemoglobin and reduced hemoglobin to the two different lights, two PI values ​​of different lights can be obtained. The red light PI value is divided by the infrared light PI value and recorded as R, then: because According to Lambert-Beer law, I = I0e -εCD , where D is the distance light penetrates in the solution, and ε is the absorption coefficient of the solution. When monochromatic light passes through human tissue, the amount of monochromatic light absorbed by venous blood and bloodless tissue is constant. However, since the heartbeat causes the optical path of arterial blood to change periodically, the amount of light absorbed by arterial blood also changes periodically. Therefore, the transmitted photoplethysmography (PPG) signal contains a large DC component value and a smaller AC component value that changes periodically with the pulsation. Therefore, it can be concluded that when monochromatic light irradiates and passes through human tissue, the formula for the transmitted light intensity is: Among them, ε0, εHbo2, and εHb represent the absorption coefficients of bloodless tissue, oxygenated hemoglobin, and reduced hemoglobin to light, respectively; C0 、 CHb o2, CHb represents the corresponding concentration coefficient; D0 Indicates the light transmission path in bloodless tissue.D Indicates the optical path of light in venous blood. ΔD It represents the optical path of light in arterial blood. From this, the DC component value of the transmitted light intensity can be obtained: AC component value of transmitted light intensity: I AC =II DC , the ratio of AC value to DC value can be calculated from the above formula: Because I AC Much smaller than I DC , taking the logarithm of both sides of the formula and finding the limit, we can calculate: Clinically, the ratio of the AC component to the DC component is called the Perfusion Index (PI), which reflects the body's blood perfusion capacity. A healthy individual's PI is generally greater than 3%, while a PI below 1% is considered poor blood oxygen perfusion. A poor perfusion algorithm can be a fundamental frequency positioning and narrowband filtering algorithm. This algorithm consists of two steps: fundamental frequency positioning and narrowband filtering. The former calculates the fundamental frequency through signal processing, while the latter performs a 0.5Hz narrowband bandpass filter based on the fundamental frequency to determine the fundamental amplitude. Simultaneously, a 0.1Hz low-pass filter is applied to the original signal to determine the DC signal amplitude. The fundamental frequency positioning step uses the PPG signal generated by infrared light as input; since the pulse wave AC signal itself is very weak during low perfusion, the original infrared light signal needs to be isolated and amplified at the beginning to isolate the DC signal and amplify the AC signal, so that the characteristics of the AC signal are more obvious and convenient for subsequent signal processing; since the average human pulse rate is about 30 to 150 times per minute, that is, a heart rate of 0.5Hz to 2.5Hz, the signal after DC isolation and amplification is then band-pass filtered with a passband of 0.5Hz to 2.5Hz, so as to retain the fundamental signal as much as possible and filter out signals outside the fundamental wave; correlation detection is widely used in weak signal extraction. It uses the difference between the periodicity of the signal to be extracted and the randomness of the noise signal to extract the periodic signal through autocorrelation operation. Although the autocorrelation signal loses the phase information compared to the original signal, the frequency is the same. Therefore, the fundamental wave signal after the band-pass filtering in the previous step is subjected to autocorrelation operation to calculate the frequency of the fundamental wave. The N-point autocorrelation operation formula is: The last step is to calculate the fundamental frequency obtained after the autocorrelation operation. This frequency is the fundamental frequency required in the fundamental frequency positioning step. When the fundamental frequency has been determined, it can be subjected to narrowband bandpass filtering to obtain the fundamental signal. The range of the narrowband bandpass filter can be set to 0.5Hz. On the one hand, the 0.5Hz range is narrow enough to obtain a good fundamental signal. On the other hand, since the fundamental frequency range is 0.5Hz-2.5Hz, only 4 narrowband bandpass filters with different parameters are used according to the 0.5Hz division, which reduces the hardware circuit overhead. The narrowband filtering step is to first isolate and amplify the PPG signal generated by the original red light and infrared light to obtain an amplified AC signal; then, according to the 0.5Hz frequency division range where the fundamental frequency is located, select the corresponding narrowband bandpass filter for filtering to obtain a fundamental signal with very little noise; find the amplitude of this fundamental signal, which is At the same time, the original red and infrared PPG signals are directly low-pass filtered with a cutoff frequency of 0.1Hz to obtain the DC signal of the PPG signal; the average value of this DC signal over a period of time can be obtained. Finally, according to the formula You can calculate the R value by substituting R into the formula SpO2 = a × R 2 +b×R+c, the current blood oxygen saturation value can be fitted, where a, b, and c are the coefficients obtained by fitting the quadratic curve with the actual SpO2 value after measuring R. Preferably, if the temperature change curve shows a trend of stable extension, the processing module determines that the monitored person is in a normal state, and can calculate the second blood oxygen data according to the corresponding normal state algorithm, and issue a second mode alarm instruction according to the second blood oxygen data below the threshold. So finally we can conclude Therefore, R can be calculated to calculate blood oxygen saturation.

[0014] According to a preferred embodiment, an alarm module is further included, which can emit a first mode alarm signal in response to the first mode alarm instruction emitted by the processing module. The first mode alarm signal can be a sound signal. Preferably, the sound is set to different decibels for differentiation. For example, it can be differentiated according to the different levels of danger of the monitored person. When the first level warning prompt condition is met, it indicates that the monitored person's blood oxygen saturation is lower than 55%, and the first mode alarm signal is prompted with a sound of 70 decibels; when the second level warning prompt condition is met, it indicates that the monitored person's blood oxygen saturation is 55% to 65%, and the first mode alarm signal is prompted with a sound of 60 decibels; when the third level warning prompt condition is met, it indicates that the monitored person's blood oxygen saturation is 65% to 80%, and the first mode alarm signal is prompted with a sound of 50 decibels; when the fourth level warning prompt condition is met, it indicates that the monitored person's blood oxygen saturation is 80% to 90%, and the first mode alarm signal is prompted with a sound of 40 decibels.

[0015] According to a preferred embodiment, the alarm module can respond to the second mode alarm instruction issued by the processing module and issue a second mode alarm signal. The second mode alarm signal can be a light. Preferably, the light is set to different colors for distinction. For example, it can be distinguished according to the different levels of danger stages of the monitored person. When the first level early warning prompt condition is met, it indicates that the monitored person's blood oxygen saturation is lower than 55%, and the second mode alarm signal is prompted in the form of red light; when the second level early warning prompt condition is met, it indicates that the monitored person's blood oxygen saturation is 55% to 65%, and the second mode alarm signal is prompted in the form of yellow light; when the third level early warning prompt condition is met, it indicates that the monitored person's blood oxygen saturation is 65% to 80%, and the second mode alarm signal is prompted in the form of green light; when the fourth level early warning prompt condition is met, it indicates that the monitored person's blood oxygen saturation is 80% to 90%, and the second mode alarm signal is prompted in the form of blue light.

[0016] According to a preferred embodiment, the sound signal can be distinguished according to the number of sound signal prompts within the same time interval, for example, according to the different degrees of danger stages of the monitored person, and the prompt is performed according to a cycle of two seconds of prompting / two seconds of silence. When the first-level early warning prompt condition is met, it indicates that the blood oxygen saturation of the monitored person is lower than 55%, and the first mode alarm signal is prompted with a cycle of four times within two seconds / two seconds of silence; when the second-level early warning prompt condition is met, it indicates that the blood oxygen saturation of the monitored person is 55%-65%, and the first mode alarm signal is prompted with a cycle of three times within two seconds / two seconds of silence; when the third-level early warning prompt condition is met, it indicates that the blood oxygen saturation of the monitored person is 65%-80%, and the first mode alarm signal is prompted with a cycle of two times within two seconds / two seconds of silence; when the fourth-level early warning prompt condition is met, it indicates that the blood oxygen saturation of the monitored person is 80%-90%, and the first mode alarm signal is prompted with a cycle of once within two seconds / two seconds of silence.

[0017] According to a preferred embodiment, the processing module is provided with a display area, which is provided with a light display window for displaying the alarm information of the light alarm. The display area is provided with a visualization display window for visually displaying the detected blood oxygen data, skin feature change data, and temperature change data. The visualization display can be in the form of numbers, bar charts, line graphs, or other display methods that can easily identify changes.

[0018] According to a preferred embodiment, the blood oxygen detection module, skin feature detection module, and temperature detection module are electrically connected to the processing module to achieve information transmission with the processing module. The processing module is connected to a power interface for connecting to an external power supply to provide power support.

[0019] The measurement principle of the present invention is a spectrophotometric method based on the Lambert-Beer law. Red light with a wavelength of 660 nm and infrared light at 940 nm are used, and the measurement is based on the phenomenon that oxyhemoglobin absorbs less 660 nm red light and more 940 nm infrared light, while hemoglobin absorbs the opposite. The ratio of infrared to red light absorption, measured spectrophotometrically, can determine the degree of hemoglobin oxygenation. Because the absorption coefficients of these two light sources are constant for skin, muscle, fat, venous blood, pigment, and bone, only the concentrations of oxyhemoglobin and hemoglobin in arterial blood flow change periodically with the arterial flow of blood, causing the signal intensity output by the photodetector to vary periodically. By processing these periodically varying signals, the corresponding blood oxygen saturation can be measured, and the pulse rate can also be calculated.

[0020] Compared with the prior art, the present invention has the following beneficial effects:

[0021] First: The device can be directly attached to the patient's skin, such as the forehead, for testing. When used clinically, it can prevent the patient from being in an agitated state, causing the traditional finger cuff device to detach, thereby causing inaccurate testing.

[0022] Second: This device can freely choose the attachment position according to the patient's condition, avoiding the situation where critically ill patients have poor peripheral circulation and inaccurate detection data at the finger site.

[0023] Third: This device uses multi-step verification and eliminates factors that affect the determination of weak perfusion status by measuring changes in skin condition and patient temperature, thereby effectively improving monitoring accuracy and reducing false alarms.

[0024] Fourth: The weak perfusion state is determined and a specific weak perfusion algorithm is used to perform a secondary calculation on the measured blood oxygen data, thereby reducing the error of the detection data. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic diagram of an existing blood oxygen saturation monitoring device;

[0026] Figure 2 This is a simplified structural diagram of a human blood oxygen saturation monitoring device of the present invention when in use;

[0027] Figure 3 It is a flow chart of the detection steps of the present invention.

[0028] Reference Signs List

[0029] 100: blood oxygen detection module; 200: alarm module; 300: display module. DETAILED DESCRIPTION

[0030] The present invention will be described in detail below with reference to the accompanying drawings.

[0031] Figure 1 This is a diagram of an existing finger-cuff-type blood oxygen saturation monitoring device.

[0032] Example 1

[0033] like Figure 2As shown, the present application proposes an adhesive attachment for sticking on the surface of the patient's skin. The adhesive attachment is made of anti-allergic material, which can prevent allergic patients from getting worse due to allergies, and the adhesive attachment can be reused after cleaning and disinfection. The blood oxygen detection module 100 is used to monitor the blood oxygen saturation of the monitored person and issue an alarm prompt instruction. The surface of the adhesive attachment is provided with a sticky substance. The blood oxygen detection module 100 is integrated into the surface of the adhesive attachment, which can be attached to the skin surface of the monitored person to avoid inaccurate monitoring due to poor peripheral circulation of the patient. The blood oxygen detection module 100 is used to monitor the blood oxygen saturation of the monitored person and issue an alarm prompt instruction when the first blood oxygen data obtained by monitoring is lower than a threshold. Preferably, the threshold is set to: blood oxygen saturation 90%, and an alarm prompt instruction is issued if it is lower than 90%. Furthermore, since the monitored person's blood oxygen saturation will be at different levels of danger, the alarm prompts are divided into multiple levels. The first-level warning prompt condition is: blood oxygen saturation is less than 55%; the second-level warning prompt condition is: blood oxygen saturation is 55%-65%; the third-level warning prompt condition is: blood oxygen saturation is 65%-80%; and the fourth-level warning prompt condition is: blood oxygen saturation is 80%-90%. The blood oxygen detection module 100 can be attached to the surface of the monitored person's skin for monitoring, thereby avoiding inaccurate monitoring due to poor peripheral circulation of the monitored person. A water-absorbing layer is provided on the adhesive part, and the water-absorbing layer is provided adjacent to the sticky material to absorb sweat to avoid the phenomenon that the stickiness of the sticky material is reduced due to moisture on the skin surface. It also includes a skin feature detection module for monitoring changes in the state of the monitored person's skin. The skin feature detection module is integrated into the adhesive part. Preferably, the skin feature detection module is capable of detecting the skin color of the monitored person, and it pre-stores standard color information. The detection steps are: s1. Determine the skin image of the same area at the first moment as the first skin image, and the skin image at the second moment as the second skin image; s2. Select the same area on the first skin image and the second skin image; s3. Divide the same area into countless pixel blocks and collect the color information of each pixel block; s4. Compare the color information of each pixel block of the collected second skin image with the color information of the pixel block of the first skin image and generate a color change curve chart based on the depth of the color. The color change curve chart can be updated in real time. The first moment and the second moment can be any moment, and there can be information from multiple moments for comparison, and the detection cycle can be to collect image information and analyze it once per second.The processing module receives skin data from the skin feature detection module in response to the alarm prompt instruction, and performs a first round of correction on the alarm prompt instruction based on the skin data, so that when the blood oxygen detection module (100) sends the alarm prompt instruction, the processor issues a first mode alarm instruction when the color of the skin state change image deepens according to the color deepening trend of the skin state change image in the skin data. When the monitored person's blood oxygen saturation is low, the monitored person's deoxyhemoglobin increases, or denatured hemoglobin appears, cyanosis will appear, and the skin will appear bluish purple. Therefore, when the monitored person's skin color darkens, it indicates that the monitored person may be in a low blood oxygen state or the condition continues to worsen. Based on this situation, after the processing module receives the alarm prompt instruction issued by the blood oxygen detection module 100, it makes a judgment based on the result of the first round of correction. If the color change curve shows a trend of extending in the dark direction, the processing module issues a first mode alarm instruction. The display area can visually display the skin state change data and temperature change data. The visual display method is a line graph.

[0034] Example 2

[0035] like Figure 2As shown, the device may also include a temperature detection module. When the processor issues a first mode alarm instruction, the temperature detection module can monitor the epidermis of the monitored person by collecting epidermal temperature data at different times. The temperature detection module can fit a temperature curve based on the high and low changes in the collected epidermal temperature data and send the temperature curve to the processing module. The temperature detection module can detect the epidermal temperature of the monitored person. The detection steps are: s1. Determine the skin temperature of the same area of ​​the monitored epidermis at the first moment as the first skin temperature and the skin temperature at the second moment as the second skin temperature; s2. Compare the collected second skin temperature with the first skin temperature and generate a temperature change curve based on the temperature. The temperature change curve can be updated in real time. The first moment and the second moment can be any moment, and there can be temperatures at multiple moments for comparison. The detection cycle can be to collect temperature data and analyze it once per second. The processing module is capable of receiving a temperature curve and performing a second round of correction on the first blood oxygen data based on the temperature curve. The second round of correction determines that the first blood oxygen data is inaccurate when the temperature curve approaches a low temperature, and determines that the first blood oxygen data is accurate when the temperature curve approaches a non-low temperature. When the monitored person's blood oxygen saturation is low, deoxyhemoglobin increases, or denatured hemoglobin appears, the monitored person's body temperature will decrease. Therefore, when the monitored person's skin temperature decreases, it indicates that the monitored person may be in a state of hypoxemia or a worsening condition. Based on this situation, after the processing module issues a first mode alarm instruction, it determines based on the results of the first round of correction that if the temperature change curve shows a trend toward a low temperature, the processing module determines that the monitored person is in a state of poor perfusion, and the first blood oxygen data is inaccurate. Therefore, the processing module calculates second blood oxygen data according to the corresponding poor perfusion algorithm, and issues a second mode alarm instruction based on the second blood oxygen data below a threshold. When the second blood oxygen data exceeds the threshold, the processing module determines that the first blood oxygen data is inaccurate. Preferably, if the temperature change curve shows a trend of steady extension, the processing module determines that the monitored person is in a normal state, and can calculate the second blood oxygen data according to the corresponding normal state algorithm, and issue a second mode alarm instruction based on the second blood oxygen data that is lower than the threshold.

[0036] Example 3

[0037] like Figure 2As shown, the processing module is provided with a display module 300, and the display module 300 is provided with a light display window for displaying the alarm information of the light alarm. The display module 300 is provided with a visual display window for visually displaying the detected blood oxygen data, skin feature change data, and temperature change data. The visual display method can be digital or a bar chart, line chart, or other display method that can easily show the changes. The blood oxygen detection module 100, skin feature detection module, and temperature detection module are electrically connected to the processing module to realize information transmission with the processing module. The processing module is connected to a power interface for connecting to an external power supply to provide power support.

[0038] Example 4

[0039] A method for monitoring human blood oxygen saturation includes an ambient light source exclusion module and a sensor capable of collecting the intensity, wavelength, and composition of the ambient light source. The sensor analyzes the intensity, wavelength, and composition of the ambient light to derive absorbance data, as well as the effect of the ambient light source on skin color, to zero the device and eliminate the influence of the ambient light source on the device. This eliminates the influence of ambient light on skin color recognition. A data model with multiple color blocks is built into the model. Computer simulation of images of different light sources illuminating different color blocks is used to compare the collected images of the skin after being illuminated by the ambient light source, thereby eliminating the influence of the ambient light source.

[0040] First, the skin feature detection module is activated. The detection module extracts image information from the patient's epidermis and performs color comparison to obtain the patient's skin color data. The processing unit has a built-in color data model that can classify the skin color data. Preferably, skin color is divided into four groups: very light skin, light skin, medium skin, and very dark skin. The device is zeroed using the color data model and, combined with the ethnicity information input by the terminal, the device is subjected to ethnicity correction. Abnormal skin color can be caused by pigment, blood vessels, or dermal scarring. A Wood's UV lamp can be used to determine skin color changes. When the ultraviolet light from a Wood's lamp penetrates the epidermis, it is absorbed by the pigment within the epidermis. In vitiligo lesions, the epidermal pigment is lost, and all light is reflected back, resulting in bluish-white patches that are brighter than the surrounding normal areas. In contrast, lesions caused by hyperpigmented diseases such as leucoderma have increased basal pigment, which absorbs ultraviolet light, making the affected area appear darker than the surrounding normal skin. Skin that appears whiter under natural light due to scarring or collagen hyperplasia has normal epidermal melanin content and, under a Wood's lamp, its color is indistinguishable from the surrounding normal skin. Similarly, skin color changes caused by vascular disease are no different from normal skin under ultraviolet light. However, due to the high content of melanin in the epidermis of black people, their skin color has a significant difference under Wood's light.

[0041] Chronic hypoxia is the most common of the three types of hypoxia in daily life. While subtle, it can easily lead to complications, accelerate aging, and even be life-threatening in severe cases. Chronic hypoxia exists in everyone to varying degrees, and its severity increases with age. Therefore, normal blood oxygen partial pressure varies among people of different ages, making false alarms prone to occur when using the same measurement method. Therefore, the skin feature detection module can also measure the smoothness of the skin surface. Sensors installed on the surface can detect the skin's surface shape, determining the patient's condition based on the number of skin wrinkles and changes in epidermal shape. A uniform epidermal shape curve indicates a young individual, while a tortuous curve indicates an elderly individual. Different alarm thresholds are used for each age group. Preferably, the normal ranges are set as follows: 20-29 years old: blood oxygen partial pressure 84-107 mmHg; 30-39 years old: blood oxygen partial pressure 81-101 mmHg; 40-49 years old: blood oxygen partial pressure 78-99 mmHg; 50-59 years old: blood oxygen partial pressure 74-94 mmHg; 60-69 years old: blood oxygen partial pressure 71-91 mmHg. An alarm is triggered only when the detected blood oxygen partial pressure data is outside the corresponding age range.

[0042] After the skin feature detection module completes the operation, the temperature detection module starts. The temperature detection module can calculate the temperature of the patient's skin surface by the amount of temperature drop during the time range of contact between the surface material and the skin. A temperature model is set in the processing unit, which can first be classified according to the measured temperature data, among which below 36.3℃ is the low temperature zone, 36.3℃~37.2℃ is the normal zone, and above 37.2℃ is the high temperature zone. Combined with the temperature data model, the device is zeroed according to the measured temperature data. Among them, the processing unit can combine the temperature data and color data for analysis to determine whether the patient is in a weak perfusion state. If the patient is in a weak perfusion state, it switches to the weak perfusion mode and uses the weak perfusion algorithm to process the blood oxygen data measured subsequently.

[0043] Example 5

[0044] like Figure 3 As shown, in actual monitoring, the detection steps can be:

[0045] S1. Ambient light source detection: detects light intensity and light composition, compares them with color blocks, and then characterizes the ambient light source;

[0046] S2. Skin color detection to determine ethnicity;

[0047] S3. The window asks whether the person is of the same race. If yes, execute Algorithm 1; otherwise, execute Algorithm 2.

[0048] S4. Skin smoothness detection and age determination;

[0049] S5. Temperature detection, determination of body temperature information;

[0050] S6. Correct all the results and calculate the blood oxygen data using the corresponding algorithm.

[0051] Example 6

[0052] The alarm module 200 can respond to the first mode alarm instruction issued by the processing module and issue a first mode alarm signal. The first mode alarm signal can be a sound signal. Preferably, the sound is set to different decibels for differentiation. For example, it can be differentiated according to the different levels of danger of the monitored person. When the first level warning prompt condition is met, it indicates that the monitored person's blood oxygen saturation is lower than 55%, and the first mode alarm signal is prompted with a sound of 70 decibels; when the second level warning prompt condition is met, it indicates that the monitored person's blood oxygen saturation is 55% to 65%, and the first mode alarm signal is prompted with a sound of 60 decibels; when the third level warning prompt condition is met, it indicates that the monitored person's blood oxygen saturation is 65% to 80%, and the first mode alarm signal is prompted with a sound of 50 decibels; when the fourth level warning prompt condition is met, it indicates that the monitored person's blood oxygen saturation is 80% to 90%, and the first mode alarm signal is prompted with a sound of 40 decibels.

[0053] It should be noted that the above-described specific embodiments are illustrative only. Those skilled in the art may devise various solutions based on the disclosure of the present invention, and such solutions fall within the scope of the present invention and are intended to be protected by the present invention. Those skilled in the art should understand that the present description and its accompanying drawings are intended to be illustrative only and are not intended to limit the scope of the claims. The scope of protection of the present invention is defined by the claims and their equivalents.

Claims

1. A human blood oxygen saturation monitoring device, comprising: The blood oxygen detection module is used to detect the patient's blood oxygen saturation to obtain the first blood oxygen data. Skin feature detection module, used to detect the patient's skin condition and generate skin data, It is characterized by: The monitoring device also includes a processing module, When the first blood oxygen data detected by the blood oxygen detection module is lower than the threshold, an alarm prompt instruction is sent to the processing module. The processing module receives skin data from the skin feature detection module in response to the alarm prompt instruction, and performs a first round of correction on the alarm prompt instruction based on the skin data, wherein the skin data is a curve fitted by the skin feature detection module by collecting skin state change images of the monitored person at different times in the same area, and then converting them into RGB color values ​​according to the color depth of the skin state change images. The skin image of the same area at the first moment is the first skin image, and the skin image at the second moment is the second skin image. The same area is selected on the first skin image and the second skin image, and the same area is divided into a number of pixel blocks. The color information of each pixel block is collected, and the color information of each pixel block of the collected second skin image is compared with the color information of the pixel block of the first skin image, and a real-time updated color change curve is generated according to the color depth. The first round of calibration is: when the blood oxygen detection module sends an alarm prompt instruction, the processing module issues a first mode alarm instruction when the color of the skin state change image in the skin data deepens according to the color trend of the skin state change image; After the processing module issues the first mode alarm instruction, it makes a judgment based on the results of the first round of calibration. If the temperature change curve shows a trend extending toward low temperature, the processing module determines that the monitored person is in a weak perfusion state, and the first blood oxygen data is inaccurate. The processing module calculates the second blood oxygen data according to the corresponding weak perfusion algorithm, and issues the second mode alarm instruction based on the second blood oxygen data that is lower than the threshold. When the second blood oxygen data is higher than the threshold, if the temperature change curve shows a trend of steady extension, the processing module determines that the monitored person is in a normal state, calculates the second blood oxygen data according to the corresponding normal state algorithm, and issues the second mode alarm instruction based on the second blood oxygen data that is lower than the threshold.

2. The human blood oxygen saturation monitoring device according to claim 1, characterized in that: It also includes a temperature detection module, which can monitor the skin of the monitored person by collecting skin temperature data at different times when the processing module issues a first mode alarm instruction, and the temperature detection module can fit a temperature curve according to the high and low changes of the collected skin temperature data and send the temperature curve to the processing module.

3. The human blood oxygen saturation monitoring device according to claim 2, characterized in that: The processing module is capable of receiving the temperature curve and performing a second round of correction on the first blood oxygen data based on the temperature curve. The second round of correction is to determine that the first blood oxygen data is inaccurate when the trend of the temperature curve approaches a low temperature, and to determine that the first blood oxygen data is accurate when the trend of the temperature curve approaches a non-low temperature.

4. The human blood oxygen saturation monitoring device according to claim 3, characterized in that: The processing module can select a preset algorithm corresponding to the correction result according to the result of the second round of correction to calculate the second blood oxygen data, wherein the preset algorithm is a weak perfusion algorithm when the processing module determines that the first blood oxygen data is inaccurate, and a normal state algorithm when the processing module determines that the first blood oxygen data is accurate.

5. The human blood oxygen saturation monitoring device according to claim 4, characterized in that: The processing module issues a second mode alarm instruction when the second blood oxygen data calculated by the preset algorithm is lower than a threshold.

6. The human blood oxygen saturation monitoring device according to claim 5, characterized in that: It also includes an alarm module (200) which can issue a first mode alarm signal in response to the first mode alarm instruction issued by the processing module.

7. The human blood oxygen saturation monitoring device according to claim 6, characterized in that: The alarm module (200) is capable of responding to the second mode alarm instruction issued by the processing module and issuing a second mode alarm signal.

8. The human blood oxygen saturation monitoring device according to claim 7, characterized in that: It also includes a display module (300), which is capable of receiving the skin data sent by the skin feature detection module and the temperature curve sent by the temperature detection module, and performing graphical display of the skin data and the temperature curve.

Images