Skin Color Adaptive Wavelength Adjustment Method and Device for Finger Pulse Oximeter

Abstract

This invention discloses a skin-tone adaptive wavelength adjustment method and device for a fingertip pulse oximeter, relating to the field of pulse oximeter technology. The method includes: when a user's finger is inserted into the detection cavity of the fingertip pulse oximeter, an image of the user's finger is obtained through an image acquisition device built into the detection cavity; skin tone analysis is performed based on the user's finger image to obtain skin tone feature values; median wavelength optimization of emitted light is performed based on the user's finger skin tone feature values ​​to obtain target wavelengths in the red light band and infrared band; and the infrared emitter is initialized according to the target wavelengths in the red light band and infrared band. This invention solves the technical problem that traditional fingertip pulse oximeters cannot adjust the detection wavelength in real time according to the user's skin tone, resulting in blood oxygen detection accuracy not matching skin tone. It achieves the technical effect of improving blood oxygen content detection accuracy and adapting to the needs of users with different skin tones by adaptively adjusting the target wavelengths in the red light band and infrared band according to the user's finger skin tone feature values.

Description

Technical Field

[0001] This application relates to the field of pulse oximeter technology, specifically to a method and device for skin color adaptive wavelength adjustment in a fingertip pulse oximeter. Background Technology

[0002] A fingertip pulse oximeter is a widely used device in medical and health monitoring fields. It assesses blood oxygen saturation by measuring the oxygen content in the blood at the fingertip. Traditional fingertip pulse oximeters typically use fixed red light wavelengths (approximately 650nm) and infrared wavelengths (approximately 850nm) for detection. While these fixed wavelengths meet the needs of most users, they have significant technical limitations in practical use. First, skin color varies significantly among users, directly affecting the penetration depth and scattering of light in the skin. For example, darker skin tones, due to higher melanin content, have a greater impact on light absorption and scattering, thus reducing the accuracy of the detection signal. This skin color difference leads to inaccuracies in the oximeter's results across different populations. Second, traditional fingertip pulse oximeters typically cannot dynamically adjust the wavelength to adapt to individual differences; their preset wavelength designs cannot cover the optimal optical response under various skin color conditions. Therefore, in some special scenarios (such as users with dark skin or abnormal blood circulation), detection accuracy and reliability may significantly decrease, affecting their widespread application in clinical and home settings. Furthermore, due to the use of fixed wavelengths, traditional pulse oximeters may struggle to guarantee the stability and accuracy of test results in high-noise environments, especially among users with darker skin tones or complex finger shapes, where this problem is more pronounced. Summary of the Invention

[0003] This application provides a skin-tone adaptive wavelength adjustment method and device for fingertip pulse oximeters, which solves the technical problem that traditional fingertip pulse oximeters cannot adjust the detection wavelength in real time according to the user's skin tone, resulting in blood oxygen detection accuracy not matching skin tone. It achieves the technical effect of improving the accuracy of blood oxygen content detection and adapting to the needs of users with different skin tones by adaptively adjusting the target wavelengths of the red light band and infrared band according to the characteristic values ​​of the user's finger skin tone.

[0004] This application provides a method for skin color adaptive wavelength adjustment of a fingertip pulse oximeter. The method includes: when a user's finger is inserted into the detection cavity of the fingertip pulse oximeter, obtaining an image of the user's finger through an image acquisition device built into the detection cavity; performing skin color analysis based on the user's finger image to obtain skin color feature values ​​of the user's finger; optimizing the median wavelength of emitted light based on the user's finger skin color feature values ​​to obtain a target wavelength in the red light band and an infrared band; and initializing the infrared emitter of the fingertip pulse oximeter according to the target wavelength in the red light band and the target wavelength in the infrared band.

[0005] This application also provides a skin color adaptive wavelength adjustment device for a fingertip pulse oximeter, comprising: an image acquisition module, which acquires an image of the user's finger by means of an image acquisition device built into the detection cavity of the fingertip pulse oximeter when the user's finger is inserted into the detection cavity; a skin color analysis module, which performs skin color analysis based on the user's finger image to obtain skin color feature values ​​of the user's finger; a wavelength median optimization module, which optimizes the median wavelength of emitted light based on the skin color feature values ​​of the user's finger to obtain a target wavelength in the red light band and a target wavelength in the infrared band; and an infrared emitter initialization module, which initializes the infrared emitter of the fingertip pulse oximeter according to the target wavelength in the red light band and the target wavelength in the infrared band.

[0006] The proposed method and device for skin-tone adaptive wavelength adjustment in fingertip pulse oximeters involves the following steps: When a user's finger is inserted into the detection chamber of the fingertip pulse oximeter, an image of the user's finger is acquired using an image acquisition device built into the detection chamber; skin tone analysis is performed on the finger image to obtain skin tone feature values; median wavelength optimization of emitted light is performed based on the skin tone feature values ​​to obtain target wavelengths in the red and infrared bands; and the infrared emitter of the fingertip pulse oximeter is initialized based on the target wavelengths in the red and infrared bands. This solves the technical problem of traditional fingertip pulse oximeters failing to adjust the detection wavelength in real time according to the user's skin tone, resulting in inaccurate blood oxygenation detection. It achieves the technical effect of improving blood oxygenation detection accuracy and adapting to the needs of users with different skin tones by adaptively adjusting the target wavelengths in the red and infrared bands based on the user's finger skin tone feature values. Attached Figure Description

[0007] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the apparatus according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0008] Figure 1 A schematic flowchart of the skin color adaptive wavelength adjustment method for a fingertip pulse oximeter provided in this application embodiment.

[0009] Figure 2 A schematic diagram of the skin color adaptive wavelength adjustment device for a fingertip pulse oximeter provided in this application embodiment.

[0010] Figure labeling: Image acquisition module 1, skin color analysis module 2, wavelength median optimization module 3, infrared emitter initialization module 4. Detailed Implementation

[0011] 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.

[0012] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0013] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or apparatuses. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0014] This application provides a method for skin color adaptive wavelength adjustment using a fingertip pulse oximeter, such as... Figure 1 As shown, the method includes: When a user inserts their finger into the detection chamber of a fingertip pulse oximeter, an image of the user's finger is obtained through an image acquisition device built into the detection chamber.

[0015] In this embodiment, when a user inserts their finger into the detection chamber of the fingertip pulse oximeter, the device's built-in image acquisition unit immediately activates to capture an image of the user's finger. The image acquisition unit is typically a high-resolution miniature camera or image sensor, capable of clearly recording the external features of the finger, including its color, shape, and size. This image data is transmitted in real-time to the device's processing unit for subsequent analysis, such as extracting skin color feature values, thereby providing fundamental data support for subsequent detection wavelength optimization. This process is usually completed imperceptibly by the user, making it both fast and accurate.

[0016] Skin color analysis is performed on the user's finger image to obtain the user's finger skin color feature value.

[0017] In one embodiment, the system terminal analyzes the captured image of a user's finger, focusing on identifying the skin color features of the finger. Specifically, the system terminal first extracts the color information from the image and processes it through color aggregation, pixel statistics, and other methods to remove background interference and focus on the finger area, calculating a feature value representing the user's finger skin color. This skin color feature value is typically a quantified numerical value that reflects the main skin color attributes of the user's finger, providing accurate data support for subsequent wavelength optimization. The entire analysis process is both efficient and accurate, ensuring reliable detection results for users with different skin colors.

[0018] Furthermore, this application provides a method for performing skin color analysis based on the user's finger image to obtain skin color feature values ​​of the user's fingers, including: Color neighborhood aggregation is performed on the user's finger image to obtain a first region, a second region, and so on up to the Nth region; finger shape matching is performed by traversing the first region, the second region, and so on up to the Nth region to obtain a finger shape calibration region, wherein the finger shape calibration region belongs to the first region, the second region, and so on up to the Nth region; the mode color feature value of the finger shape calibration region is extracted and set as the skin color feature value of the user's finger.

[0019] Preferably, after acquiring the user's finger image, the system terminal converts the image from the RGB color space to the Lab color space, as the Lab color space better represents color differences perceived by the human eye, which helps improve the clustering effect. Subsequently, each pixel in the image is converted into a color vector, for example, using the three channel values ​​(L, a, b) of the Lab color space as feature vectors. In the color clustering stage, the K-means clustering algorithm is used to group the color vectors. For this, the number of clusters K is set, which can be determined empirically or dynamically using methods such as the elbow method to determine a suitable K value. K cluster centers are then randomly initialized, each center representing a primary color. For each pixel, its Euclidean distance to all cluster centers is calculated, and it is assigned to the cluster of the nearest cluster center. After the initial assignment, the mean value of pixels within each cluster is recalculated, and this is used as the new cluster center. This process is repeated until the change in the number of cluster centers is less than a set convergence threshold, or the set number of iterations is reached. After clustering, color neighborhood regions are formed based on the pixels of each category, and each neighborhood region is assigned a unique number, resulting in the first region, the second region, and so on up to the Nth region. Simultaneously, the generated neighborhood regions are optimized by removing regions that are too small or isolated (such as regions with an area below a set threshold), thereby reducing interference from invalid regions. Then, all color neighborhood regions are traversed, and finger shape matching analysis is performed one by one to identify the region where the user's finger is located. In this process, the system terminal constructs a finger shape mold based on user input and preset parameters, and performs shape matching between this mold and each color neighborhood region, calculating their similarity to obtain a similarity value for each region. Then, finger regions are filtered based on the similarity value threshold to obtain finger shape calibration regions. This finger shape calibration region can be any one of the first, second, or Nth regions. Within the determined finger shape calibration region, the system terminal performs statistical analysis on all pixel color values ​​of that region, extracting the main color feature of the region by calculating the mode of the color values ​​(i.e., the color value that appears most frequently), ensuring that this feature value accurately represents the skin color of the user's finger. The system terminal uses this modal color feature value as the final skin tone feature value of the user's finger, providing basic data for subsequent wavelength optimization. Through the above steps, the skin tone feature value of the user's finger can be accurately extracted, ensuring that wavelength adjustment adapts to individual skin tone differences and improving the reliability and accuracy of detection.

[0020] Furthermore, this application provides a method for traversing the first region, the second region, and up to the Nth region to perform finger shape matching to obtain a finger shape marking region, including: The system receives the user's finger width input from the user terminal and obtains the finger length that the detection cavity can accommodate. Based on the user's finger width and the finger length that can accommodate, a finger mold is constructed. The system traverses the first region, the second region, and up to the Nth region, and performs similarity analysis with the finger mold to obtain a first similarity value, a second similarity value, and up to the Nth similarity value. The system extracts regions from the first similarity value, the second similarity value, and up to the Nth similarity value that are greater than or equal to the similarity value threshold, and sets them as candidate calibration regions. When the number of candidate calibration regions is 1, the candidate calibration region is set as the finger calibration region. When the number of candidate calibration regions is not 1, the system updates the user's finger image and performs a skin color analysis loop.

[0021] Optionally, when a user's finger is inserted into the detection cavity, the system terminal receives the user's finger width input from the user terminal, typically through measurement or directly provided by the user. Simultaneously, it automatically obtains the finger-accommodating length of the detection cavity based on internal parameters or sensors. These two data points serve as the basis for constructing the finger mold. Using the user's finger width and the finger-accommodating length of the detection cavity, a finger mold is constructed using 3D modeling tools. Subsequently, the system begins traversing the color neighborhood regions (first region, second region, up to the Nth region) in the user's finger image. For each color neighborhood region, the system terminal compares it one by one with the constructed finger mold, performing a similarity analysis. Specifically, the system terminal first extracts the boundary of each color neighborhood region, obtaining a set of boundary points that fully describe the region's external contour. Simultaneously, the boundary point set of the finger mold has been generated using 3D modeling tools, simulating the natural contour of the finger. For each color neighborhood region, its boundary point set is compared with the boundary point set of the finger mold for shape similarity analysis. Here, Hausdorff distance is used to quantify the shape difference between the region boundary and the mold boundary. Frechet distance, Euclidean distance, etc., can also be used. Hausdorff distance is a commonly used distance metric used to calculate the nearest distance between the farthest matching point between two point sets, effectively describing the maximum deviation between the region and the mold. By calculating the Hausdorff distance between each color neighborhood region and the finger mold, and then subtracting the ratio of this distance to the maximum distance from 1, a similarity value is obtained as the similarity value of the region. This similarity value is between 0 and 1, where 1 indicates that the shapes are completely identical, and the closer the value is to 1, the closer the shape of the region is to the shape of the finger mold. Then, the system terminal traverses all color neighborhood regions, calculates the similarity value one by one, and records the result of each region, thus obtaining the first similarity value, the second similarity value, and so on up to the Nth similarity value. These similarity values ​​are compared with a preset similarity value threshold, and regions that are greater than or equal to the similarity value threshold are selected and marked as candidate calibration regions. If the number of candidate calibration regions is 1, then the region is directly set as the finger calibration region. If the number of candidate calibration regions is greater than one, it indicates that there are multiple possible matching regions. At this point, the system terminal will enter a further analysis phase, updating the user's finger image (e.g., by enhancing contrast or refining edge features) and re-executing the skin color analysis loop. This loop will be repeated until a unique finger shape calibration region is selected, ensuring the accuracy of the target region. Through this process, the specific region where the user's finger is located can be efficiently identified, providing precise region localization for subsequent skin color analysis and wavelength optimization.

[0022] Based on the user's finger skin color feature value, the median wavelength of emitted light is optimized to obtain the target wavelength in the red light band and the target wavelength in the infrared band.

[0023] In one embodiment, the red light band is typically defined in the 600-750nm range, with a conventional preset value of 650nm, which is the band where deoxyhemoglobin absorbs light most efficiently. The infrared band is typically in the 850-1000nm range, with a conventional preset value of 850nm, which is the band where oxyhemoglobin absorbs light most efficiently. Although these preset values ​​provide good detection results under standard conditions, differences in user skin color significantly affect the penetration depth and scattering state of light in skin tissue, thus impacting detection accuracy. To address this issue, the system terminal performs median optimization of the light wavelengths in the aforementioned bands based on the user's finger skin color characteristics. In this process, the red light wavelength restriction range and the infrared wavelength restriction range are reconstructed using the median value based on the user's finger skin color characteristics, resulting in reconstructed red light wavelength restriction ranges and infrared wavelength restriction ranges. For example, for the red light band (600-750nm), if the skin color characteristics indicate that the optimal penetration range should be biased towards a higher wavelength, the restriction range might be reconstructed to 660-700nm. After obtaining the red light wavelength reconstruction limit range and the infrared wavelength reconstruction limit range, the system terminal will find the optimal median wavelength from these ranges as the target wavelengths for the red light band and the infrared band. This method allows for dynamic adjustment of the red and infrared emission wavelengths to better suit the user's skin tone characteristics, thereby maximizing light absorption efficiency and improving the accuracy and reliability of blood oxygen detection.

[0024] Furthermore, this application provides a method for optimizing the median wavelength of emitted light based on the user's finger skin color feature values ​​to obtain target wavelengths in the red light band and infrared band, including: Obtain the red light wavelength restriction range and the infrared wavelength restriction range; based on the user's finger skin color feature value, perform median reconstruction on the red light wavelength restriction range and the infrared wavelength restriction range to obtain the red light wavelength reconstruction restriction range and the infrared wavelength reconstruction restriction range; extract the target wavelength of the red light band with the smallest deviation from 650nm from the red light wavelength reconstruction restriction range; extract the target wavelength of the infrared band with the smallest deviation from 850nm from the infrared wavelength reconstruction restriction range.

[0025] Preferably, the system terminal defines an initial wavelength range based on standard light absorption characteristics. The red light wavelength restriction range can be 600-750nm, covering the region with the highest absorbance of deoxyhemoglobin. The infrared wavelength restriction range can be 850-1000nm, covering the region with the highest absorbance of oxyhemoglobin. Subsequently, based on the skin color characteristics of the user's finger, the system terminal performs median reconstruction of the restriction ranges for red and infrared wavelengths based on the penetration ratio calculated by the ratio of the first penetration depth to the user's finger thickness. This is because dark skin may reduce the penetration depth of light, while light skin reflects more light; therefore, the restriction ranges need to be reconstructed to improve penetration efficiency. After median reconstruction, new red light wavelength reconstruction restriction ranges and infrared wavelength reconstruction restriction ranges are obtained respectively. Within the red light wavelength reconstruction restriction range, the system terminal searches for the wavelength with the smallest deviation from the standard preset value of 650nm using an absolute difference method as the target wavelength for the red light band. Similarly, within the infrared wavelength reconstruction restriction range, it searches for the wavelength with the smallest deviation from the standard preset value of 850nm as the target wavelength for the infrared band. After completing the above steps, the system terminal outputs the determined target wavelengths in the red and infrared bands as the final outputs to initialize the device's light emission module. Through dynamic optimization, the emitted light wavelengths are more closely matched to the user's skin tone characteristics, thereby improving the accuracy and reliability of the detection.

[0026] Furthermore, this application provides a method for reconstructing the red light wavelength restriction range and the infrared wavelength restriction range based on the user's finger skin color feature value, to obtain the red light wavelength reconstruction restriction range and the infrared wavelength reconstruction restriction range, including: Extract the lower limit wavelength of the red light wavelength restriction range and the upper limit wavelength of the infrared wavelength restriction range to construct a first median optimization range; take the median of the first median optimization range to obtain a first median wavelength; predict the penetration depth based on the user's finger skin color feature value and the first median wavelength to obtain a first penetration depth; receive the user's finger thickness at the user end; calculate the ratio of the first penetration depth to the user's finger thickness and set it as the penetration ratio; when the penetration ratio belongs to the penetration ratio threshold range, set the first median wavelength as the lower limit wavelength threshold, and reconstruct the red light wavelength restriction range and the infrared wavelength restriction range using the lower limit wavelength threshold as the lower limit wavelength to obtain the red light wavelength reconstruction restriction range and the infrared wavelength reconstruction restriction range.

[0027] Optionally, the system terminal first extracts the lower limit wavelength (e.g., 600nm) from the red light wavelength restriction range and the upper limit wavelength (e.g., 1000nm) from the infrared wavelength restriction range. These two values ​​will serve as the basis for subsequent optimization calculations. Then, using the extracted lower limit wavelength (e.g., 600nm) and upper limit wavelength (e.g., 1000nm) as boundaries, a unified range containing information from both wavelength bands is formed, namely the first median optimization range. Subsequently, the median is calculated by taking the median of the first median optimization range. For example, for the range [600nm, 1000nm], the first median wavelength is (600+1000) / 2=800nm. Then, combining the user's finger skin color feature value and the first median wavelength, a pre-trained penetration depth prediction model is used for prediction. This model is trained based on experimental data. The model processes the received user's finger skin color feature value and the first median wavelength, outputting the first penetration depth, representing the depth to which light penetrates the user's finger at the first median wavelength. Next, the system receives finger thickness data input from the user, typically obtained directly through a measuring device or provided by the user. For example, the finger thickness might be 8mm. The system terminal calculates the ratio of the first penetration depth to the user's finger thickness and uses this as the transmittance ratio. This ratio reflects whether light can effectively penetrate the user's finger at the current wavelength. For example, if the first penetration depth is 6mm and the user's finger thickness is 8mm, the transmittance ratio is 6 / 8 = 0.75. After calculating the transmittance ratio, the system terminal checks whether it falls within a preset transmittance ratio threshold range (e.g., 0.7-0.9). If the transmittance ratio is not within the threshold range, the system terminal adjusts the limit range and recalculates, which may include narrowing the range or adjusting the median. If the transmittance ratio is within the threshold range, the first median wavelength is used as the lower limit wavelength threshold of the red light wavelength limit range. Simultaneously, a similar adjustment is made to the infrared wavelength limit range to improve penetration efficiency. Through these steps, the optimized red light wavelength reconstruction limit range and infrared wavelength reconstruction limit range are finally obtained. These new restriction ranges are better suited to the skin color characteristics and thickness of the user's fingers, providing support for the subsequent extraction of wavelength targets.

[0028] Furthermore, this application provides a method for predicting penetration depth based on the user's finger skin color feature value and the first median wavelength to obtain a first penetration depth, including: Collect finger red light penetration experiment data, which includes a skin color feature vector variable dataset, a penetrating red light wavelength variable dataset, and a penetration depth calibration dataset. Use the penetration depth calibration dataset as output supervision, and use the skin color feature vector variable dataset and the penetrating red light wavelength variable dataset as input to perform random forest configuration to obtain a penetration depth prediction model. Based on the penetration depth prediction model, process the user's finger skin color feature value and the first median wavelength to obtain the first penetration depth.

[0029] Optionally, the system terminal establishes a penetration depth prediction model by collecting experimental data on red light penetration from fingers. This experimental data includes a skin color feature vector variable dataset, a red light wavelength variable dataset, and a penetration depth calibration dataset. The experimental data is collected using experimental equipment to ensure coverage of various skin colors and wavelengths. Data reliability is guaranteed by measuring reflectance, absorbance, and penetration depth calibration values ​​using a spectral analyzer and a photodetector. After collection, the data is cleaned and formatted, outliers are removed, missing data is filled in, and normalization is performed. Skin color feature vectors and red light wavelengths are used as input variables, and penetration depth calibration values ​​are used as output targets. Subsequently, the dataset is divided into training, validation, and test sets to ensure that each subset covers different skin color features and wavelength ranges. During model training, random forest is selected as the core algorithm for the penetration depth prediction model; deep neural networks can also be used. Before training, model parameters are set, including the number of decision trees, the maximum depth of each tree, and the random sampling ratio. At the start of training, multiple subsets are generated by randomly sampling with replacement from the training set; each subset is used to train one decision tree. For each decision tree, the feature space is progressively split based on the skin color feature vector and the red light wavelength variable, selecting the feature and split point with the highest node purity after splitting. Node purity is measured using mean squared error (MSE) to ensure optimal fitting of the predicted continuous variables. This splitting process continues until the maximum depth is reached or the number of node samples is less than a preset threshold. Afterward, the predictions from all decision trees are fused using the average to form the final random forest model output. During training, the model's hyperparameters are optimized using a validation set, including adjusting the number of decision trees, the sample and feature sampling ratio, and the maximum tree depth. Optimization can use grid search or random search methods to select the optimal parameter combination, while calculating mean squared error (MSE) or root mean squared error (RMSE) to validate model performance. After model validation, the final model's prediction performance is evaluated using a test set, using mean absolute error (MAE), mean squared error (MSE), and coefficient of determination (CR). As a performance metric, the model's generalization ability under different skin tones and wavelengths is ensured. Finally, based on the constructed random forest model, predictions are made using actual user finger data. Taking the user's finger skin tone features and the first median wavelength as input, the model processes the input data using the learned relationship between skin tone and wavelength on penetration depth, outputting the first penetration depth. This result reflects the effective penetration depth of light under the current skin tone features and wavelength conditions, providing accurate data support for subsequent wavelength optimization, thereby ensuring the accuracy and adaptability of blood oxygen detection.

[0030] Furthermore, the method also includes: When the transmittance is greater than the upper limit of the transmittance threshold interval, a second median optimization interval is constructed based on the lower limit wavelength of the red light wavelength restriction interval and the first median wavelength; median reconstruction loop analysis is performed according to the second median optimization interval.

[0031] Optionally, if the transmittance exceeds the upper limit of a preset transmittance threshold range (e.g., the threshold range is 0.7-0.9, and the current transmittance is 0.95), it indicates that the penetration depth of the current light wavelength is too high relative to the finger thickness, potentially affecting detection accuracy. In this case, the red light wavelength restriction range and the first median wavelength need to be re-optimized to determine a more suitable wavelength range. First, the lower limit wavelength of the red light wavelength restriction range and the first median wavelength are extracted, and a new optimization range, namely the second median optimization range, is constructed using these two values ​​as boundaries. This range represents a red light wavelength range that may be more suitable for the user's skin tone characteristics and finger thickness. Subsequently, the aforementioned median reconstruction analysis is performed again within this second median optimization range. This analysis process will continue to loop, with the system terminal calculating a new median wavelength and evaluating its penetration depth each time, gradually converging the optimization range until the transmittance falls into the preset threshold range. The finally determined median wavelength will serve as a key parameter of the reconstructed restriction range, used to optimize the red light wavelength range to better suit the characteristics of the current user's finger and detection needs.

[0032] The infrared emitter of the fingertip pulse oximeter is initialized according to the target wavelength of the red light band and the target wavelength of the infrared band.

[0033] In one embodiment, the system terminal initializes the infrared emitter of the fingertip pulse oximeter based on the determined target wavelengths for the red and infrared bands. The infrared emitter within the fingertip pulse oximeter typically consists of a tunable light source or filter assembly, capable of dynamically adjusting the emitted light wavelength within a certain range. The system terminal first inputs the target wavelengths for the red and infrared bands as initialization parameters into the control module. Subsequently, the control module adjusts the wavelength output of the infrared emitter according to the input target wavelengths. For example, by changing the excitation frequency of the emitting light source or adjusting the settings of the wavelength selector (such as a liquid crystal tunable filter or an acousto-optic tunable filter), the emitter outputs precise red and infrared wavelengths. After initialization, the emitter can stably output red and infrared light consistent with the target wavelengths, providing the necessary light source for subsequent pulse oximetry. This process ensures the device's adaptive adjustment to the user's skin color and finger characteristics, contributing to improved detection accuracy and applicability.

[0034] In the above text, refer to Figure 1 A method for skin color adaptive wavelength adjustment using a fingertip pulse oximeter according to an embodiment of the present invention is described in detail. Next, reference will be made to... Figure 2A skin color adaptive wavelength adjustment device for a fingertip pulse oximeter according to an embodiment of the present invention is described.

[0035] The skin-tone adaptive wavelength adjustment device for a fingertip pulse oximeter according to an embodiment of the present invention solves the technical problem that traditional fingertip pulse oximeters cannot adjust the detection wavelength in real time according to the user's skin tone, resulting in blood oxygen detection accuracy not matching skin tone. It achieves the technical effect of improving blood oxygen content detection accuracy and adapting to the needs of users with different skin tones by adaptively adjusting the target wavelengths of the red and infrared bands based on the characteristic values ​​of the user's finger skin tone. The fingertip pulse oximeter skin-tone adaptive wavelength adjustment device includes: an image acquisition module 1, a skin tone analysis module 2, a wavelength median optimization module 3, and an infrared emitter initialization module 4.

[0036] The system comprises: an image acquisition module 1, used to acquire an image of the user's finger when the user's finger is inserted into the detection cavity of the fingertip pulse oximeter via an image acquisition device built into the detection cavity; a skin color analysis module 2, used to perform skin color analysis based on the user's finger image to obtain skin color feature values ​​of the user's finger; a wavelength median optimization module 3, used to optimize the median wavelength of emitted light based on the user's finger skin color feature values ​​to obtain target wavelengths in the red light band and infrared band; and an infrared emitter initialization module 4, used to initialize the infrared emitter of the fingertip pulse oximeter according to the target wavelengths in the red light band and infrared band.

[0037] Furthermore, the skin color analysis module 2 also includes: Color neighborhood aggregation is performed on the user's finger image to obtain a first region, a second region, and so on up to the Nth region; finger shape matching is performed by traversing the first region, the second region, and so on up to the Nth region to obtain a finger shape calibration region, wherein the finger shape calibration region belongs to the first region, the second region, and so on up to the Nth region; the mode color feature value of the finger shape calibration region is extracted and set as the skin color feature value of the user's finger.

[0038] Furthermore, the skin color analysis module 2 also includes: The system receives the user's finger width input from the user terminal and obtains the finger length that the detection cavity can accommodate. Based on the user's finger width and the finger length that can accommodate, a finger mold is constructed. The system traverses the first region, the second region, and up to the Nth region, and performs similarity analysis with the finger mold to obtain a first similarity value, a second similarity value, and up to the Nth similarity value. The system extracts regions from the first similarity value, the second similarity value, and up to the Nth similarity value that are greater than or equal to the similarity value threshold, and sets them as candidate calibration regions. When the number of candidate calibration regions is 1, the candidate calibration region is set as the finger calibration region. When the number of candidate calibration regions is not 1, the system updates the user's finger image and performs a skin color analysis loop.

[0039] Furthermore, the wavelength median optimization module 3 also includes: Obtain the red light wavelength restriction range and the infrared wavelength restriction range; based on the user's finger skin color feature value, perform median reconstruction on the red light wavelength restriction range and the infrared wavelength restriction range to obtain the red light wavelength reconstruction restriction range and the infrared wavelength reconstruction restriction range; extract the target wavelength of the red light band with the smallest deviation from 650nm from the red light wavelength reconstruction restriction range; extract the target wavelength of the infrared band with the smallest deviation from 850nm from the infrared wavelength reconstruction restriction range.

[0040] Furthermore, the wavelength median optimization module 3 also includes: Extract the lower limit wavelength of the red light wavelength restriction range and the upper limit wavelength of the infrared wavelength restriction range to construct a first median optimization range; take the median of the first median optimization range to obtain a first median wavelength; predict the penetration depth based on the user's finger skin color feature value and the first median wavelength to obtain a first penetration depth; receive the user's finger thickness at the user end; calculate the ratio of the first penetration depth to the user's finger thickness and set it as the penetration ratio; when the penetration ratio belongs to the penetration ratio threshold range, set the first median wavelength as the lower limit wavelength threshold, and reconstruct the red light wavelength restriction range and the infrared wavelength restriction range using the lower limit wavelength threshold as the lower limit wavelength to obtain the red light wavelength reconstruction restriction range and the infrared wavelength reconstruction restriction range.

[0041] Furthermore, the wavelength median optimization module 3 also includes: When the transmittance is greater than the upper limit of the transmittance threshold interval, a second median optimization interval is constructed based on the lower limit wavelength of the red light wavelength restriction interval and the first median wavelength; median reconstruction loop analysis is performed according to the second median optimization interval.

[0042] Furthermore, the wavelength median optimization module 3 also includes: Collect finger red light penetration experiment data, which includes a skin color feature vector variable dataset, a penetrating red light wavelength variable dataset, and a penetration depth calibration dataset. Use the penetration depth calibration dataset as output supervision, and use the skin color feature vector variable dataset and the penetrating red light wavelength variable dataset as input to perform random forest configuration to obtain a penetration depth prediction model. Based on the penetration depth prediction model, process the user's finger skin color feature value and the first median wavelength to obtain the first penetration depth.

[0043] The skin color adaptive wavelength adjustment device for a fingertip pulse oximeter provided in this embodiment of the invention can execute the skin color adaptive wavelength adjustment method for a fingertip pulse oximeter provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0044] Although this application makes various references to certain modules in the apparatus according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not intended to limit the scope of protection of this invention.

[0045] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for skin color adaptive wavelength adjustment using a fingertip pulse oximeter, characterized in that, include: When a user inserts their finger into the detection chamber of the fingertip pulse oximeter, an image of the user's finger is obtained through an image acquisition device built into the detection chamber; Skin color analysis is performed on the user's finger image to obtain the user's finger skin color feature values; Based on the user's finger skin color feature value, the median wavelength of emitted light is optimized to obtain the target wavelength in the red light band and the target wavelength in the infrared band. The infrared emitter of the fingertip pulse oximeter is initialized according to the target wavelength of the red light band and the target wavelength of the infrared band; Specifically, based on the user's finger skin color feature value, the median wavelength of emitted light is optimized to obtain the target wavelength in the red light band and the target wavelength in the infrared band, including: Obtain the red light wavelength restriction range and the infrared wavelength restriction range; Based on the user's finger skin color feature value, the median reconstruction of the red light wavelength restriction range and the infrared wavelength restriction range is performed to obtain the red light wavelength reconstruction restriction range and the infrared wavelength reconstruction restriction range. Extract the target wavelength of the red light band that has the smallest deviation from 650nm from the red light wavelength reconstruction restriction range; Extract the target wavelength of the infrared band that has the smallest deviation from 850nm from the infrared wavelength reconstruction restriction range; Specifically, based on the user's finger skin color feature value, the median reconstruction of the red light wavelength restriction range and the infrared wavelength restriction range is performed to obtain the red light wavelength reconstruction restriction range and the infrared wavelength reconstruction restriction range, including: Extract the lower limit wavelength of the red light wavelength restriction range, extract the upper limit wavelength of the infrared wavelength restriction range, and construct the first median optimization range; The first median wavelength is obtained by taking the median value of the first median optimization interval; Based on the user's finger skin color feature value and the first median wavelength, the penetration depth is predicted to obtain the first penetration depth; The thickness of the user's finger shape received from the user terminal; Calculate the ratio of the first penetration depth to the thickness of the user's finger shape, and set it as the penetration ratio; When the transmittance falls within the transmittance threshold range, the first median wavelength is set as the lower limit wavelength threshold. Using the lower limit wavelength threshold as the lower limit wavelength, the red light wavelength restriction range and the infrared wavelength restriction range are reconstructed to obtain the red light wavelength reconstruction restriction range and the infrared wavelength reconstruction restriction range. This also includes: When the transmittance is greater than the upper limit of the transmittance threshold range, a second median optimization range is constructed based on the lower limit wavelength of the red light wavelength restriction range and the first median wavelength. Perform median reconstruction cyclic analysis based on the second median optimization interval.

2. The method as described in claim 1, characterized in that, Skin color analysis is performed on the user's finger image to obtain the user's finger skin color feature values, including: Color neighborhood aggregation is performed on the user's finger image to obtain the first region, the second region, and so on up to the Nth region; The finger shape matching is performed by traversing the first region, the second region up to the Nth region to obtain the finger shape marking region, wherein the finger shape marking region belongs to the first region, the second region up to the Nth region; Extract the mode color feature value of the finger shape calibration area and set it as the skin color feature value of the user's finger.

3. The method as described in claim 2, characterized in that, Traverse the first region, the second region, up to the Nth region for finger shape matching to obtain the finger shape calibration region, including: Receive the user's finger width input from the user terminal, and obtain the finger length that the detection cavity can accommodate; Based on the user's finger width and the finger's accommodating length, construct a finger mold; Traverse the first region, the second region up to the Nth region, and perform similarity analysis with the finger mold respectively to obtain the first similarity value, the second similarity value up to the Nth similarity value; Extract the regions from the first similarity value, the second similarity value up to the Nth similarity value that are greater than or equal to the similarity value threshold, and set them as candidate calibration regions; When the number of candidate calibration regions is 1, the candidate calibration region is set as the finger-shaped calibration region; When the number of candidate calibration regions is not 1, update the user's finger image and perform a skin color analysis loop.

4. The method as described in claim 1, characterized in that, Based on the user's finger skin color feature value and the first median wavelength, a first penetration depth is predicted, including: Collect finger red light penetration experiment data, wherein the finger red light penetration experiment data includes skin color feature vector variable dataset, penetrating red light wavelength variable dataset and penetration depth calibration dataset; Using the penetration depth calibration dataset as output supervision, and the skin color feature vector variable dataset and the penetration red light wavelength variable dataset as input, a random forest configuration is performed to obtain a penetration depth prediction model. Based on the penetration depth prediction model, the user's finger skin color feature value and the first median wavelength are processed to obtain the first penetration depth.

5. A skin color adaptive wavelength adjustment device for a fingertip pulse oximeter, characterized in that, The device is used to implement the skin color adaptive wavelength adjustment method for a fingertip pulse oximeter as described in any one of claims 1-4, comprising: Image acquisition module: When a user's finger is inserted into the detection chamber of the fingertip pulse oximeter, an image of the user's finger is obtained through the image acquisition device built into the detection chamber; Skin color analysis module: Performs skin color analysis based on the user's finger image to obtain the user's finger skin color feature values; Median wavelength optimization module: Based on the user's finger skin color feature value, optimize the median wavelength of emitted light to obtain the target wavelength in the red light band and the target wavelength in the infrared band; Infrared transmitter initialization module: Initializes the infrared transmitter of the fingertip pulse oximeter according to the target wavelength of the red light band and the target wavelength of the infrared band; The wavelength median optimization module further includes: obtaining a red light wavelength restriction range and an infrared wavelength restriction range; performing median reconstruction on the red light wavelength restriction range and the infrared wavelength restriction range based on the user's finger skin color feature value to obtain a red light wavelength reconstruction restriction range and an infrared wavelength reconstruction restriction range; extracting the target wavelength of the red light band with the smallest deviation from 650nm from the red light wavelength reconstruction restriction range; and extracting the target wavelength of the infrared band with the smallest deviation from 850nm from the infrared wavelength reconstruction restriction range. The wavelength median optimization module further includes: extracting the lower limit wavelength of the red light wavelength restriction range and the upper limit wavelength of the infrared wavelength restriction range to construct a first median optimization range; taking the median of the first median optimization range to obtain a first median wavelength; predicting the penetration depth based on the user's finger skin color feature value and the first median wavelength to obtain a first penetration depth; receiving the user's finger thickness at the user end; calculating the ratio of the first penetration depth to the user's finger thickness and setting it as the penetration ratio; when the penetration ratio belongs to the penetration ratio threshold range, setting the first median wavelength as the lower limit wavelength threshold, and using the lower limit wavelength threshold as the lower limit wavelength to reconstruct the red light wavelength restriction range and the infrared wavelength restriction range to obtain the red light wavelength reconstruction restriction range and the infrared wavelength reconstruction restriction range. The wavelength median optimization module further includes: when the transmittance is greater than the upper limit of the transmittance threshold interval, constructing a second median optimization interval based on the lower limit wavelength of the red light wavelength restriction interval and the first median wavelength; and performing median reconstruction loop analysis based on the second median optimization interval.

Images