Non-invasive diagnostic data acquisition method and system

Abstract

The invention provides a non-invasive diagnostic data acquisition method and system, and relates to the technical field of data acquisition, and the method comprises the steps: obtaining original optical acquisition data and original contact pressure data, and carrying out the time alignment processing; contact pressure change recognition processing is executed, and contact pressure influence subdomain data is generated; according to the contact pressure influence domain data, performing optical path disturbance positioning processing on original optical acquisition data in the alignment data to generate optical disturbance fragment data; according to the optical interference fragment data, executing optical compensation construction processing on the corresponding optical interference fragment to generate domain-divided compensation optical data; according to the sub-domain compensation optical data, sub-domain splicing processing is executed between the contact pressure influence sub-domains, and continuous correction optical data is generated; according to the continuous correction optical data, target data extraction processing is executed, and non-invasive diagnosis data is generated; according to the invention, the autonomy and accuracy of non-invasive diagnostic data acquisition are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data acquisition, in particular to a method and system for collecting non-invasive diagnostic data. BACKGROUND

[0002] The existing non-invasive diagnostic data acquisition technology mainly uses optical sensing (such as near-infrared spectroscopy NIRS), electrophysiological sensing (such as photoplethysmography PPG), or millimeter wave radar-based vital sign detection to measure the changes in human tissue, hemodynamic parameters or physiological signals in a non-contact or surface contact manner. For example, the photoplethysmography-based acquisition system usually uses a light-emitting diode and a photosensitive diode to calculate the user's heart rate, blood oxygen saturation and other physiological indicators by detecting the intensity changes of transmitted or reflected light. This kind of system usually needs to fix the sensor on the skin surface to realize real-time acquisition of non-invasive diagnostic data by collecting continuous optical waveforms and combining pre-set optical path models, calibration coefficients or empirical algorithms.

[0003] However, when the above optical or electrophysiological sensing system is applied to a scene where the human body is in motion (for example, the user is slightly walking or hand shaking), the collected signal is easily affected by baseline drift and waveform distortion caused by dynamic contact pressure changes. For example, in a wearable photoplethysmography device, a small relative displacement between the sensor and the skin will cause a change in the optical path length, resulting in a low-frequency fluctuation in the reflected light intensity that is unrelated to the real pulse wave. In actual research, this low-frequency interference may partially overlap with the frequency band of the real physiological signal, making it difficult for traditional filtering algorithms to effectively distinguish, resulting in heart rate calculation jumps, waveform peak shifts or pulse waveform distortion, affecting the reliability of subsequent diagnostic models. Therefore, in the data acquisition involving dynamic body scenes, the traditional system often has difficulty in ensuring the stability and accuracy of non-invasive diagnostic data. SUMMARY

[0004] The purpose of the present application is to provide a method and system for collecting non-invasive diagnostic data, which aims to solve the problems mentioned in the background.

[0005] To solve the above technical problems, the technical solutions of the present application are as follows: In a first aspect, a method for collecting non-invasive diagnostic data, the method comprising: obtaining original optical acquisition data and original contact pressure data, and performing time alignment processing to form aligned data; performing contact pressure change identification processing according to the original contact pressure data in the aligned data, and generating contact pressure influence sub-domain data through contact pressure change amplitude and contact pressure change rate; According to the contact pressure influence sub-domain data, the original optical acquisition data in the alignment data is subjected to light path disturbance positioning processing, the light intensity slope change synchronous with the contact pressure change is identified, and optical interference segment data is generated; According to the optical interference segment data, optical compensation construction processing is performed on the corresponding optical interference segment, the change direction, change amplitude of the interference segment and the pressure change trend of the contact pressure influence sub-domain to which the interference segment belongs are obtained, and sub-domain compensation optical data is generated; According to the sub-domain compensation optical data, sub-domain splicing processing is performed between each contact pressure influence sub-domain, the light intensity difference, phase difference and waveform period difference between adjacent sub-domains are obtained, and continuous correction optical data is generated; According to the continuous correction optical data, target data extraction processing is performed, and non-invasive diagnosis data is generated.

[0006] Preferably, according to the original contact pressure data in the alignment data, contact pressure change identification processing is performed, and contact pressure influence sub-domain data is generated by contact pressure change amplitude and contact pressure change rate, including: The adjacent sampling point difference of the original contact pressure data in the alignment data is calculated to generate pressure change amplitude data; According to the pressure change amplitude data, the change rate is calculated according to the change amount of adjacent time interval, and the pressure change rate data is generated; According to the pressure change amplitude data and the pressure change rate data, the preset pressure change amplitude threshold and the preset pressure change rate threshold are compared respectively, and the continuous time segment in the comparison result which meets any one of the preset threshold matching conditions is recorded as candidate pressure change segment data; According to the time continuity of the candidate pressure change segment data, the candidate pressure change segment with a time interval exceeding a preset interval threshold is split to generate a plurality of pressure change sub-segment data; According to the start point and end point of the pressure change sub-segment data, the corresponding sub-segment boundary point data is generated, and the contact pressure influence sub-domain data is generated according to the sub-segment boundary point data.

[0007] Preferably, according to the contact pressure influence sub-domain data, the original optical acquisition data in the alignment data is subjected to light path disturbance positioning processing, the light intensity slope change synchronous with the contact pressure change is identified, and optical interference segment data is generated, including: According to the contact pressure influence sub-domain data, the corresponding original optical acquisition data in each sub-domain is extracted to form sub-domain optical data; The light intensity difference of the adjacent sampling points of the sub-domain optical data is calculated to generate light intensity change slope data; According to the light intensity change slope data and the pressure change trend in the corresponding sub-domain, light intensity synchronous change point data is generated; According to the continuity of the light intensity synchronous change point data on the time axis, the continuous synchronous change point is combined as optical disturbance segment data; According to the slope change of the optical disturbance segment data at the boundary, the interference boundary point is identified, and interference segment boundary point data is generated; According to the optical disturbance segment data and the interference segment boundary point data, optical interference segment data is generated.

[0008] Preferably, according to the optical interference segment data, optical compensation construction processing is performed on the corresponding optical interference segment, the change direction, the change amplitude of the interference segment and the pressure change trend of the contact pressure influence subdomain to which it belongs are determined, and subdomain compensation optical data is generated, including: According to the optical interference segment data, the change direction data of each interference segment is determined, and compensation trend data opposite to the direction is generated according to the change direction data; According to the change amplitude data of the interference segment in the optical interference segment data, the target compensation amount data is generated by distributing uniformly according to the time length or proportionally according to the pressure change trend; According to the target compensation amount data, compensation transition zone data is constructed at the start and end positions of the interference segment, so that the compensation amount gradually increases or gradually decreases from zero in the compensation transition zone, and compensation transition data is generated; According to the compensation trend data, the target compensation amount data and the compensation transition data, compensation processing is performed on the optical interference segment, and compensated interference segment data is generated; According to the compensated interference segment data and the optical acquisition data not affected by the interference, time sequence splicing is performed to generate subdomain compensation optical data.

[0009] Preferably, according to the change amplitude data of the interference segment in the optical interference segment data, the target compensation amount data is generated by distributing uniformly according to the time length or proportionally according to the pressure change trend, including: According to the optical interference segment data, the change amplitude data of the interference segment is extracted, and the interference segment time interval data is generated according to the start time and end time of the interference segment; According to the interference segment time interval data, the change amplitude data is divided into a plurality of compensation subzones according to the time sequence, so that each compensation subzone corresponds to a time distribution segment of the change amplitude data, and compensation subzone time period data is generated; According to the compensation subzone time period data, the change amplitude data is uniformly distributed and mapped to each compensation subzone, and uniform distribution compensation amount data is generated; According to the pressure change trend of the contact pressure influence subdomain to which the interference segment belongs, the change amplitude data is distributed according to the change relationship of each compensation subzone according to the pressure change trend, and trend compensation amount data is generated; Based on the compatibility conditions between uniformly distributed compensation data and trend compensation data, a compensation distribution method that better matches the waveform recovery characteristics of the interference segment is selected from the two to generate target compensation data.

[0010] Preferably, compensation transition zone data is constructed based on the start and end positions of the interference segment according to the target compensation amount data, so that the compensation amount gradually increases or decreases from zero within the compensation transition zone, generating compensation transition data, including: The starting point and ending point data of the interference segment are determined based on the optical interference segment data, and the boundary point data of the interference segment are generated based on the starting point and ending point data. Based on the boundary point data of the interference segment, the boundary compensation area is divided into several compensation transition sub-regions, and compensation transition sub-region data is generated. Based on the compensation transition sub-region data, the target compensation amount data is gradually processed in a pattern of gradually increasing from zero to the target compensation amount or gradually decreasing from the target compensation amount, to generate compensation transition sequence data. Based on the compensation gradient sequence data, continuous adjustment processing is performed between adjacent sampling points to ensure that the compensation amount changes smoothly point by point within the compensation transition zone, thereby generating compensation gradient adjustment data; Based on the compensation gradual adjustment data, construct complete compensation transition zone data.

[0011] Preferably, compensation processing is performed on the optical interference segments based on compensation trend data, target compensation amount data, and compensation transition data to generate compensated interference segment data, including: Based on the compensation trend data, the optical interference segment data is subjected to orientation correction processing, and the compensation direction is adjusted to be opposite to the change direction of the interference segment to generate orientation correction data. Based on the orientation correction data, the target compensation data is used to perform compensation allocation processing on each sampling point inside the interference segment. The compensation amount is allocated according to the time sequence of the sampling point position to generate internal compensation allocation data. Based on the internal compensation allocation data and compensation transition data, a compensation gradient superposition process is performed at the boundary of the interference segment to ensure that the compensation amount maintains a continuous transition between the internal part and the boundary part of the interference segment, thereby generating compensation superposition structure data. Compensation and fusion processing is performed on the entire interference segment based on the compensated overlay structure data. By fusing the compensated overlay structure data with the original interference segment data in the time series, compensated interference segment data is generated.

[0012] Secondly, a non-invasive diagnostic data acquisition system, the system comprising: The data alignment module is used to acquire raw optical acquisition data and raw contact pressure data, and perform time alignment processing to form aligned data; a contact pressure change identification module, configured to perform a contact pressure change identification process according to the original contact pressure data in the alignment data, to generate contact pressure influence sub-domain data through a contact pressure change amplitude and a contact pressure change rate; a light path disturbance positioning module, configured to perform a light path disturbance positioning process on the original optical acquisition data in the alignment data according to the contact pressure influence sub-domain data, to generate optical interference segment data by identifying a light intensity slope change that occurs synchronously with the contact pressure change; an optical compensation construction module, configured to perform an optical compensation construction process on a corresponding optical interference segment according to the optical interference segment data, to generate sub-domain compensation optical data through a change direction, a change amplitude of the interference segment, and a pressure change trend of a contact pressure influence sub-domain to which the interference segment belongs; a sub-domain splicing module, configured to perform a sub-domain splicing process between each contact pressure influence sub-domain according to the sub-domain compensation optical data, to generate continuous correction optical data through a light intensity difference value, a phase difference value, and a waveform period difference value between adjacent sub-domains; a target data extraction module, configured to perform a target data extraction process according to the continuous correction optical data, to generate non-invasive diagnosis data.

[0013] The above scheme of the present application at least includes the following beneficial effects: (1) Improve data stability and accuracy The present application forms accurate alignment data by performing time alignment processing on the original optical acquisition data and the original contact pressure data. This processing method makes the two types of data maintain a strict corresponding relationship on the time axis, providing a reliable basis for subsequent interference identification and compensation. Compared with the prior art, the time alignment method can effectively reduce the optical signal distortion and baseline drift caused by the relative displacement of the sensor, and significantly improve the stability and accuracy during the data acquisition process.

[0014] (2) Effectively isolate dynamic interference and real physiological signals The traditional non-invasive diagnosis system is easily disturbed by dynamic contact pressure changes in a motion state. The present application analyzes the contact pressure data, identifies the pressure change amplitude and rate, and generates contact pressure influence sub-domain data. This method can accurately divide the influence intervals of different pressure changes, facilitate subsequent independent interference identification and compensation in the sub-domain, effectively avoid the phenomenon of dynamic interference and physiological signal overlap in the traditional system, and improve the recognition of real physiological signals.

[0015] (3) Strongly targeted optical interference compensation The present application is based on the synchronization of contact pressure changes and optical signal changes, uses light intensity slope changes to identify optical interference segments, and performs targeted compensation on these interference segments. By analyzing the change direction, amplitude and pressure change trend of the interference segments, a compensation amount is generated and smoothed through a compensation transition zone, so that the recovered optical signal can more truly reflect the physiological waveform, avoiding waveform distortion or failure caused by simple filtering in traditional systems.

[0016] (Four) Enhance data splicing processing capability In the present application, based on the compensation of optical data in different domains, the optical data in different domains is spliced and processed, and the light intensity difference, phase difference and waveform period difference between adjacent domains are adjusted to generate continuous corrected optical data. This splicing method can ensure smooth connection of optical data between different domains, avoid discontinuity or sudden change in traditional systems during processing, and keep the final data smooth and coherent on the time axis, improving the overall availability and diagnostic accuracy of the data.

[0017] (Five) Improve the reliability and practicality of non-invasive diagnostic data The present application generates continuous corrected optical data by accurately compensating and processing interference segments, and extracts target data for non-invasive diagnosis. This method avoids signal distortion caused by sensor movement, making the final non-invasive diagnostic data more stable and reliable, especially suitable for physiological signal acquisition in motion state. Compared with the prior art, the present application can better acquire physiological information such as heart rate and blood oxygen saturation in dynamic scenes, providing more accurate and effective input for subsequent diagnostic models.

[0018] (Six) Suitable for practical application scenarios By implementing the data processing method in the present application, the non-invasive diagnostic data acquisition system can better adapt to dynamic changes in practical application scenarios, such as hand shaking and slight walking in motion state. This scheme can accurately identify and compensate for optical interference caused by contact pressure changes, providing more stable and efficient technical support for health monitoring in actual use. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 is a flowchart of the non-invasive diagnostic data acquisition method provided by the embodiment of the present application. DETAILED DESCRIPTION

[0020] Exemplary embodiments of the present disclosure will be described in greater detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it is understood that the present disclosure can be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the present disclosure to those skilled in the art.

[0021] As Figure 1 shown, the embodiment of the present application proposes a method for collecting non-invasive diagnostic data, the method comprising: obtaining original optical acquisition data and original contact pressure data, and performing time alignment processing to form aligned data; According to the original contact pressure data in the aligned data, performing contact pressure change identification processing, and generating contact pressure influence sub-domain data through contact pressure change amplitude and contact pressure change rate; According to the contact pressure influence sub-domain data, performing optical path disturbance positioning processing on the original optical acquisition data in the aligned data, and generating optical interference segment data by identifying the light intensity slope change that appears synchronously with the contact pressure change; According to the optical interference segment data, performing optical compensation construction processing on the corresponding optical interference segment, and generating sub-domain compensation optical data through the change direction, change amplitude of the interference segment and the pressure change trend of the contact pressure influence sub-domain to which it belongs; According to the sub-domain compensation optical data, performing sub-domain splicing processing between each contact pressure influence sub-domain, and generating continuous corrected optical data through the light intensity difference, phase difference and waveform period difference between adjacent sub-domains; According to the continuous corrected optical data, performing target data extraction processing to generate non-invasive diagnostic data.

[0022] In the embodiment of the present application, by performing time alignment processing on the original optical acquisition data and the original contact pressure data, the two types of data maintain a corresponding relationship on the time axis, thereby providing a common analysis basis for subsequent identification of the correlation between contact pressure change and optical signal change, and facilitating improvement of the coherence of the overall data processing process. On this basis, by identifying the change amplitude and change rate of the contact pressure data, further generating contact pressure influence sub-domains, the data intervals under different pressure states in the acquisition process can be distinguished, and the data structure for subsequent independent interference identification and compensation by sub-domain is established.

[0023] According to the contact pressure influence sub-domain, the light intensity slope change synchronized with the pressure change is identified from the aligned optical acquisition data, so that the optical anomaly in the acquisition waveform caused by pressure fluctuation can be accurately located in the corresponding sub-domain. By synchronism comparison and slope change judgment, these abnormal positions can be combined into optical interference segments, providing a clear target area for subsequent compensation operation, so that the compensation operation has higher pertinence.

[0024] In the optical compensation construction process, by calculating the change direction, change amplitude and corresponding pressure change trend of the interference segment, a compensation trend opposite to the interference change is formed, and the compensation amount and compensation transition zone are constructed accordingly, so that the recovery processing can take into account the continuity of the internal and boundary parts of the segment. Further, by performing splicing operation between each contact pressure influence sub-domain and adjusting according to the light intensity difference, phase difference and period difference, the optical data after final recovery forms a continuous whole waveform between each sub-domain, reducing the splicing inconsistency phenomenon caused by sub-domain compensation. Finally, the target diagnostic result is extracted from the continuous corrected optical data, so that the final output non-invasive diagnostic data has more stable waveform characteristics and more reliable parameter basis.

[0025] For example, in the actual scene of collecting photoelectric plethysmography data of human fingers, pressure increase or decrease caused by slight movement of fingers may occur during the collection process. When such situations occur, the contact pressure data will first reflect the corresponding changes, and different pressure influence intervals are divided accordingly. Subsequently, in the optical data within the same interval, the light intensity change synchronized with the pressure can be identified, which is marked as an interference segment. The compensation amount is generated according to the change trend of these interference segments, and a gentle transition is constructed at the front and rear positions, so that the corrected data is continuous between each section. Then, the pulse waveform or other diagnostic data is extracted from the corrected data, which helps to obtain more accurate non-invasive physiological information when interference exists.

[0026] In a preferred embodiment of the present application, the original optical acquisition data and the original contact pressure data are obtained, and time alignment processing is performed to form aligned data, which specifically includes: The optical signal of the detection site is continuously collected by the optical sensor, and the sampled light intensity values are recorded in sequence according to the sampling time to form the original optical collection data. At the same time, the pressure change between the sensor and the skin is collected by the pressure sensor, and the pressure value of each sampling is recorded in sequence as the original contact pressure data. The sampling time of the two types of data is extracted as a time sequence parameter, and by comparing the starting sampling time and sampling interval of the two types of data, the data with higher sampling density is resampled in sequence to make the sampling interval consistent with the other group of data. Then, the two groups of sampling values are corresponded according to the adjusted time sequence, so that the optical sampling point and the corresponding pressure sampling point are consistent, thereby generating the aligned data with complete overall structure, so that the subsequent steps can be analyzed and processed based on the same time reference.

[0027] In a preferred embodiment of the present application, according to the domain-compensated optical data, a domain splicing process is performed between each contact pressure influence domain, and continuous corrected optical data is generated through the light intensity difference, phase difference and waveform period difference between adjacent domains, specifically including: The domain-compensated optical data corresponding to all contact pressure influence domains is rearranged in time sequence, and the starting position and ending position of each domain are recorded. For the two adjacent domains, the light intensity values at both ends are extracted at the domain boundary and the difference between the two is calculated to determine whether there is a mutation in the compensated light intensity. The phase change trend of the two waveforms is compared, and whether the phase of the two waveforms is offset is determined according to the order of the peak positions on the two waveforms. Further, the period lengths of the two waveforms are compared to confirm whether the rhythm of the waveforms between the domains is consistent. According to the size of the light intensity difference, the phase difference and the period difference, the time period that needs to be adjusted is determined, and operations such as waveform translation, local stretching and point-by-point buffer adjustment are performed on the domain with larger difference to make the two waveforms gradually consistent at the domain boundary. Finally, all the domains are spliced in turn to form continuous corrected optical data without obvious boundary, which is used for subsequent diagnostic data extraction.

[0028] In a preferred embodiment of the present application, according to the continuous corrected optical data, a target data extraction process is performed to generate non-invasive diagnostic data, specifically including: The continuous correction optical data is preprocessed, and first, peak points, valley points, and structural features such as rising and falling sections of the overall waveform are identified. According to the waveform structure, the waveform segment corresponding to a continuous cardiac cycle is extracted, and the main morphological parameters of each cycle are analyzed, such as peak height, waveform rising time, falling time, and cycle length. The characteristic data of multiple cycles are statistically analyzed, and parameters related to physiological information of the collection area that change periodically are identified, such as the cycle frequency or amplitude variation trend of the pulse wave signal. In combination with the waveform characteristics in the continuous section, by screening and combining these structural feature parameters, non-invasive diagnostic data for subsequent diagnostic purposes are generated, so that the analysis results can reflect the physiological fluctuations of the collection object under the current collection state.

[0029] In a preferred embodiment of the present application, according to the original contact pressure data in the alignment data, a contact pressure change identification process is performed, and contact pressure influence sub-domain data is generated by contact pressure change amplitude and contact pressure change rate, including: The original contact pressure data in the alignment data is subjected to adjacent sample point difference calculation to generate pressure change amplitude data; According to the pressure change amplitude data, the change rate is calculated according to the change amount of the adjacent time interval to generate pressure change rate data; According to the pressure change amplitude data and the pressure change rate data, the preset pressure change amplitude threshold and the preset pressure change rate threshold are compared respectively, and the continuous time segment that meets any preset threshold matching condition in the comparison result is recorded as candidate pressure change segment data; According to the time continuity of the candidate pressure change segment data, the candidate pressure change segment with a time interval exceeding a preset interval threshold is split to generate multiple pressure change segment data; According to the start point and the end point of the pressure change segment data, corresponding segment boundary point data is generated, and contact pressure influence sub-domain data is generated according to the segment boundary point data.

[0030] In the embodiment of the present application, by calculating the difference between adjacent sampling points of the original contact pressure data, the instantaneous amplitude of pressure change can be embodied in the form of data, which is convenient for judging whether pressure fluctuation occurs during the acquisition process. By calculating the amplitude change amount in combination with the adjacent time interval, the pressure change rate can be further obtained, which expands the pressure change process from single amplitude description to amplitude and rate representation, which helps to identify different pressure states of short-time rapid change and long-time gentle change. Then, comparing the above two types of change data with the preset amplitude threshold and rate threshold can preliminarily distinguish the continuous time segments in different pressure change modes and form candidate regions. Splitting the candidate regions according to time continuity can avoid small segment misjudgment caused by sampling jitter, and make the final pressure change segmentation more stable and consistent. By generating the segmentation boundary points, different pressure change modes can be clearly divided by clear boundaries, thereby forming a complete contact pressure influence domain, so that the subsequent optical data interference recognition can be analyzed within the clear domain range.

[0031] In a preferred embodiment of the present application, the method for setting the preset pressure change amplitude threshold comprises: According to the collected original contact pressure data, first, statistical analysis of the data is performed. By calculating the standard deviation, maximum value and minimum value of the contact pressure data sampled multiple times within a period of time, the fluctuation range of the sampling data is determined. Then, a pressure change amplitude threshold is set, which can be selected based on actual application requirements. For example, in normal use, the change amplitude of the contact pressure should generally be within an acceptable range, and changes exceeding this range are considered abnormal. In the implementation process, the threshold value can be selected with reference to the sensitivity of the device or the maximum pressure change fluctuation of the typical application scenario. The threshold value is usually set to twice the standard deviation of the contact pressure data or other appropriate multiples, to ensure sufficient sensitivity to sudden large changes. Finally, according to the selected amplitude threshold, if the original contact pressure change amplitude exceeds the threshold, it is considered as a significant pressure change event and is processed subsequently.

[0032] In a preferred embodiment of the present application, the method for setting the preset pressure change rate threshold comprises: The average, maximum and minimum of the pressure change rate are calculated by statistical analysis of the change rate of the contact pressure data. First, the pressure change rate between every two consecutive sampling points is calculated, which is the change amount of pressure per unit time. Then, based on these rate data, a suitable rate threshold is selected. The threshold is usually set to be several times the average or standard deviation of the pressure change rate data, ensuring that the threshold is sufficient to identify those relatively sharp pressure changes. For example, if the sampling interval is 1 millisecond, the rate threshold can be set to twice or more the standard rate. The setting of the rate threshold can also be adjusted according to the application scenario, and if the device requires a response to rapid changes in pressure, the threshold can be set lower. Finally, if the pressure change rate exceeds the set threshold, the pressure change at that moment is considered to be a sudden or rapid change, and the subsequent potential change segment is marked for further analysis and processing.

[0033] In a preferred embodiment of the present application, the method for setting the preset interval threshold specifically includes: According to the time interval information of the collected contact pressure data, the time interval of consecutive sampling points is first counted, and the average and standard deviation of the time interval are calculated. Then, a preset interval threshold is set, which can be selected according to actual needs. Generally, if the sampling time interval is too long (for example, the time interval between sampling points is greater than the set threshold), it may represent loss or abnormality in the sampling process, resulting in intermittent pressure changes. In order to avoid this situation, the threshold can be set to the average value of the time interval plus two or more times the standard deviation, to ensure that the threshold is effective within the normal fluctuation range. For example, if the average value of the sampling time interval is 100 milliseconds and the standard deviation is 20 milliseconds, the set interval threshold can be 140 milliseconds or higher. If the time interval between consecutive sampling time points exceeds the preset threshold, it is considered that the data in this segment has discontinuity or abnormality, and further data splitting and processing are performed.

[0034] In a preferred embodiment of the present application, the original contact pressure data in the aligned data is subjected to adjacent sampling point difference calculation to generate pressure change amplitude data, specifically including: Each contact pressure sampling point is read from the aligned data in sequence to obtain a time-ordered sequence of all pressure values. For two pressure values in adjacent positions, the numerical difference between them is calculated, and the difference is taken as the pressure change amplitude at the corresponding time. Specifically, the pressure value of the latter sampling point is directly subtracted from the pressure value of the former sampling point to obtain the numerical size and direction of the pressure change. All adjacent sampling point change amplitudes are recorded in time order to form a set of pressure change amplitude data describing the amplitude of pressure change over time, which is used for subsequent quantitative analysis of the degree of pressure change.

[0035] In a preferred embodiment of the present application, the pressure change rate data is generated according to the change amount of adjacent time intervals based on the pressure change amplitude data, and specifically includes: The change amplitude value of each sample point is read from the pressure change amplitude data, and the corresponding sampling time when the change amplitude is recorded is obtained. The sampling times of two consecutive change amplitudes are differentiated to determine the time interval therebetween. Then, the change amplitude value is compared with the corresponding time interval, and the change of the change amplitude within the time interval is expressed as the change amount within a unit time, thereby generating the pressure change rate corresponding to each sample point. All the calculated change rates are recorded in time sequence, so that the speed of change of the pressure value in the time dimension is quantitatively reflected, thereby providing basic data for subsequent identification of the rapid change or slow change section.

[0036] In a preferred embodiment of the present application, the pressure change amplitude data and the pressure change rate data are compared with the preset pressure change amplitude threshold and the preset pressure change rate threshold, respectively, and the continuous time segments that meet any one of the preset threshold matching conditions in the comparison results are recorded as candidate pressure change section data, and specifically includes: The pressure change amplitude data of each sample time is judged, and when the change amplitude is higher than the preset pressure change amplitude threshold, the time point is marked as a point where the pressure change may exist. Similarly, the pressure change rate data of each sample time is judged, and when the change rate is higher than the preset pressure change rate threshold, the time point is marked as a point where the pressure change may exist. Subsequently, the continuity of these marked time points is analyzed according to the time sequence, adjacent or short-interval marked time points are combined into continuous time segments, and these segments are recorded as candidate pressure change section data for preliminary identification of the section where the pressure fluctuation may occur.

[0037] In a preferred embodiment of the present application, the candidate pressure change section data is split according to the time continuity, and the candidate pressure change section with a time interval exceeding a preset interval threshold is split to generate a plurality of pressure change section data, and specifically includes: Each time segment in the candidate pressure change section data is traversed, and the sampling time interval between any two adjacent points in the segment is detected. When a certain adjacent time interval is greater than a preset interval threshold, the segment is cut at the interval position, so that two independent time sections are formed before and after the segment. The above process is repeated for all segments, and finally the candidate segment that is too long or has internal discontinuity can be divided into a plurality of pressure change section data with higher continuity, so that the pressure change within each section has more consistent change characteristics, thereby providing a more accurate structural basis for subsequent generation of domain boundaries.

[0038] In a preferred embodiment of the present application, the corresponding segment boundary point data is generated according to the start and end points of the pressure change segment data, and the contact pressure influence subdomain data is generated according to the segment boundary point data, specifically including: The start sampling time and the end sampling time of each pressure change segment are extracted respectively, and the two time points are defined as the corresponding segment boundary points. The boundary points of all segments are sorted in time sequence, so that the interval boundaries of each pressure change segment are arranged in time to form a complete boundary point sequence. According to the boundary point sequence, the alignment data is divided into multiple independent pressure influence intervals, each interval corresponding to a pressure change subdomain, so that each pressure change mode has a clear interval range in the data structure. Finally, these subdomains are marked and output as contact pressure influence subdomain data, providing an accurate segmentation basis for subsequent optical data interference recognition by subdomain.

[0039] In a preferred embodiment of the present application, according to the contact pressure influence subdomain data, the original optical acquisition data in the alignment data is subjected to optical path disturbance positioning processing, and by identifying the light intensity slope change that appears synchronously with the contact pressure change, optical interference segment data is generated, including: According to the contact pressure influence subdomain data, the corresponding original optical acquisition data is extracted in each subdomain to form subdomain optical data; The light intensity difference value of adjacent sampling points of the subdomain optical data is calculated to generate light intensity change slope data; According to the light intensity change slope data and the pressure change trend in the corresponding subdomain, light intensity synchronous change point data is generated; According to the continuity of the light intensity synchronous change point data on the time axis, the continuous synchronous change points are combined into optical disturbance segment data; According to the slope change of the optical disturbance segment data at the boundary, the interference boundary point is identified, and interference segment boundary point data is generated; According to the optical disturbance segment data and the interference segment boundary point data, optical interference segment data is generated.

[0040] In the embodiment of the present application, by using the contact pressure influence domain to extract the optical acquisition data, the optical waveform and the pressure change interval correspondence relationship can be maintained, and cross-pressure state mixed analysis can be avoided. By calculating the light intensity difference value of the domain optical data, the change trend of the original waveform can be converted into the slope data, and the sudden increase and sudden decrease trend in the light intensity change can be more easily identified. By synchronously comparing the light intensity change slope and the pressure change trend in the corresponding domain, the optical change points related to the pressure change can be screened out, thereby excluding the real fluctuations caused by physiological factors and improving the accuracy of interference identification. By judging the time continuity of the synchronous change points, multiple single-point abnormalities can be merged into an optical disturbance segment with complete interference characteristics, so that the abnormal segment has a continuous structure on the time axis. Further, according to the slope change of the disturbance segment at the boundary, the interference segment boundary points can be generated, and the start and end positions of the interference can be accurately marked. The final optical interference segment data enables the subsequent compensation step to directly act on the abnormal segment, reducing the impact on the normal waveform.

[0041] In a preferred embodiment of the present application, according to the contact pressure influence domain data, the corresponding original optical acquisition data is extracted in each domain to form domain optical data, which specifically includes: Each domain in the contact pressure influence domain data is read according to its recorded start time point and end time point, and compared with the optical sampling time recorded in the alignment data to determine all sampling points in the optical data that match the time range of the domain. These sampling points are stored in time sequence to form the domain optical data corresponding to the domain. The above extraction process is repeated for all pressure change domains, so that the original optical acquisition data is split into multiple domain optical sequences according to the domain structure, providing an independent basis for subsequent interference identification in each domain.

[0042] In a preferred embodiment of the present application, the light intensity difference value of adjacent sampling points of the domain optical data is calculated to generate light intensity change slope data, which specifically includes: The light intensity values of each sampling point in the domain optical data are read in sequence, and the numerical difference between the adjacent next sampling point is compared, and the light intensity value of the latter sampling point is subtracted from the light intensity value of the former sampling point to obtain the change amount of the light intensity in the time interval. The change amount is recorded according to the sampling time sequence of the optical data, so that each change amount corresponds to a clear time position, thereby forming the light intensity change slope data. Through the expression of the light intensity difference between adjacent sampling points, the fast-changing segment and the slow-changing segment of the original waveform in the time dimension can be effectively distinguished, providing a basis feature for subsequent optical interference judgment.

[0043] In a preferred embodiment of the present application, the light intensity change slope data and the pressure change trend in the corresponding domain are synchronously compared to generate light intensity synchronous change point data, which specifically includes: The pressure change trend of the sub-domain is read from the contact pressure influence sub-domain data, and the trend is recorded in time sequence. Then the trend is compared with the light intensity change slope data on the same time axis to determine whether the direction of light intensity change is consistent with the direction of pressure change, and whether the magnitude of light intensity change is positively or negatively correlated with the amplitude of pressure change. When the light intensity change slope is synchronized with the pressure change trend in terms of change trend, change direction or change amplitude, the corresponding sampling position is recorded as a light intensity synchronous change point. Through this comparison process, the optical change points related to pressure changes in the sub-domain can be screened out, providing a reliable basis for subsequent interference segment identification.

[0044] In a preferred embodiment of the present application, according to the continuity of the light intensity synchronous change point data on the time axis, the continuous synchronous change points are combined into optical disturbance segment data, specifically including: All light intensity synchronous change points are sorted in order of sampling time, and the sampling time interval between adjacent two synchronous change points is determined point by point. When the interval between adjacent two points is less than or equal to a preset continuity threshold, the two points are considered as continuous parts in the same interference trend. With the passage of time, all synchronous change points that meet the continuity condition are combined into an optical disturbance segment. When the time interval of two synchronous change points exceeds the set threshold, the current disturbance segment is terminated, and a new disturbance segment is recorded from the next point. The generated optical disturbance segment data can accurately represent the interference segment affected by synchronous pressure.

[0045] In a preferred embodiment of the present application, according to the slope change of the optical disturbance segment data at the boundary, the interference boundary points are identified, and the interference segment boundary point data is generated, specifically including: The light intensity change slope between the first sampling point and the second sampling point in each optical disturbance segment is read, and it is determined whether the change trend deviates from the overall change trend inside the disturbance segment. When the deviation exceeds the set boundary deviation threshold, the position is marked as the starting boundary point of the disturbance segment. Similarly, the light intensity change slope between the last sampling point of the disturbance segment and its previous sampling point is read, and when its change trend is inconsistent with the overall trend inside the disturbance segment, the position is marked as the end boundary point of the disturbance segment. Through this judgment method, the accurate start and end positions of the interference segment can be determined, and the range of the interference is accurately expressed.

[0046] In a preferred embodiment of the present application, the optical disturbance segment data is generated according to the optical disturbance segment data and the interference segment boundary point data, specifically including: The start point and the end point in the interference segment boundary point data are taken as the boundary range of the interference segment, and all the sampling points in the boundary range are extracted from the optical disturbance segment data, so that the sampling values in the interference segment are arranged in time sequence. The extracted sequence is taken as the core content of the interference segment, and the start point, the end point and the corresponding optical value of the interference segment are taken to form complete optical interference segment data. The optical interference segment data can completely describe the position, duration and change form of the interference, and provide accurate input for subsequent optical compensation construction.

[0047] In a preferred embodiment of the present application, according to the optical interference segment data, optical compensation construction processing is performed on the corresponding optical interference segment, the change direction, the change amplitude of the interference segment and the pressure change trend of the domain to which the interference segment belongs are used to generate domain compensation optical data, including: Change direction data of each interference segment is determined according to the optical interference segment data, and compensation trend data opposite to the direction is generated according to the change direction data; Target compensation amount data is generated according to the change amplitude data of the interference segment in the optical interference segment data, and is uniformly distributed according to time length or proportionally distributed according to the pressure change trend; Compensation transition zone data is constructed at the start and end positions of the interference segment according to the target compensation amount data, so that the compensation amount gradually increases or decreases from zero in the compensation transition zone, and compensation transition data is generated; Compensation processing is performed on the optical interference segment according to the compensation trend data, the target compensation amount data and the compensation transition data, and compensated interference segment data is generated; Domain compensation optical data is generated by splicing the compensated interference segment data and the optical acquisition data not affected by the interference in time sequence.

[0048] In the embodiment of the present application, the main change trend of the interference waveform can be determined by analyzing the change direction in the optical interference segment data, so that the subsequent compensation direction is consistent with the interference in reverse, and the basis for restoring the original waveform form is provided. The compensation amount reflecting the interference degree and the pressure change law can be generated by processing the change amplitude of the interference segment and the pressure change trend in the domain, so that the compensation operation has both quantitative basis and trend consistency. The compensation transition zone is constructed, so that the compensation amount gradually changes at the start and end positions of the interference segment, and the waveform mutation caused by direct compensation is avoided. Further compensation processing is performed on the interference segment, and the change mode of the compensation trend, the compensation amount and the transition zone is comprehensively applied to the original waveform, so that the interference segment is adjusted point by point in time, and the compensation result closer to the real physiological waveform is formed. By splicing the compensated interference segment and the optical acquisition data not affected by the interference, domain compensation optical data maintaining continuity is obtained, and a relatively smooth basic signal is provided for subsequent cross-domain splicing.

[0049] In a preferred embodiment of the present application, the change direction data of each interference segment is determined according to the optical interference segment data, and the compensation trend data opposite to the direction thereof is generated according to the change direction data, specifically comprising: All sampling points of the interference segment are read from the optical interference segment data in sequence, and the light intensity values of the sampling points are arranged in time sequence to form an interference waveform sequence. For the light intensity change between each adjacent sampling point, it is judged whether the change amount is positive or negative, so as to determine the change direction of the segment at the time position. Then the change directions of all positions in the interference segment are counted and analyzed to judge whether the interference segment presents an upward change trend, a downward change trend or a compound trend of first rising and then falling. According to the obtained direction trend, the compensation trend is designed to be a trend opposite to the interference trend, for example, when the interference segment presents an overall upward trend, the compensation trend is made to present a downward trend; when the interference presents a downward trend, the compensation trend is made to present an upward trend. The finally generated compensation trend data can be used as the direction reference for subsequent compensation amount calculation and compensation operation.

[0050] In a preferred embodiment of the present application, the time sequence splicing is performed on the compensated interference segment data and the optical acquisition data not affected by the interference to generate the domain-compensated optical data, specifically comprising: First, the compensated interference segment data is sorted according to the original acquisition time, and the order thereof on the time axis is kept unchanged. Then, the optical acquisition data not affected by the interference is read from the aligned data, and is compared with the compensated interference segment according to the sampling time sequence to identify the time interval corresponding to the compensation segment. In the interval where the compensation segment is located, the original interference sampling points are replaced by the compensated sampling points, so that the interference segment is completely covered by the waveform after compensation; outside the compensation segment, the original normal waveform is kept unchanged. Then, the spliced data is subjected to a whole coherence check to ensure that there is no fault at the boundary position between the compensation segment and the normal segment, and the light intensity change at the boundary is kept smooth through necessary fine adjustment. The finally obtained domain-compensated optical data can be used as the basis for subsequent cross-domain splicing, so that the whole data maintains the continuity and structural consistency of the time sequence after the compensation processing.

[0051] In a preferred embodiment of the present application, the target compensation amount data is generated according to the change amplitude data of the interference segment in the optical interference segment data, and is distributed according to the time length or the pressure change trend, comprising: The change amplitude data of the interference segment is extracted from the optical interference segment data, and the interference segment time interval data is generated according to the start time and the end time of the interference segment; According to the interference fragment time interval data, the change amplitude data is divided into multiple compensation sub-regions in time sequence, so that each compensation sub-region corresponds to a time distribution fragment of the change amplitude data, and compensation sub-region time period data is generated; According to the compensation sub-region time period data, the change amplitude data is uniformly distributed and mapped to each compensation sub-region according to time, and uniform compensation amount data is generated; According to the pressure change trend of the contact pressure influence sub-domain to which the interference fragment belongs, the change relationship of the change amplitude data in each compensation sub-region is distributed, and trend compensation amount data is generated; According to the adaptability condition of the uniform compensation amount data and the trend compensation amount data, the compensation amount distribution mode that is more suitable for the waveform recovery characteristics of the interference fragment is selected from the two, and target compensation amount data is generated.

[0052] In the embodiment of the application, by extracting the change amplitude data in the optical interference fragment and combining the start and end times of the interference fragment to generate time interval data, the range of the interference fragment on the time axis can be quantitatively expressed, and subsequent processing can be conveniently split in time sequence. The time interval is divided into multiple compensation sub-regions, so that the change amplitude can be distributed in a finer granularity, avoiding the problem of insufficient or excessive local compensation caused by using a single compensation amount for the entire fragment. By mapping the change amplitude according to the time relationship to generate uniform compensation amount, the compensation demand of the interference when the interference is uniformly distributed in the fragment can be met. By distributing the change amplitude according to the pressure change trend, it is more adaptive in the case where the interference and the pressure trend have a strong correlation, so that the compensation amount can reflect the direction and strength of the interference change. Combined with the adaptability selection of the uniform method and the trend method, the accuracy of the compensation amount generation can be improved, and the subsequent compensation processing is more in line with the actual change structure of the interference fragment, providing a more reasonable compensation amount basis for waveform recovery.

[0053] In a preferred embodiment of the application, the change amplitude data of the interference fragment is extracted from the optical interference fragment data, and the interference fragment time interval data is generated according to the start time and end time of the interference fragment, specifically including: The complete sampling sequence of the interference fragment is obtained from the optical interference fragment data, and the light intensity values of each sampling point are recorded in time sequence. Then the light intensity difference values of adjacent sampling points are read to determine the light intensity change amplitude at each time within the interference fragment, and the above change amplitude is recorded in sequence as change amplitude data. The sampling times of the first sampling point and the last sampling point of the interference fragment are further read, and the two time points are recorded as the start time and the end time of the interference fragment. According to the two time points, the time interval corresponding to the interference fragment is generated, so that the range of the interference fragment in the time dimension is clearly defined, providing a time reference for the calculation and distribution of the subsequent compensation amount.

[0054] In a preferred embodiment of the present application, according to the time interval data of the interference segment, the change amplitude data is divided into a plurality of compensation sub-regions in time sequence, so that each compensation sub-region corresponds to a time distribution segment of the change amplitude data, and compensation sub-region time period data is generated, specifically including: First, according to the time interval length of the interference segment, the interval is divided into a plurality of continuous and equal-length sub-regions, so that each sub-region occupies the same length on the time axis. Then, the change amplitude data is corresponded to each sub-region in the sampling time sequence, and the change amplitude data within a certain time period is divided into the corresponding compensation sub-region. The internal data of each compensation sub-region is recorded, and the time range, start and end sampling points and internal data list of the compensation sub-region are established, so that each compensation sub-region has a complete structure in time dimension and data distribution dimension.

[0055] In a preferred embodiment of the present application, according to the compensation sub-region time period data, the change amplitude data is uniformly distributed and mapped to each compensation sub-region, and uniform compensation amount data is generated, specifically including: According to the time length of all compensation sub-regions, the compensation distribution proportion of each compensation sub-region in the entire interference segment is determined. Then, according to the total amount of change amplitude data, the total amount is evenly distributed to each compensation sub-region according to the time distribution proportion, so that each compensation sub-region obtains an initial compensation value proportional to its time length. Then, for the sampling points in each sub-region, the initial compensation value is averaged again according to the number of sampling points, so that each sampling point obtains the same compensation amount. The finally formed uniform compensation amount data is a set of compensation amount sequences distributed continuously in time, which provides a reference for the comparison of subsequent trend compensation.

[0056] In a preferred embodiment of the present application, according to the pressure change trend of the contact pressure influence sub-domain to which the interference segment belongs, the change amplitude data is distributed according to the change relationship of each compensation sub-region, and trend compensation amount data is generated, specifically including: The pressure change trend of the interference segment belonging to the sub-domain is read from the contact pressure influence sub-domain data, and the trend is unfolded into a trend change sequence in time sequence. According to the trend change sequence, the change of the pressure value in each time sub-region is judged, for example, the sub-region with faster pressure rise is considered to be more significantly affected by the interference. Then, the change amplitude data is distributed according to the pressure change degree of different sub-regions, so that the area with larger pressure change gets more compensation amount, and the area with smaller pressure change gets less compensation amount. For the sampling points in each sub-region, the compensation amount is gradually transitioned from the start position to the end position according to the time position of the sampling points in the sub-region, so that the trend compensation amount data presents a distribution structure consistent with the pressure change trend.

[0057] In a preferred embodiment of the present application, according to the adaptability condition of the uniform compensation amount data and the trend compensation amount data, the compensation amount distribution mode that is more suitable for the waveform recovery characteristics of the interference segment is selected from the two, and the target compensation amount data is generated, which specifically includes: First, the uniform compensation amount data and the trend compensation amount data are respectively analyzed in structure, and it is checked whether the two distribution modes match the main change direction, change position and change intensity of the interference segment. For example, when the change of the interference segment is relatively uniform in the entire time interval, the uniform compensation mode is preferred; when the change of the interference segment has a clear synchronous relationship with the pressure trend, the trend compensation mode is preferred. Then the compensation amount sequence of the selected compensation mode is sorted in time sequence to form complete target compensation amount data, and it is ensured that the data is consistent with the time interval and change amplitude of the interference segment, so that it can be used as input data for the construction of the compensation transition zone and subsequent compensation processing.

[0058] In a preferred embodiment of the present application, the compensation transition zone data is constructed according to the target compensation amount data at the start and end positions of the interference segment, so that the compensation amount gradually increases or gradually decreases from zero in the compensation transition zone, and the compensation transition data is generated, including: The start point data and the end point data of the interference segment are determined according to the optical interference segment data, and the interference segment boundary point data is generated according to the start point data and the end point data; According to the interference segment boundary point data, the boundary compensation zone division processing is performed on the target compensation amount data, the boundary area is divided into a plurality of compensation transition sub-zones, and the compensation transition sub-zone data is generated; According to the compensation transition sub-zone data, the target compensation amount data is gradually processed according to the mode of gradually increasing from zero to the target compensation amount or gradually decreasing from the target compensation amount, and the compensation gradual change sequence data is generated; According to the compensation gradual change sequence data, the continuity adjustment processing between adjacent sampling points is performed, so that the compensation amount keeps smooth change point by point in the compensation transition zone, and the compensation gradual change adjustment data is generated; According to the compensation gradual change adjustment data, the complete compensation transition zone data is constructed.

[0059] In the embodiment of the present application, by determining the start and end positions from the optical interference segment, clear boundary range can be provided for subsequent construction of compensation transition zone, so that the compensation transformation has a position reference. According to the waveform variation between the boundary positions, the compensation transition sub-zone is divided, so that the transition zone has a structure, and different compensation processing can be performed according to the variation characteristics in different sub-zones. By generating a compensation gradual change sequence in a gradually increasing or gradually decreasing mode, it can be ensured that the compensation amount changes gently in the transition zone, avoiding sudden changes to the waveform. The gradual change sequence is continuously adjusted, so that the compensation difference of adjacent sampling points remains stable, which helps to form a more natural change curve and reduce the jump error. The finally constructed compensation transition zone data can establish a smooth connection structure between the interference segment boundary and the internal compensation amount of the segment, provide a smooth transition for subsequent overall compensation of the segment, and help to improve the continuity and integrity of the waveform after compensation processing.

[0060] In a preferred embodiment of the present application, the start point data and the end point data of the interference segment are determined according to the optical interference segment data, and the interference segment boundary point data is generated according to the start point data and the end point data, specifically including: The complete time sequence and the corresponding light intensity value are extracted from the optical interference segment data. The light intensity change of each sampling point is checked in sequence, and when the light intensity fluctuates obviously and the change direction is different from or suddenly changes from the previous period, it is marked as a potential start or end point of the interference. Specifically, the first sampling point where the fluctuation occurs is regarded as the start point of the interference segment, and the last sampling point where the fluctuation occurs is regarded as the end point of the interference segment. If the amplitude of the light intensity change in the interference segment reaches a set threshold value and the change trend changes obviously, these points are further confirmed as the boundary points of the interference segment. By this method, the start and end times of the interference segment are determined according to the variation characteristics of the light intensity value, and the boundary point data of the interference segment is generated for subsequent processing.

[0061] In a preferred embodiment of the present application, according to the interference segment boundary point data, the boundary compensation zone division processing is performed on the target compensation amount data, the boundary region is divided into a plurality of compensation transition sub-zones, and the compensation transition sub-zone data is generated, specifically including: According to the previously generated interference segment boundary point data, the starting point and the ending point of the interference segment are first determined, and then a boundary compensation zone is demarcated between the two positions. Within the compensation zone, it is divided into a plurality of continuous compensation transition sub-zones, and each compensation transition sub-zone corresponds to a time period. In order to ensure the smoothness of the compensation transition, the time length of each sub-zone is usually set according to the speed of the compensation amount change or the characteristics of the pressure change. For regions with rapid changes, the compensation transition sub-zone time is short, while for regions with slow changes, longer sub-zone time is used. The range, time length and change rule of the compensation amount of each compensation transition sub-zone are recorded as basic data in the compensation process, and compensation transition sub-zone data is generated.

[0062] In a preferred embodiment of the present application, according to the compensation transition sub-zone data, the target compensation amount data is gradually processed in a gradually increasing or gradually decreasing mode from zero to the target compensation amount to generate compensation gradual change sequence data, which specifically includes: First, the initial value of the target compensation amount is set to zero from the starting position of each compensation transition sub-zone, and the compensation amount is gradually increased according to the compensation amount increase trend in the zone until the compensation amount reaches the target value. Similarly, at the end position of the compensation transition zone, the compensation amount gradually decreases to zero, ensuring smooth transition of the compensation amount in the transition zone and avoiding sudden changes. According to the time span of each sub-zone, the compensation amount increase rate and the pressure change trend, a continuous compensation amount change sequence is generated to make the compensation amount gradually change in the entire zone, ensuring smooth and non-sudden changes in the compensation effect. In this way, the generated compensation gradual change sequence data can naturally reflect the correction process of the optical waveform, avoiding distortion or discontinuity in the compensation process.

[0063] In a preferred embodiment of the present application, according to the compensation gradual change sequence data, a continuity adjustment process between adjacent sampling points is performed to make the compensation amount gradually change smoothly in the compensation transition zone, and compensation gradual change adjustment data is generated, which specifically includes: The compensation gradual change sequence data is read, and the difference between each pair of adjacent sampling points in the sequence is checked to ensure that the compensation amount change of each two adjacent sampling points does not exceed a set threshold. If the difference is too large, the compensation amount of the region is adjusted by linear interpolation to make the change more smooth. During the adjustment process, if the change amplitude of the compensation amount is large, the length of the transition region is appropriately increased and the change speed is slowed down; if the change amplitude is small, the transition zone is appropriately shortened and the compensation amount change is accelerated. In this way, the adjusted compensation gradual change sequence can gradually increase or decrease the compensation amount in the time dimension, ensuring smooth transition in the compensation process, thereby generating the final compensation gradual change adjustment data.

[0064] In a preferred embodiment of the present application, according to the compensation gradual change adjustment data, complete compensation transition zone data is constructed, which specifically includes: The compensation amount values at each time point are extracted from the compensation gradual adjustment data, and these compensation amount values are corresponded to the time periods of the compensation transition zone according to the order of the time sequence. Through the change trend and time length of the compensation amount, the data structure of the entire compensation transition zone is constructed, and the change of the compensation amount in each time period is recorded. The finally formed compensation transition zone data will contain the complete compensation process from the start point to the end point of the compensation zone, ensuring the smooth change of the compensation amount in each time period and meeting the optical waveform correction requirements. The compensation transition zone data can be used by the subsequent compensation processing module as a smooth transition zone of the compensation waveform, ensuring that there is no unnatural jump or mutation after the waveform correction.

[0065] In a preferred embodiment of the present application, the compensation processing is performed on the optical interference segment according to the compensation trend data, the target compensation amount data and the compensation transition data to generate the compensated interference segment data, including: According to the compensation trend data, the direction correction processing is performed on the optical interference segment data, the compensation direction is adjusted to be opposite to the change direction of the interference segment, and the direction correction data is generated; According to the direction correction data, the compensation distribution processing is performed on each sampling point inside the interference segment by using the target compensation amount data, the compensation amount is distributed according to the time sequence of the sampling point position, and the internal compensation distribution data is generated; According to the internal compensation distribution data and the compensation transition data, the compensation gradual adjustment superposition processing is performed on the boundary of the interference segment, so that the compensation amount remains continuous transition between the internal part and the boundary part of the interference segment, and the compensation superposition structure data is generated; According to the compensation superposition structure data, the compensation fusion processing is performed on the whole interference segment, and the compensated interference segment data is generated by fusing the compensation superposition structure data in the time sequence with the original interference segment data.

[0066] In the embodiment of the present application, by using the compensation trend data to correct the direction of the optical interference segment, the compensation direction can be opposite to the interference direction, ensuring that the compensation can produce a counteracting effect rather than enhancing the interference. According to the target compensation amount data, the compensation amount of each sampling point in the interference segment is sequentially distributed, so that the compensation size of each sampling point has a clear basis, avoiding local waveform distortion caused by uneven compensation in the segment. In combination with the compensation transition zone data, gradient superposition processing is performed at the boundary position of the interference segment, so that the internal compensation zone and the boundary compensation zone are smoothly connected in intensity and change trend, reducing the possibility of boundary jump before and after compensation. By fusing the direction correction data, the internal compensation distribution data and the compensation superposition structure data, the overall compensation of the interference segment is performed, which can realize unified adjustment between different data structures, so that the compensated waveform can not only correct the deviation caused by the interference, but also maintain the stability of the internal form. The processed compensated interference segment can form a continuous signal structure with the undisturbed waveform, providing more reliable basic data for subsequent data splicing and diagnostic analysis.

[0067] In a preferred embodiment of the present application, according to the compensation trend data, the direction correction processing of the optical interference segment data is performed to adjust the compensation direction to be opposite to the change direction of the interference segment, to generate direction correction data, specifically including: The change information of the compensation direction is read from the compensation trend data, and the change trend of the interference segment (such as rising, falling, etc.) is identified. Then, according to the compensation trend data, it is determined that the compensation should be opposite to the direction of the original interference segment. For example, when the interference segment presents an upward trend, the compensation trend should be downward; when the interference segment presents a downward trend, the compensation trend should be upward. Then, read each sampling point in the optical interference segment data, and apply the opposite direction of the compensation amount to each sampling point, modify the value of the compensation amount, so that it meets the correction requirement of being opposite to the interference direction. In this way, the direction correction data is generated, which can correct the optical interference segment to be consistent with the direction of the compensation trend, ensuring that the corrected waveform after compensation meets the actual needs.

[0068] In a preferred embodiment of the present application, according to the direction correction data, the compensation distribution processing of each sampling point in the interference segment is performed by using the target compensation amount data, the compensation amount is distributed according to the time sequence of the sampling point position, and the internal compensation distribution data is generated, specifically including: According to the compensation direction of each time point in the direction correction data, the target compensation amount data is distributed to each sampling point in time sequence. First, the total amount of compensation amount data and the time range of distribution are confirmed. Then, according to the position and time distribution of the sampling points, the compensation amount data is distributed to each sampling point, so that the size of the compensation amount is proportional to the time position it is in. For example, when the sampling point is at the starting position of the time period, the compensation amount distributed is small, and when it is close to the end position, the compensation amount gradually increases (or gradually decreases, depending on the compensation trend). Finally, the compensation amount of each sampling point is recorded, so that the compensation data forms a continuous compensation distribution process within the interference segment, ensuring that the compensation effect is consistent with the change trend of the optical waveform.

[0069] In a preferred embodiment of the present application, according to the internal compensation distribution data and the compensation transition data, the compensation fade-in superposition processing is performed at the boundary of the interference segment, so that the compensation amount remains continuous transition between the internal part and the boundary part of the interference segment, and the compensation superposition structure data is generated, which specifically includes: First, read the internal compensation distribution data and obtain the distribution of the compensation amount at each time point, and pay special attention to the change process of the compensation amount. According to the compensation transition data, determine the transition range of the compensation amount at the boundary of the interference segment. At the beginning and end of the compensation segment, gradually adjust the compensation amount so that it gradually increases or decreases, to ensure that the compensation amount remains smooth transition at the boundary and does not appear suddenly. According to this fade-in mode, gradually apply the distribution of the compensation amount, expand from the inside of the segment to the boundary, and generate a compensation superposition effect at the boundary part, so that the compensation amount at the boundary naturally connects with the compensation amount in the internal part, without producing faults or excessive adjustment. Finally, the generated compensation superposition structure data can smoothly connect the compensation amount at the beginning and end on the time axis, ensuring that the compensation effect is continuous and stable within the entire interference segment.

[0070] In a preferred embodiment of the present application, according to the compensation superposition structure data, the compensation fusion processing is performed on the entire interference segment, and by fusing the compensation superposition structure data with the original interference segment data in time sequence, the compensated interference segment data is generated, which specifically includes: First, the compensation superposition structure data is time-aligned with the original interference segment data to ensure that the two data sequences have the same time reference. Then, according to the compensation amount generated in the compensation superposition structure data, the compensation amount is added to each corresponding sampling point in the original interference segment data point by point. For the sampling points in the compensation section, the original optical signal is replaced with the compensation amount; for the unaffected part, the original data remains unchanged. In particular, attention is paid to the superposition manner of the compensation amount to ensure that the introduction of the compensation amount does not cause excessive correction or distortion of the data. Finally, by smoothing the adjusted compensation interference segment data, the compensated data forms a continuous and smooth waveform on the time axis, consistent with the original data, and the light intensity change is corrected in the interference area, generating the final compensated interference segment data.

[0071] Embodiments of the present application also provide a non-invasive diagnostic data acquisition system, which comprises: a data alignment module for acquiring original optical acquisition data and original contact pressure data and performing time alignment processing to form aligned data; a contact pressure change identification module for performing contact pressure change identification processing on the original contact pressure data in the aligned data to generate contact pressure influence sub-domain data through the contact pressure change amplitude and the contact pressure change rate; an optical path disturbance positioning module for performing optical path disturbance positioning processing on the original optical acquisition data in the aligned data according to the contact pressure influence sub-domain data to generate optical interference segment data by identifying the light intensity slope change that appears synchronously with the contact pressure change; an optical compensation construction module for performing optical compensation construction processing on the corresponding optical interference segment according to the optical interference segment data to generate sub-domain compensation optical data through the change direction, change amplitude of the interference segment and the pressure change trend of the contact pressure influence sub-domain to which the interference segment belongs; a sub-domain splicing module for performing sub-domain splicing processing between each contact pressure influence sub-domain according to the sub-domain compensation optical data to generate continuous correction optical data through the light intensity difference, phase difference and waveform period difference between adjacent sub-domains; a target data extraction module for performing target data extraction processing according to the continuous correction optical data to generate non-invasive diagnostic data.

[0072] It should be noted that the system corresponds to the above method, and all implementation manners in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0073] The embodiment of the application also provides a computing device, comprising a processor and a memory storing a computer program, the computer program being executed by the processor to perform the method described above. All implementation manners in the method embodiment are applicable to this embodiment and can achieve the same technical effects.

[0074] The embodiment of the application also provides a computer readable storage medium storing instructions, which, when executed on a computer, cause the computer to perform the method described above. All implementation manners in the method embodiment are applicable to this embodiment and can achieve the same technical effects.

[0075] The present case is applied to a wearable non-invasive diagnostic device, aiming at the heart rate and blood oxygen saturation collection requirements of users in a light exercise scene (such as walking, slight hand movement), through the "contact pressure-optical signal linkage correction" technology of the application to eliminate motion interference, and then matching the corrected diagnostic data with the cloud standard physiological database, outputting health assessment suggestions, solving the problems of data distortion and low diagnostic reference value of traditional devices during exercise.

[0076] Case core device and data source: 1. Collection device: an intelligent bracelet integrating a near-infrared optical sensor (emission wavelength 660nm / 940nm, sampling frequency 50Hz) and a thin film pressure sensor (sampling frequency 50Hz, measurement range 0-5N), worn at the base of the user's index finger.

[0077] 2. Cloud standard database: storing physiological data of 100,000+ healthy people and common cardiovascular disease populations, including: normal heart rate range (resting 60-100 times / min, light exercise 90-130 times / min); normal blood oxygen saturation range (95%-100%); typical pulse waveform template (grouped by age: 20-30 years old, 31-50 years old, 51 years old and above, containing waveform peak / valley ratio, rising slope, cycle stability and other characteristic parameters); data quality evaluation threshold after interference correction (such as waveform similarity ≥85% for effective matching).

[0078] 3. Test object: a 35-year-old male user, with daily light exercise (30 minutes of walking per day), no known cardiovascular disease, in a walking state (walking speed 4km / h) during testing, and the collection duration is 30 seconds.

[0079] Case implementation steps (combined with the technical solution of the application): Step 1: Raw data collection and time alignment The optical sensor collects the finger skin reflected light intensity signal to generate raw optical collection data (30 seconds x 50Hz = 1500 sampling points, containing light intensity fluctuation signal); The pressure sensor collects the contact pressure data between the bracelet and the skin, and generates original contact pressure data (1500 sampling points, the pressure fluctuates between 0.8N and 2.5N due to walking); According to the method of the present application, the pressure data is resampled (the sampling interval is adjusted to 20ms) based on the sampling clock of the optical sensor, and time-aligned data (1 pressure sampling point corresponds to each optical sampling point) is formed.

[0080] Step 2: Contact pressure sub-domain and optical interference positioning The difference between adjacent sampling points of the aligned pressure data is calculated to generate pressure change amplitude data (the maximum change amplitude is 0.6N); the pressure change rate data (the maximum rate is 0.3N / ms) is calculated; The preset pressure change amplitude threshold is 0.2N, and the rate threshold is 0.1N / ms, and 3 continuous candidate pressure change segments (corresponding to the pressure fluctuation caused by hand swing when the user walks, each segment lasts for 1.2-1.8 seconds) are screened out; After splitting according to time continuity, 3 contact pressure influence sub-domains are generated (sub-domain 1: 0-1.5 seconds, sub-domain 2: 8.2-9.8 seconds, sub-domain 3: 22.5-24.0 seconds); In each sub-domain, the optical data is extracted, the light intensity change slope is calculated, and the synchronization with the pressure change trend (rise / fall) is compared to locate 3 optical interference segments (completely synchronized with the pressure change segment, the light intensity slope mutation amplitude is ≥0.05 lux / ms).

[0081] Step 3: Optical compensation and sub-domain splicing For each interference segment: 1. Determine the interference direction (the light intensity in sub-domain 1 shows a downward trend, and the compensation trend is set to be upward); 2. Extract the interference amplitude (the maximum light intensity decrease in sub-domain 1 is 0.8lux), and distribute the compensation amount according to the pressure change trend (first rise and then fall) to generate target compensation amount data; 3. Build a 5-sample-point (0.1s) compensation transition zone at the start and end positions of the interference segment, and the compensation amount gradually changes from 0 to the target value; Fuse the compensation trend, compensation amount and transition zone data to compensate the interference segment, and then splice it with the optical data that is not disturbed to generate sub-domain compensation optical data; Calculate the light intensity difference (the maximum difference is 0.1lux), the phase difference (≤5ms), and the period difference (≤0.02s) between adjacent sub-domains, realize sub-domain splicing through local waveform translation, and generate continuous corrected optical data (no obvious boundary mutation, smooth waveform).

[0082] Step 4: Target diagnostic data extraction Extract the pulse waveform features from the continuous corrected optical data: Heart rate: Count the number of pulse cycles in 30 seconds, calculate 105 beats per minute; Oxygen saturation: Based on the ratio of light intensity at 660 nm and 940 nm, combined with the corrected peak ratio, calculate 97%; Pulse waveform feature parameters: Peak-to-valley ratio = 3.2, rising slope = 0.08 lux / ms, cycle stability coefficient of variation = 3.5%; The above parameter combination is the final non-invasive diagnostic data, uploaded to the cloud.

[0083] Step 5: Cloud data matching and health advice generation 1. Matching dimensions and algorithms: Dimension 1: Heart rate interval matching (user 105 beats per minute, cloud normal interval for 31-50 year-old men during light exercise 90-130 beats per minute); Dimension 2: Oxygen saturation matching (97%, cloud normal interval 95%-100%); Dimension 3: Pulse waveform similarity matching (using dynamic time warping (DTW) algorithm, compare the user's corrected pulse waveform with the standard waveform template of 31-50 year-old men during light exercise, similarity = 92%); Matching degree calculation: Weighted score (heart rate 30% + oxygen saturation 30% + waveform similarity 40%), final matching degree = 93%.

[0084] 2. Matching degree classification and advice: Classification standard: Matching degree ≥ 85% is "high matching", 60%-84% is "medium matching", and < 60% is "low matching"; This matching result: High matching (93%); Health advice: "Your current heart rate (105 beats per minute), oxygen saturation (97%) and pulse waveform during light exercise meet the standard of 31-50 year-old healthy men, with no obvious abnormalities. It is recommended to maintain the current exercise intensity and continue monitoring; if subsequent exercise heart rate exceeds 130 beats per minute or oxygen saturation is less than 95%, please stop exercising and rest, and if necessary, go to the hospital for cardiovascular function examination." Case technical effect verification: 1. Interference correction effect: Before correction, optical interference caused by exercise resulted in a heart rate calculation error of ±15 beats per minute, and an oxygen saturation fluctuation of ±3%; After correction, the heart rate error was ≤±2 beats per minute, and the oxygen saturation fluctuation was ≤±0.5%, with an improvement of more than 85% in data stability; 2. Matching accuracy: After correction, the matching degree with the cloud standard data improved from 62% before correction to 93%, effectively avoiding "misjudgment caused by interference data"; 3. Practicality: adapt to light exercise scene, solve the pain point of traditional wearable device "data unavailable during exercise", provide real-time and reliable health evaluation reference for users.

[0085] The above is the preferred embodiment of the present application, it should be pointed out that, for the ordinary skilled in the art, without departing from the principles of the present application described, can also be made several improvements and refinements, these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A method for acquiring non-invasive diagnostic data, characterized in that, The method includes: The raw optical acquisition data and raw contact pressure data are acquired and time-aligned to form aligned data. Based on the original contact pressure data in the alignment data, perform contact pressure change identification processing, and generate contact pressure influence domain data by the contact pressure change amplitude and contact pressure change rate. Based on the domain data affected by contact pressure, optical path disturbance localization processing is performed on the original optical acquisition data in the alignment data. By identifying the changes in light intensity slope that occur synchronously with the changes in contact pressure, optical interference fragment data is generated. Based on the optical interference fragment data, optical compensation construction processing is performed on the corresponding optical interference fragments. By analyzing the change direction and amplitude of the interference fragments and the pressure change trend of the domain affected by the contact pressure of the corresponding fragments, domain compensation optical data is generated. Based on the domain-compensated optical data, domain-segmentation processing is performed between the domains affected by each contact pressure. Continuous correction optical data is generated by using the light intensity difference, phase difference, and waveform period difference between adjacent domains. Based on continuously calibrated optical data, target data extraction and processing are performed to generate non-invasive diagnostic data.

2. The method for acquiring non-invasive diagnostic data according to claim 1, characterized in that, Based on the original contact pressure data in the aligned data, contact pressure change identification processing is performed. Contact pressure influence domain data is generated by analyzing the magnitude and rate of contact pressure change, including: The difference between adjacent sampling points in the original contact pressure data in the alignment data is calculated to generate pressure change amplitude data. Based on the pressure change amplitude data, the rate of change is calculated according to the change in adjacent time intervals to generate pressure change rate data; Based on the pressure change amplitude data and pressure change rate data, compare them with the preset pressure change amplitude threshold and preset pressure change rate threshold respectively, and record the continuous time segment that meets the matching condition of any preset threshold in the comparison results as candidate pressure change segment data. Based on the temporal continuity of the candidate pressure change segment data, candidate pressure change segments with time intervals exceeding a preset interval threshold are split into multiple pressure change segment data. Based on the start and end points of the pressure change segment data, corresponding segment boundary point data are generated, and contact pressure influence domain data are generated based on the segment boundary point data.

3. The method for acquiring non-invasive diagnostic data according to claim 1, characterized in that, Based on the contact pressure-affected domain data, optical path disturbance localization processing is performed on the raw optical acquisition data in the aligned data. By identifying changes in light intensity slope that occur synchronously with changes in contact pressure, optical interference fragment data is generated, including: Based on the influence of contact pressure on the domain data, the corresponding original optical acquisition data is extracted in each domain to form domain optical data. The light intensity difference between adjacent sampling points of the domain optical data is calculated to generate light intensity change slope data. By comparing the slope data of light intensity change with the pressure change trend in the corresponding domain, synchronous data of light intensity change points are generated. Based on the continuity of the synchronous change point data of light intensity on the time axis, the continuous synchronous change points are combined into optical disturbance segment data. Based on the slope change of the optical disturbance segment data at the boundary, the interference boundary points are identified, and interference segment boundary point data is generated. Optical interference segment data is generated based on the optical disturbance segment data and the boundary point data of the interference segment.

4. The method for acquiring non-invasive diagnostic data according to claim 1, characterized in that, Based on the optical interference fragment data, optical compensation construction processing is performed on the corresponding optical interference fragments. By analyzing the direction and magnitude of the interference fragment changes and the pressure change trend of the corresponding contact pressure, domain-specific compensated optical data is generated, including: The change direction data of each interference segment is determined based on the optical interference segment data, and compensation trend data opposite to its direction is generated based on the change direction data. Based on the variation amplitude data of the interference segments in the optical interference segment data, the target compensation amount data is generated by uniformly distributing the data according to the time length or proportionally distributing it according to the pressure change trend. Based on the target compensation amount data, compensation transition zone data is constructed at the start and end positions of the interference segment. The compensation amount is gradually increased or decreased from zero within the compensation transition zone to generate compensation transition data. Compensation processing is performed on the optical interference segments based on compensation trend data, target compensation amount data, and compensation transition data to generate compensated interference segment data. The compensated interference fragment data is combined with the undisturbed optical acquisition data in a time series to generate domain-compensated optical data.

5. The method for acquiring non-invasive diagnostic data according to claim 4, characterized in that, Based on the variation amplitude data of the interference segments in the optical interference segment data, the target compensation amount data is generated by uniformly distributing the data according to the time length or proportionally distributing it according to the pressure change trend, including: Extract the variation amplitude data of the interference segments from the optical interference segment data, and generate the time interval data of the interference segments based on the start and end times of the interference segments; Based on the time interval data of the interference segment, the change amplitude data is divided into multiple compensation sub-regions in chronological order, so that each compensation sub-region corresponds to a time distribution segment of the change amplitude data, thus generating compensation sub-region time period data; Based on the time period data of the compensation sub-region, the change amplitude data is evenly distributed over time and mapped to each compensation sub-region to generate uniformly distributed compensation amount data; Based on the pressure change trend of the contact pressure influence domain to which the interference segment belongs, the change amplitude data is distributed according to the relationship of the pressure change trend in each compensation sub-region to generate trend compensation amount data. Based on the compatibility conditions between uniformly distributed compensation data and trend compensation data, a compensation distribution method that better matches the waveform recovery characteristics of the interference segment is selected from the two to generate target compensation data.

6. The method for acquiring non-invasive diagnostic data according to claim 4, characterized in that, Based on the target compensation amount data, compensation transition zone data is constructed at the start and end positions of the interference segment. The compensation amount gradually increases or decreases from zero within the compensation transition zone, generating compensation transition data, including: The starting point and ending point data of the interference segment are determined based on the optical interference segment data, and the boundary point data of the interference segment are generated based on the starting point and ending point data. Based on the boundary point data of the interference segment, the boundary compensation area is divided into several compensation transition sub-regions, and compensation transition sub-region data is generated. Based on the compensation transition sub-region data, the target compensation amount data is gradually processed in a pattern of gradually increasing from zero to the target compensation amount or gradually decreasing from the target compensation amount, to generate compensation transition sequence data. Based on the compensation gradient sequence data, continuous adjustment processing is performed between adjacent sampling points to ensure that the compensation amount changes smoothly point by point within the compensation transition zone, thereby generating compensation gradient adjustment data; Based on the compensation gradual adjustment data, construct complete compensation transition zone data.

7. The method for acquiring non-invasive diagnostic data according to claim 4, characterized in that, Based on the compensation trend data, target compensation amount data, and compensation transition data, compensation processing is performed on the optical interference segments to generate compensated interference segment data, including: Based on the compensation trend data, the optical interference segment data is subjected to orientation correction processing, and the compensation direction is adjusted to be opposite to the change direction of the interference segment to generate orientation correction data. Based on the orientation correction data, the target compensation data is used to perform compensation allocation processing on each sampling point inside the interference segment. The compensation amount is allocated according to the time sequence of the sampling point position to generate internal compensation allocation data. Based on the internal compensation allocation data and compensation transition data, a compensation gradient superposition process is performed at the boundary of the interference segment to ensure that the compensation amount maintains a continuous transition between the internal part and the boundary part of the interference segment, thereby generating compensation superposition structure data. Compensation and fusion processing is performed on the entire interference segment based on the compensated overlay structure data. By fusing the compensated overlay structure data with the original interference segment data in the time series, compensated interference segment data is generated.

8. A non-invasive diagnostic data acquisition system, characterized in that, The system, used in any one of claims 1 to 7, comprises: The data alignment module is used to acquire raw optical acquisition data and raw contact pressure data, and perform time alignment processing to form aligned data; The contact pressure change identification module is used to perform contact pressure change identification processing based on the original contact pressure data in the alignment data, and generate contact pressure influence domain data through the contact pressure change amplitude and contact pressure change rate. The optical path disturbance positioning module is used to perform optical path disturbance positioning processing on the original optical acquisition data in the aligned data based on the contact pressure influence domain data. By identifying the light intensity slope change that occurs synchronously with the contact pressure change, it generates optical interference fragment data. The optical compensation construction module is used to perform optical compensation construction processing on the corresponding optical interference segments based on the optical interference segment data. It generates regional compensation optical data by considering the change direction and amplitude of the interference segments and the pressure change trend of the region affected by the contact pressure of the interference segments. The domain stitching module is used to perform domain stitching processing between each contact pressure-affected domain based on the domain compensation optical data. It generates continuous correction optical data by using the light intensity difference, phase difference, and waveform period difference between adjacent domains. The target data extraction module is used to perform target data extraction processing based on continuously corrected optical data to generate non-invasive diagnostic data.

9. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images