A method for judging weak vital signs based on infrared pair tube light technology

Abstract

The application discloses a kind of weak vital sign discrimination methods based on infrared pair tube light technology, it is related to vital sign detection technology field, the method is applied to the detection device that can be worn on the body of the corpse finger without heart rate, temperature and touch no vital sign, based on the optical signal after infrared light and finger biological tissue interaction, the discrimination of weak vital sign is realized, and when identifying weak life state such as pseudo-death, automatic triggering alarm, the method comprises: collecting and decomposing mixed light signal obtains direct current and alternating component;Based on the time domain stability of alternating component signal, the preliminary effective signal segment is screened, and the microcirculation physiological coupling coefficient is calculated to distinguish target signal segment;The normalized cross-correlation function peak value of alternating component signal in target signal segment is calculated as cross-correlation peak value intensity, and the steady-state volume index is calculated according to the mean value of direct current component signal;When cross-correlation peak value intensity and steady-state volume index both satisfy preset threshold condition, it is determined that there is weak vital sign.

Description

Technical Field

[0001] This invention relates to the field of vital sign detection technology, specifically a method for judging weak vital signs based on infrared photodiode illumination technology. Background Technology

[0002] Vital signs monitoring is a key technology in clinical medicine, emergency medicine, and health management. In extreme situations such as cardiac arrest, severe trauma, and drowning, individuals may enter a state of "apparent death," where traditional examination methods fail to detect vital signs, but faint life activities still exist. Timely and accurate identification of these faint vital signs is crucial for initiating resuscitation and reducing mortality.

[0003] Infrared light detection technology, especially methods based on photoplethysmography, has become an important technique in the field of non-invasive vital sign monitoring. This technology typically uses light of a specific wavelength to irradiate biological tissue and obtains physiological parameters such as heart rate by detecting changes in light absorption or scattering signals caused by changes in blood volume. It boasts advantages such as being non-invasive, continuous, and easy to operate, playing a crucial role in routine health monitoring, exercise physiology, and clinical surveillance. However, in extreme physiological and clinical states such as apparent death, deep shock, severe blood loss, or deep postoperative sedation, an individual's vital signs may become extremely weak, with significantly reduced physiological signal amplitudes and potentially exhibiting atypical or nearly absent periodic characteristics. In such scenarios, background noise caused by the tissue itself, ambient light, or motion is relatively amplified, leading to a sharp deterioration in the signal-to-noise ratio of the effective physiological signal. Traditional signal processing methods typically rely on the clear identification and amplitude analysis of periodic pulse waveforms; however, when the signal periodicity is weak or disordered, their analytical efficiency and reliability face severe challenges. Therefore, in order to achieve effective monitoring of vital signs under a wider range of extreme conditions, it is urgent to develop a new method that can robustly distinguish weak and potentially non-periodic vital sign signals in environments with extremely low signal-to-noise ratios, so as to improve the detection capability and reliability of relevant monitoring equipment under critical conditions. Summary of the Invention

[0004] In view of the shortcomings of the prior art described above, the purpose of this invention is to provide a method for judging weak vital signs based on infrared photodiode illumination technology, so as to solve the above-mentioned technical problems.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for judging weak vital signs based on infrared photodiode illumination technology, comprising: S1: Synchronously acquire the dual-wavelength mixed light signal reflected by biological tissue and decompose it into DC component signal and AC component signal corresponding to each wavelength. S2: Based on the time-domain stability of dual-wavelength AC component signals, preliminary effective signal segments are selected from continuous signals; S3: Calculate the normalized cross-energy spectrum of the dual-wavelength AC component signal segment in the preliminary effective signal segment within the preset physiological frequency band, and perform weighted summation to obtain the microcirculation physiological coupling coefficient. Based on the microcirculation physiological coupling coefficient, identify the target signal segment with physiological rationality. S4: Calculate the normalized cross-correlation function of the dual-wavelength AC component signal segment in the target signal segment, and extract the peak value of the normalized cross-correlation function as the cross-correlation peak intensity. S5: Calculate the steady-state volume index based on the ratio of the mean values ​​of the dual-wavelength DC component signal segments in the target signal segment through a preset mapping relationship; S6: Compare the cross-correlation peak intensity and steady-state volume index with the corresponding preset thresholds respectively. When the comparison results of the two meet the preset conditions at the same time, it is determined that there are weak vital signs.

[0006] The present invention is further configured such that S1 includes: A first light-emitting device that emits a preset first wavelength and a second light-emitting device that emits a preset second wavelength are synchronously driven at a preset sampling frequency. A single photodetector synchronously receives dual-wavelength mixed optical signals after being processed by biological tissue and converts them into mixed analog electrical signals. By performing cross-impedance amplification, gain adjustment, and analog-to-digital conversion on the hybrid analog electrical signal, a hybrid discrete digital sequence is obtained. Based on the driving timing of the first light-emitting device and the second light-emitting device, a first discrete digital sequence corresponding to the first wavelength and a second discrete digital sequence corresponding to the second wavelength are separated from the mixed discrete digital sequence; A low-pass digital filter with a preset cutoff frequency is used to filter the first discrete digital sequence and the second discrete digital sequence respectively to obtain the first wavelength DC component signal and the second wavelength DC component signal. Subtracting the first wavelength DC component signal from the first discrete digital sequence yields the first wavelength AC component signal, and subtracting the second wavelength DC component signal from the second discrete digital sequence yields the second wavelength AC component signal.

[0007] The present invention is further configured such that S2 includes: An analysis window of a preset duration is slid across a continuous sequence of the first wavelength AC component signal and the second wavelength AC component signal with a preset step size. For each analysis window position, the first window segment of the first wavelength AC component signal and the second window segment of the second wavelength AC component signal are respectively extracted; The stability indexes of the first window segment and the second window segment are calculated respectively, wherein the stability indexes are obtained based on the reciprocal of the normalized root mean square error between each window segment and its corresponding smoothed window segment. The stability indices of the first window segment and the second window segment are compared with preset stability thresholds respectively; When the stability indices of both the first and second window segments are greater than the stability threshold, the first wavelength AC component signal segment, the second wavelength AC component signal segment corresponding to the current analysis window position, and the first wavelength DC component signal segment and the second wavelength DC component signal segment within the same time interval are extracted together as preliminary effective signal segments.

[0008] The present invention is further configured such that S3 includes: Perform short-time Fourier transform on the first wavelength AC component signal segment and the second wavelength AC component signal segment contained in each preliminary effective signal segment to obtain the corresponding first time-frequency matrix and second time-frequency matrix; Within the preset physiological frequency band, the normalized cross-energy spectrum at each time frame and frequency point is calculated based on the first time-frequency matrix and the second time-frequency matrix. Based on the preset physiological feature weighting function, the normalized cross-energy spectrum within the preset physiological frequency band is weighted and summed to obtain the microcirculation physiological coupling coefficient of the preliminary effective signal segment. When the microcirculation physiological coupling coefficient is greater than or equal to the preset physiological coupling threshold, the preliminary effective signal segment is determined to be physiologically reasonable and is identified as the target signal segment.

[0009] The present invention is further configured such that S4 includes: For each target signal segment, calculate the normalized cyclic cross-correlation function sequence of the first wavelength AC component signal segment and the second wavelength AC component signal segment contained therein within the preset time delay range; Extract the peak value with the largest absolute value from the normalized cyclic cross-correlation function sequence; Based on the statistical properties of the normalized cyclic cross-correlation function sequence, the significance threshold is determined; If the absolute value of the peak value is greater than the significance threshold, then the absolute value of the peak value is taken as the cross-correlation peak intensity of the target signal segment; If the absolute value of the peak is less than or equal to the significance threshold, the cross-correlation peak intensity of the target signal segment is set to zero.

[0010] The present invention is further configured such that S5 includes: For each target signal segment, extract the first wavelength DC component signal segment and the second wavelength DC component signal segment contained therein; Calculate the average values ​​of the first wavelength DC component signal segment and the second wavelength DC component signal segment after removing the preset beginning and end transition regions; Calculate the ratio of the average value of the first wavelength DC component signal segment to the average value of the second wavelength DC component signal segment, and map the ratio to the corresponding steady-state volume index according to a preset mapping relationship.

[0011] The present invention is further configured such that S6 includes: The cross-correlation peak intensity corresponding to each target signal segment is compared with a preset intensity threshold. The steady-state volume index corresponding to each target signal segment is compared with the preset lower and upper thresholds. When the peak intensity of the cross-correlation is greater than the intensity threshold and the steady-state volume index is within the numerical range defined by the lower threshold and the upper threshold, it is determined that there are transient weak vital signs at the corresponding moment of the target signal segment.

[0012] The present invention is further configured to count the cumulative number of times that are determined to have momentary weak vital signs within a sliding time window of a preset time length; If the cumulative number of times exceeds the preset threshold and the current judgment result is that there are momentary weak vital signs, then it is determined that the weak vital signs continue to exist and an alarm signal is triggered. If the cumulative number of alarms is less than or equal to the threshold number, the alarm signal that has been triggered will not be triggered or will be revoked.

[0013] The present invention is further configured such that the triggering alarm signal is a triggering acoustic and optical alarm device.

[0014] The present invention is further configured such that the method further includes generating visual indication information, the visual indication information being used to characterize the determination result of weak vital signs.

[0015] This invention provides a method for identifying weak vital signs based on infrared photodiode illumination technology. The method involves: S1: Simultaneously acquiring dual-wavelength mixed light signals reflected from biological tissue and decomposing them into DC and AC components corresponding to each wavelength; S2: Based on the temporal stability of the dual-wavelength AC component signals, filtering out preliminary effective signal segments from the continuous signal; S3: Calculating the normalized cross-energy spectrum of the dual-wavelength AC component signal segments within a preset physiological frequency band in the preliminary effective signal segments, and performing weighted summation to obtain the microcirculation physiological coupling coefficient, identifying physiologically reasonable target signal segments based on the microcirculation physiological coupling coefficient; S4: Calculating the normalized cross-correlation function of the dual-wavelength AC component signal segments in the target signal segments, and extracting the peak value of the normalized cross-correlation function as the cross-correlation peak intensity; S5: Calculating the steady-state volume index based on the ratio of the means of the dual-wavelength DC component signal segments in the target signal segments through a preset mapping relationship; S6: Comparing the cross-correlation peak intensity and the steady-state volume index with corresponding preset thresholds. When both comparison results simultaneously meet preset conditions, the presence of weak vital signs is determined. The beneficial effects include: 1. By constructing a multi-level processing mechanism based on synchronous acquisition and separation of dual-wavelength signals, time-domain stability screening, time-frequency domain physiological coupling analysis, time-domain cross-correlation verification, and steady-state optical property analysis, we have achieved effective extraction of weak feature signals related to microcirculation blood flow from a strong noise background, thereby achieving high-sensitivity detection of extremely weak vital signs. Even when the tested individual is in an extreme state such as "apparent death" where traditional methods are difficult to respond to, this method can still reliably identify weak vital signs based on the temporal correlation and optical property analysis of blood. 2. By introducing the microcirculation physiological coupling coefficient and combining the initial screening of time-domain stability with the verification of cross-correlation peak intensity, a multi-joint discrimination mechanism was constructed, which effectively improved the specificity and anti-interference ability of the method. This mechanism can reliably distinguish signals from real physiological activities from non-physiological interference with similar stability, thereby suppressing misjudgments caused by motion artifacts, sudden changes in ambient light or equipment noise, and ensuring the high reliability of alarm signals. 3. By adopting a steady-state volume index that does not rely on periodic pulse signals, which is generated by mapping the dual-wavelength DC light absorption ratio through a preset model, it is possible to effectively evaluate the steady-state optical properties of blood even in extreme physiological states where blood circulation is extremely weak and periodic pulse signals are almost absent.

[0016] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1 A flowchart illustrating an exemplary embodiment of the present invention is provided, showing a method for identifying weak vital signs based on infrared photodiode illumination technology; Figure 2 , Figure 3 and Figure 4 A schematic diagram of a vital signs detection device; Among them, 1 is a ring, 2 is a PCB board, 3 is a silicone pad, 4 is a button, 5 is a switch, 6 is a photodetector, 7 is the first light-emitting device, 8 is the second light-emitting device, 9 is a buzzer, and 10 is a light-emitting diode. Detailed Implementation

[0018] The embodiments of the present invention will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0019] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0020] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0021] A method for identifying weak vital signs based on infrared photodiode illumination technology, such as... Figure 1 As shown, it includes: S1: Synchronously acquire the dual-wavelength mixed light signal reflected by biological tissue and decompose it into DC component signal and AC component signal corresponding to each wavelength. S2: Based on the time-domain stability of dual-wavelength AC component signals, preliminary effective signal segments are selected from continuous signals; S3: Calculate the normalized cross-energy spectrum of the dual-wavelength AC component signal segment in the preliminary effective signal segment within the preset physiological frequency band, and perform weighted summation to obtain the microcirculation physiological coupling coefficient. Based on the microcirculation physiological coupling coefficient, identify the target signal segment with physiological rationality. S4: Calculate the normalized cross-correlation function of the dual-wavelength AC component signal segment in the target signal segment, and extract the peak value of the normalized cross-correlation function as the cross-correlation peak intensity. S5: Calculate the steady-state volume index based on the ratio of the mean values ​​of the dual-wavelength DC component signal segments in the target signal segment through a preset mapping relationship; S6: Compare the cross-correlation peak intensity and steady-state volume index with the corresponding preset thresholds respectively. When the comparison results of the two meet the preset conditions at the same time, it is determined that there are weak vital signs.

[0022] The present invention is further configured such that S1 includes: The first light-emitting device (7) that emits a preset first wavelength and the second light-emitting device (8) that emits a preset second wavelength are synchronously driven at a preset sampling frequency. The dual-wavelength mixed light signal after being acted upon by biological tissue is received synchronously by a single photodetector (6) and converted into a mixed analog electrical signal; By performing cross-impedance amplification, gain adjustment, and analog-to-digital conversion on the hybrid analog electrical signal, a hybrid discrete digital sequence is obtained. Based on the driving timing of the first light-emitting device (7) and the second light-emitting device (8), the first discrete digital sequence corresponding to the first wavelength and the second discrete digital sequence corresponding to the second wavelength are separated from the mixed discrete digital sequence; A low-pass digital filter with a preset cutoff frequency is used to filter the first discrete digital sequence and the second discrete digital sequence respectively to obtain the first wavelength DC component signal and the second wavelength DC component signal. The first wavelength DC component signal is subtracted from the first discrete digital sequence to obtain the first wavelength AC component signal, and the second wavelength DC component signal is subtracted from the second discrete digital sequence to obtain the second wavelength AC component signal; specifically, the first light-emitting device (7), the second light-emitting device (8), and the photodetector (6) are miniaturized and integrated into the inner optical window of a wearable vital signs detection device. The vital signs detection device is worn on the finger of a deceased person who has no heart rate, no body temperature, and no vital signs. Its light-emitting surface and light-receiving surface are configured to form a tight optical contact with the wearer's finger skin surface during use to establish a stable optical coupling channel; the first light-emitting device (7) The first light-emitting device (8) is configured to emit a light signal of a preset first wavelength located in the red spectral region, and the second light-emitting device (8) is configured to emit a light signal of a preset second wavelength located in the near-infrared spectral region. A driving circuit controlled by a precision clock source alternately and synchronously illuminates the two light-emitting devices according to a preset fixed timing and frequency. The emitted light enters the finger biological tissue through the optical window and undergoes absorption, scattering and reflection processes inside the biological tissue. The photodetector (6) is configured to receive the mixed light signal containing dual-wavelength light intensity attenuation information reflected from inside the finger biological tissue and convert the mixed light signal into a continuous mixed analog electrical signal output, thereby completing the process from optical signal to digital signal. The initial conversion from analog to digital signal; the mixed analog electrical signal is converted into a mixed analog voltage signal via a transimpedance amplifier. To accommodate differences in skin optical characteristics and wearing conditions among individuals, a programmable gain amplifier is used to adaptively adjust the gain of this mixed analog voltage signal to ensure that its dynamic range matches the input range of the subsequent analog-to-digital converter (ADC); the conditioned mixed analog voltage signal is then fed into the ADC, whose operating clock is strictly synchronized with the time-division driving timing of the light-emitting devices, and converts the continuous mixed analog voltage signal into a discrete digital point sequence with a preset quantization precision, thereby generating a mixed discrete digital sequence arranged in time sequence; based on the two light-emitting devices being... The precise synchronization timing of the lighting is analyzed and separated from the mixed discrete digital sequence to obtain the first discrete digital sequence corresponding to the lighting period of the first light-emitting device (7) and the second discrete digital sequence corresponding to the lighting period of the second light-emitting device (8). This timing synchronization separation mechanism ensures the correspondence and accuracy of the dual-wavelength optical signal data acquisition. In order to separate the background part that characterizes the static light absorption characteristics of tissue and the fluctuation part that characterizes the dynamic blood flow changes in the signal, the separated first discrete digital sequence and the second discrete digital sequence are digitally filtered respectively. A low-pass digital filter with a flat passband frequency response is used, and its cutoff frequency is preset to an extremely low value to retain the components in the signal that change extremely slowly without distortion.To eliminate phase distortion or time delay introduced by the filtering process, zero-phase filtering technology is used. The first and second discrete digital sequences are processed sequentially as follows: a first digital filter is performed in the forward time direction; the result sequence of the first filter is inverted on the time axis; the inverted sequence is then filtered a second time using a digital filter with the same transfer function as the first filter; the result sequence of the second filter is inverted on the time axis again, thus obtaining the final filtered sequence without phase distortion. The sequence obtained through this zero-phase filtering process is the DC component signal, which includes a first wavelength DC component signal and a second wavelength DC component signal. Its physical meaning mainly represents the stable absorption of light by tissues such as skin, subcutaneous tissue, and non-pulsatile blood. The first and second discrete digital sequences are subtracted point-by-point from their corresponding DC component signals obtained through low-pass zero-phase filtering to obtain the corresponding difference sequence. This difference sequence is the AC component signal, which includes a first wavelength AC component signal and a second wavelength AC component signal. The AC component signal contains information about periodic or non-periodic light absorption fluctuations caused by changes in microvascular blood volume. ;

[0023] The present invention is further configured such that S2 includes: An analysis window of a preset duration is slid across a continuous sequence of the first wavelength AC component signal and the second wavelength AC component signal with a preset step size. For each analysis window position, the first window segment of the first wavelength AC component signal and the second window segment of the second wavelength AC component signal are respectively extracted; The stability indexes of the first window segment and the second window segment are calculated respectively, wherein the stability indexes are obtained based on the reciprocal of the normalized root mean square error between each window segment and its corresponding smoothed window segment. The stability indices of the first window segment and the second window segment are compared with preset stability thresholds respectively; When the stability indices of both the first and second window segments are greater than the stability threshold, the first wavelength AC component signal segment, the second wavelength AC component signal segment corresponding to the current analysis window position, and the first wavelength DC component signal segment and the second wavelength DC component signal segment within the same time interval are extracted together as preliminary effective signal segments. Specifically, by using a preset fixed-length time window and a sliding step size smaller than the window length, the signal sequence is sequentially traversed along the time axis of the dual-wavelength AC component signal sequence to achieve continuous traversal of the signal sequence. For each analysis window position, signal segments whose time coordinates fall completely within the start and end range of the current analysis window are extracted from the continuous sequences of the first wavelength AC component signal and the continuous sequences of the second wavelength AC component signal, respectively, to obtain the first and second window segments that are precisely aligned in the time dimension. The calculation of the stability index is based on the principle that each window segment has a high similarity to its corresponding smoothed window segment. The specific calculation process is as follows: First, each window segment is subjected to first-order hysteresis smoothing to suppress high-frequency random fluctuations in the signal; then, the similarity between the window segment and the smoothed window segment is calculated. The root mean square error (RMSE) is obtained by taking the square root of the sum of squares of the differences between data points. This RMS error is then normalized by dividing the RMS error by the peak-to-peak value of the window segment, where the peak-to-peak value is the difference between the maximum and minimum signal values ​​within the window segment. Finally, the reciprocal of this normalized RMS error is used as the stability index of the window segment. A larger stability index indicates a lower degree of deviation from the smoothness trend of the window segment, meaning a more stable signal shape. The screening process is performed independently for each analysis window position. The stability indices of the first and second window segments calculated for the current analysis window are compared with preset stability thresholds. Only when the stability indices of both the first and second window segments are greater than the stability threshold are the dual-wavelength signal quality within the current analysis window determined to meet the preset stability requirements. Once the determination is successful, the first and second wavelength AC component signal segments corresponding to the analysis window position are extracted, and the first and second wavelength DC component signal segments within the same time interval are simultaneously extracted. These four signal segments are then encapsulated and identified as a preliminary valid signal segment.

[0024] The present invention is further configured such that S3 includes: Perform short-time Fourier transform on the first wavelength AC component signal segment and the second wavelength AC component signal segment contained in each preliminary effective signal segment to obtain the corresponding first time-frequency matrix and second time-frequency matrix; Within the preset physiological frequency band, the normalized cross-energy spectrum at each time frame and frequency point is calculated based on the first time-frequency matrix and the second time-frequency matrix. Based on the preset physiological feature weighting function, the normalized cross-energy spectrum within the preset physiological frequency band is weighted and summed to obtain the microcirculation physiological coupling coefficient of the preliminary effective signal segment. When the microcirculation physiological coupling coefficient is greater than or equal to a preset physiological coupling threshold, the preliminary effective signal segment is determined to have physiological rationality and is identified as the target signal segment. Specifically, for each preliminary effective signal segment, a short-time Fourier transform is performed on the first wavelength AC component signal segment and the second wavelength AC component signal segment contained therein. This transform is achieved by segmented windowing processing using a sliding window with a preset length and preset overlap. This includes: windowing the AC component signal segment using a preset window function and then sliding the analysis window along the time axis; performing a fast Fourier transform on each windowed AC component signal segment to analyze each AC component signal segment. The signal spectrum components within the segment include their amplitude and phase information. By traversing the entire AC component signal segment, the one-dimensional time-domain signal sequence is expanded into a two-dimensional time-frequency matrix. One dimension of the two-dimensional time-frequency matrix corresponds to the time frame defined by the sliding position of the analysis window, and the other dimension corresponds to the frequency points decomposed by the Fourier transform. Each element of the two-dimensional time-frequency matrix is ​​a complex number, whose magnitude represents the intensity or energy of the signal at that time frame and frequency point, and whose argument represents the phase of that frequency component. After this processing, the first wavelength AC component signal segment and the second wavelength AC component signal segment are respectively converted into a first time-frequency matrix and a second time-frequency matrix with the same structure, and the two time-frequency matrices together form a complete sequence. The distribution of dual-wavelength signal energy in a two-dimensional time and frequency plane was characterized. Within a preset physiological frequency band, the normalized cross-energy spectrum was calculated between all pairs of complex elements with the same time frame index and frequency point index in the first time-frequency matrix corresponding to the first wavelength AC component signal segment and the second time-frequency matrix corresponding to the second wavelength AC component signal segment. For each pair of complex elements, the calculation process was as follows: First, the modulus of the conjugate product of the first and second complex elements was calculated and multiplied by two. Then, the sum of the squares of the first and second complex element moduli was calculated and a preset minimum normal quantity was added to the sum to prevent division by zero errors. Finally, the obtained double modulus value is divided by the sum of squared moduli corrected by a minimal normal number. The result is the normalized cross-energy spectrum value at that specific time frame and frequency point. This normalized cross-energy spectrum value is used to quantify the consistency of the energy fluctuation mode of the dual-wavelength signal at this time frame and frequency point. A normalized cross-energy spectrum value approaching one indicates a higher coupling degree, while a value approaching zero indicates a lower correlation. Based on a preset physiological feature weighting function defined in the frequency dimension, the normalized cross-energy spectrum values ​​at all time frames and frequency points within the preset physiological frequency band are weighted and summed to obtain the microcirculation physiological coupling coefficient. Here, the frequency dimension refers to the coordinate axis representing the frequency component in the time-frequency matrix.The physiological feature weighting function assigns a corresponding weight to each frequency point within a preset physiological frequency band. This function exhibits a bell-shaped distribution within the preset physiological frequency band, peaking at the center frequency. Its function value smoothly decreases to zero from the peak towards both edges of the band, and remains zero at frequency points outside the preset physiological frequency band. The weighted summation process includes: traversing each time frame and frequency point within the preset physiological frequency band, multiplying the corresponding normalized cross-energy spectrum value by the physiological feature weighting function value corresponding to that frequency point, summing all the product results, and then applying this weighted summation. The summation and division by the sum of the weighted function values ​​of physiological features at all frequency points within the preset physiological frequency band yields the microcirculation physiological coupling coefficient. This coefficient reflects the overall coordination and consistency of the energy fluctuation patterns of the dual-wavelength signals within the preset physiological frequency band in the initially effective signal segment. For each initially effective signal segment, its corresponding microcirculation physiological coupling coefficient is compared with a preset physiological coupling threshold. When the microcirculation physiological coupling coefficient is greater than or equal to the threshold, the initially effective signal segment is deemed physiologically reasonable and marked as the target signal segment.

[0025] The present invention is further configured such that S4 includes: For each target signal segment, calculate the normalized cyclic cross-correlation function sequence of the first wavelength AC component signal segment and the second wavelength AC component signal segment contained therein within the preset time delay range; Extract the peak value with the largest absolute value from the normalized cyclic cross-correlation function sequence; Based on the statistical properties of the normalized cyclic cross-correlation function sequence, the significance threshold is determined; If the absolute value of the peak value is greater than the significance threshold, then the absolute value of the peak value is taken as the cross-correlation peak intensity of the target signal segment; If the absolute value of the peak is less than or equal to the significance threshold, the cross-correlation peak intensity of the target signal segment is set to zero. Specifically, for each target signal segment, extract the first wavelength AC component signal segment and the second wavelength AC component signal segment that are time-aligned and have the same number of data points. Set one of the signal segments as a fixed reference sequence and the other as a moving sequence. Within a preset time delay range determined based on the maximum physiological time delay, the time delay is measured in discrete sampling points. The specific range of this time delay is obtained by dividing the maximum physiological time delay by the sampling time interval of the signal and rounding it down, thereby generating a set of discrete time delay index values. This set of discrete time delay index values ​​defines the calculation of the normalization loop. The cross-correlation function needs to cover all time delay points. For each time delay index value within the preset time delay range, the moving sequence is cyclically shifted along the time axis by the number of data points corresponding to that time delay index value. The sum of the point-by-point products of the shifted moving sequence and the fixed reference sequence is calculated. This sum of products is divided by the square root of the product of the sum of the squares of the fixed reference sequence and the sum of the squares of the moving sequence, thus obtaining the normalized cross-correlation coefficient under that time delay index value. By traversing all time delay index values ​​within the preset time delay range and repeating the above normalized cross-correlation coefficient calculation process, a sequence of normalized cyclic cross-correlation functions with time delay index values ​​as independent variables and normalized cross-correlation coefficients as dependent variables is obtained. The absolute value of each point in the sequence represents the similarity of the waveform shape of the dual-wavelength AC signal under the corresponding time delay index value; traversing all the normalized cross-correlation coefficients corresponding to the time delay index values ​​in the normalized cyclic cross-correlation function sequence, the normalized cross-correlation coefficient with the largest absolute value is determined, and this normalized cross-correlation coefficient is defined as the peak value of the normalized cyclic cross-correlation function, and the time delay index value corresponding to this peak value is recorded; the absolute value of the peak value represents the maximum waveform similarity that the first wavelength AC component signal segment and the second wavelength AC component signal segment can achieve under the optimal time delay alignment state; to effectively distinguish between the random peaks caused by random noise and the significant correlation peaks generated by real physiological activities in the normalized cross-correlation function sequence under low signal-to-noise ratio conditions, it is necessary to... The extracted peak values ​​are subjected to a statistical significance test. This test constructs a dynamic significance threshold based on the statistical characteristics of the normalized cyclic cross-correlation function sequence itself. The specific method is as follows: calculate the arithmetic mean and standard deviation of the normalized cyclic cross-correlation function sequence after taking the absolute values ​​of all time delay index values, add a preset multiple of the standard deviation to the arithmetic mean, and define the result as the significance threshold. The statistical basis of this test method is that, under the null hypothesis that random noise dominates and there is no real correlation between signals, the probability that the absolute value of the normalized cross-correlation coefficient is greater than the significance threshold is lower than the preset significance level. If the absolute value of the observed peak value is greater than the significance threshold, the null hypothesis can be rejected at the preset confidence level, and thus the peak value is determined to be statistically significant.The extracted absolute peak value is compared with a significance threshold. If the absolute peak value is greater than the stability threshold, it is determined that there is a significant temporal correlation in the dual-wavelength AC signal within the current target signal segment, and this absolute peak value is used as the cross-correlation peak intensity to quantify the waveform similarity of the dual-wavelength AC signal under optimal time delay alignment. If the absolute peak value is less than or equal to the stability threshold, it is determined that no significant temporal correlation is detected within the current target signal segment, and the cross-correlation peak intensity of the target signal segment is set to zero.

[0026] The present invention is further configured such that S5 includes: For each target signal segment, extract the first wavelength DC component signal segment and the second wavelength DC component signal segment contained therein; Calculate the average values ​​of the first wavelength DC component signal segment and the second wavelength DC component signal segment after removing the preset beginning and end transition regions; The ratio of the average value of the first wavelength DC component signal segment to the average value of the second wavelength DC component signal segment is calculated, and the ratio is mapped to the corresponding steady-state volume index according to a preset mapping relationship. Specifically, for each target signal segment that passes the physiological rationality judgment, the first wavelength DC component signal segment and the second wavelength DC component signal segment contained therein are extracted. The DC component signal segment represents the overall non-pulsating absorption or reflection intensity of the first wavelength light and the second wavelength light by biological tissue within the corresponding time window. This intensity information comprehensively reflects the comprehensive attenuation effect of various tissue components such as skin, subcutaneous tissue, and non-pulsating blood on incident light. To enhance the robustness of the calculation, the DC component signal needs to be evaluated before calculating the average value. The signal segment undergoes preprocessing, specifically by removing data points from the beginning and end of the segment in a predetermined transition zone to eliminate edge effects, thus retaining the middle stable segment for characterizing steady-state properties. After removing the transition zone, the average values ​​of the first and second wavelength DC component signal segments of the remaining stable portion are calculated. These two average values ​​quantify the average reflected light intensity of the tissue to the first and second wavelengths of light within the stable observation time corresponding to the target signal segment. This light intensity, after logarithmic transformation, corresponds to the average absorption degree of light by the tissue. The average value of the first wavelength DC component signal segment is divided by the average value of the second wavelength DC component signal segment to calculate the ratio of the two wavelength DC light intensities. The DC intensity ratio eliminates the potential fluctuations in the absolute intensity of the light source, directly reflecting the relative intensity ratio of the two wavelengths of light after absorption by biological tissue. This ratio is closely related to the overall optical properties of blood in the tissue. By mapping the dual-wavelength DC intensity ratio, it is converted into a steady-state volume index characterizing the steady-state optical properties of blood. This mapping is based on a pre-established and calibrated mathematical model, typically in the form of a polynomial function. The specific coefficients in the model are determined by fitting in vitro simulation experiments with clinical data from known life states. The in vitro simulation experiments establish a fundamental mapping between optical signals and volume parameters by setting different simulated blood volume conditions on tissue phantoms with known optical properties. The clinical data fitting process involves collecting samples of the dual-wavelength DC light intensity ratio from individuals with clearly defined weak vital signs and those without. These sample data are then numerically fitted to their corresponding life status labels to ultimately determine the model coefficients. Through this calibration process, the model can map the input dual-wavelength DC light intensity ratio to a calibrated steady-state volume index range. This ensures that the steady-state volume index calculated under living conditions is primarily concentrated within a preset range, while under non-living conditions, the steady-state volume index exhibits a distribution significantly different from this range. This steady-state volume index comprehensively reflects the absorption characteristics of dual-wavelength light by the blood volume and its components in tissues under steady-state observation conditions.

[0027] The present invention is further configured such that S6 includes: The cross-correlation peak intensity corresponding to each target signal segment is compared with a preset intensity threshold. The steady-state volume index corresponding to each target signal segment is compared with the preset lower and upper thresholds. When the cross-correlation peak intensity is greater than the intensity threshold and the steady-state volume index is within the numerical range defined by the lower and upper thresholds, it is determined that there are transient weak vital signs at the corresponding moment of the target signal segment. The invention is further configured to count the cumulative number of times transient weak vital signs are determined to exist within a sliding time window of a preset time length. If the cumulative number is greater than a preset number threshold and the determination result at the current moment is that transient weak vital signs exist, then it is determined that the weak vital signs continue to exist, and an alarm signal is triggered. If the cumulative number is less than or equal to the number threshold, then the triggered alarm signal is not triggered or is revoked. Specifically, based on preset two-dimensional determination conditions, an instantaneous single-event determination is performed on each target signal segment. The determination requires that the following two independent conditions be met simultaneously: the cross-correlation peak intensity corresponding to the target signal segment is greater than the preset intensity threshold, and its steady-state volume index is within the numerical range defined by the preset lower and upper thresholds. The target signal segment is determined to exist if and only if both of the above independent conditions are met simultaneously. If a momentary weak vital signs are present at the corresponding time, a positive judgment flag is generated; if any condition is not met, a negative judgment flag is generated. To overcome the problem that the reliability of momentary judgment conclusions is easily affected by random noise or sudden artifacts, improve the robustness and credibility of decision-making, and effectively prevent false alarms caused by occasional interference events, a continuous verification mechanism based on a time window is introduced. This verification mechanism maintains a sliding time window covering a recent preset historical period, periodically checks all momentary judgment events occurring within this sliding time window, and counts the cumulative number of momentary positive judgment flags within the window as a criterion for assessing whether vital signs are continuously present. If the cumulative number of momentary positive judgment flags within the sliding time window is greater than a preset threshold, and the momentary judgment result at the current time is positive, then the weak vital signs are determined to be continuously present, and an alarm trigger command is generated to activate or maintain the alarm state of the alarm device; otherwise, it is determined that the condition for the continuous presence of weak vital signs is not met, and the alarm state is canceled or maintained.

[0028] The present invention is further configured such that the triggering alarm signal is a triggering acoustic and optical alarm device; specifically, when it is determined that weak vital signs continue to exist, a high-level effective alarm signal is triggered and generated; the alarm signal is transmitted to the alarm driving circuit of the vital signs detection device, and after receiving the alarm signal, the alarm driving circuit synchronously drives the acoustic alarm unit and the optical alarm unit of the device to perform preset warning actions: driving the buzzer (9) to emit a warning sound with a preset audio mode, and driving the light-emitting diode (10) to flash according to a preset frequency and duty cycle, thereby realizing the synchronous triggering of acoustic alarm and optical alarm.

[0029] The present invention is further configured such that the method includes generating visual indication information, which is used to characterize the determination result of weak vital signs; specifically, based on the determination result of weak vital signs, the visual display unit of the driving device generates corresponding visual indication information; when it is determined that weak vital signs continue to exist, the visual display unit is controlled to present a first visual state, which is to drive the indicator light to illuminate with a first preset color; when it is determined that the condition of weak vital signs continuing to exist is not met, the visual display unit is driven to present a second visual state, which is to drive the indicator light to illuminate with a second preset color.

[0030] In another embodiment of the invention, such as Figure 2 , Figure 3 and Figure 4 As shown, a vital sign detection device is provided that is suitable for the aforementioned method of judging weak vital signs based on infrared photodiode illumination technology; as... Figure 2 As shown, the main body of the device is a ring (1) suitable for wearing on a finger, with a silicone pad (3) covering its outer side to protect the internal electronic components and enhance the structural durability of the device; the device integrates a PCB board (2), such as Figure 3 As shown, the PCB board (2) is arranged with core functional components for implementing the method of the present invention, including: a photodetector (6) for receiving mixed light signals reflected back from the biological tissue of the finger, a buzzer (9) for emitting an acoustic alarm, a light-emitting diode (10) for emitting an optical alarm, a switch (5) for controlling the on / off state of the circuit, and a button (4) for user operation; wherein, the switch (5) is soldered and fixed on the PCB board (2), and its trigger end is set upwards, and the button (4) is installed on the silicone pad (3) of the device shell, and its position is precisely aligned with the trigger end of the switch (5), so that the mechanical force generated by pressing the button (4) from the outside can directly trigger the internal switch (5) to act; Figure 4As shown, the core functional components also include a first light-emitting device (7) for emitting light of a first wavelength and a second light-emitting device (8) for emitting light of a second wavelength. The first light-emitting device (7) and the second light-emitting device (8) are arranged side by side on the side of the photodetector (6). The first light-emitting device (7), the second light-emitting device (8) and the photodetector (6) together constitute a reflective optical detection module pointing to the biological tissue area of ​​the finger.

[0031] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for identifying weak vital signs based on infrared photodiode illumination technology, characterized in that, include: S1: Synchronously acquire the dual-wavelength mixed light signal reflected by biological tissue and decompose it into DC component signal and AC component signal corresponding to each wavelength. S2: Based on the time-domain stability of dual-wavelength AC component signals, preliminary effective signal segments are selected from continuous signals; S3: Calculate the normalized cross-energy spectrum of the dual-wavelength AC component signal segment in the preliminary effective signal segment within the preset physiological frequency band, and perform weighted summation to obtain the microcirculation physiological coupling coefficient. Based on the microcirculation physiological coupling coefficient, identify the target signal segment with physiological rationality. S4: Calculate the normalized cross-correlation function of the dual-wavelength AC component signal segment in the target signal segment, and extract the peak value of the normalized cross-correlation function as the cross-correlation peak intensity. S5: Calculate the steady-state volume index based on the ratio of the mean values ​​of the dual-wavelength DC component signal segments in the target signal segment using a preset mapping relationship; S5 includes: for each target signal segment, extracting the first-wavelength DC component signal segment and the second-wavelength DC component signal segment contained therein; calculating the average value of the first-wavelength DC component signal segment and the second-wavelength DC component signal segment after removing the preset beginning and end transition regions; calculating the ratio of the average value of the first-wavelength DC component signal segment to the average value of the second-wavelength DC component signal segment, and mapping the ratio to the corresponding steady-state volume index according to the preset mapping relationship; S6: Compare the cross-correlation peak intensity and steady-state volume index with the corresponding preset thresholds respectively. When the comparison results of the two meet the preset conditions at the same time, it is determined that there are weak vital signs.

2. The method for judging weak vital signs based on infrared photodiode illumination technology according to claim 1, characterized in that, S1 includes: A first light-emitting device that emits a preset first wavelength and a second light-emitting device that emits a preset second wavelength are synchronously driven at a preset sampling frequency. A single photodetector synchronously receives dual-wavelength mixed optical signals after being processed by biological tissue and converts them into mixed analog electrical signals. By performing cross-impedance amplification, gain adjustment, and analog-to-digital conversion on the hybrid analog electrical signal, a hybrid discrete digital sequence is obtained. Based on the driving timing of the first light-emitting device and the second light-emitting device, a first discrete digital sequence corresponding to the first wavelength and a second discrete digital sequence corresponding to the second wavelength are separated from the mixed discrete digital sequence; A low-pass digital filter with a preset cutoff frequency is used to filter the first discrete digital sequence and the second discrete digital sequence respectively to obtain the first wavelength DC component signal and the second wavelength DC component signal. Subtracting the first wavelength DC component signal from the first discrete digital sequence yields the first wavelength AC component signal, and subtracting the second wavelength DC component signal from the second discrete digital sequence yields the second wavelength AC component signal.

3. The method for judging weak vital signs based on infrared photodiode illumination technology according to claim 1, characterized in that, S2 includes: An analysis window of a preset duration is slid across a continuous sequence of the first wavelength AC component signal and the second wavelength AC component signal with a preset step size. For each analysis window position, the first window segment of the first wavelength AC component signal and the second window segment of the second wavelength AC component signal are respectively extracted; The stability indexes of the first window segment and the second window segment are calculated respectively, wherein the stability indexes are obtained based on the reciprocal of the normalized root mean square error between each window segment and its corresponding smoothed window segment. The stability indices of the first window segment and the second window segment are compared with preset stability thresholds respectively; When the stability indices of both the first and second window segments are greater than the stability threshold, the first wavelength AC component signal segment, the second wavelength AC component signal segment corresponding to the current analysis window position, and the first wavelength DC component signal segment and the second wavelength DC component signal segment within the same time interval are extracted together as preliminary effective signal segments.

4. The method for judging weak vital signs based on infrared photodiode illumination technology according to claim 1, characterized in that, S3 includes: Perform short-time Fourier transform on the first wavelength AC component signal segment and the second wavelength AC component signal segment contained in each preliminary effective signal segment to obtain the corresponding first time-frequency matrix and second time-frequency matrix; Within the preset physiological frequency band, the normalized cross-energy spectrum at each time frame and frequency point is calculated based on the first time-frequency matrix and the second time-frequency matrix. Based on the preset physiological feature weighting function, the normalized cross-energy spectrum within the preset physiological frequency band is weighted and summed to obtain the microcirculation physiological coupling coefficient of the preliminary effective signal segment. When the microcirculation physiological coupling coefficient is greater than or equal to the preset physiological coupling threshold, the preliminary effective signal segment is determined to be physiologically reasonable and is identified as the target signal segment.

5. The method for judging weak vital signs based on infrared photodiode illumination technology according to claim 1, characterized in that, S4 includes: For each target signal segment, calculate the normalized cyclic cross-correlation function sequence of the first wavelength AC component signal segment and the second wavelength AC component signal segment contained therein within the preset time delay range; Extract the peak value with the largest absolute value from the normalized cyclic cross-correlation function sequence; Based on the statistical properties of the normalized cyclic cross-correlation function sequence, the significance threshold is determined; If the absolute value of the peak value is greater than the significance threshold, then the absolute value of the peak value is taken as the cross-correlation peak intensity of the target signal segment; If the absolute value of the peak is less than or equal to the significance threshold, the cross-correlation peak intensity of the target signal segment is set to zero.

6. The method for judging weak vital signs based on infrared photodiode illumination technology according to claim 1, characterized in that, S6 includes: The cross-correlation peak intensity corresponding to each target signal segment is compared with a preset intensity threshold. The steady-state volume index corresponding to each target signal segment is compared with the preset lower and upper thresholds. When the peak intensity of the cross-correlation is greater than the intensity threshold and the steady-state volume index is within the numerical range defined by the lower threshold and the upper threshold, it is determined that there are transient weak vital signs at the corresponding moment of the target signal segment.

7. The method for judging weak vital signs based on infrared photodiode illumination technology according to claim 6, characterized in that, The cumulative number of times that were determined to have momentary weak vital signs within a sliding time window of a preset time length is counted. If the cumulative number of times exceeds the preset threshold and the current judgment result is that there are momentary weak vital signs, then it is determined that the weak vital signs continue to exist and an alarm signal is triggered. If the cumulative number of alarms is less than or equal to the threshold number, the alarm signal that has been triggered will not be triggered or will be revoked.

8. The method for judging weak vital signs based on infrared photodiode illumination technology according to claim 7, characterized in that, The alarm signal is used to trigger acoustic and optical alarm devices.

9. The method for judging weak vital signs based on infrared photodiode illumination technology according to claim 1, characterized in that, The method further includes generating visual indication information, which is used to characterize the determination result of weak vital signs.

Images