Gravity separation system based on platelet concentration

Abstract

The invention relates to the technical field of blood separation, in particular to a platelet concentration-based gravity separation system, which comprises a tension disturbance extraction module, a viscosity response positioning module, a sedimentation retarding judgment module, a platelet export judgment module and a concentration layer interface output module. According to the method, real-time extraction of tension change frequency and mutation amplitude is introduced, and a normalization processing mode of disturbance polymerization indexes is combined, so that a mechanical unstable area under tiny disturbance in the blood standing sedimentation process is effectively described, and the sensitive recognition capability on potential layered behavior changes is enhanced. Through continuous tracking of a viscosity behavior curve in a disturbance time period, an accurate time anchor point is provided for judgment of a subsequent liquid column motion state. On the basis of microscopic changes of the liquid column height and the rate, through combined judgment of the acceleration change trend and the attenuation amplitude, the settlement state is distinguished no longer depending on a macroscopic parameter threshold value, but is dynamically regulated and controlled through the change trend and a fluctuation sequence.

Description

Technical Field

[0001] The present invention relates to the technical field of blood separation, and in particular to a gravity separation system based on platelet concentration. Background Art

[0002] The technical field of blood separation includes related devices and processes for separating different components in blood by physical or chemical methods. The core content of this technical field is to effectively separate and extract blood components such as red blood cells, white blood cells, platelets, and plasma through means such as centrifugation, filtration, and gravitational sedimentation.

[0003] Among them, a gravity separation system based on platelet concentration refers to a device and method for extracting and concentrating platelets from whole blood by gravitational sedimentation. It includes subjecting a blood sample to static sedimentation treatment, controlling the sedimentation time and temperature to promote the enrichment of platelets in plasma, achieving natural stratification of platelets from red blood cells and white blood cells through density differences, and using a specific channel structure to export the separated platelet layer.

[0004] Traditional platelet separation methods based on gravitational sedimentation only rely on static placement and time control to drive the natural stratification process, lacking the ability to respond to dynamic change signals during the separation process. During the sedimentation process, there are obvious fluctuations in the interface change speeds of various blood components. Simply judging the export timing based on the static time is prone to premature or delayed misjudgments, thereby affecting the separation purity and concentration efficiency. In existing methods, there is a lack of quantification means for disturbance phenomena, and it is difficult to accurately capture and analyze the liquid surface tension fluctuations caused by environmental changes or minor vibrations, resulting in a blurred platelet layer recognition area, increasing the operation dependence and separation risk. At the same time, there is a lack of microscopic modeling ability for liquid column acceleration and descent trend, and it is impossible to judge whether the sedimentation process enters a deceleration or stable section, often causing mis-export before the separation interface is clear, affecting the platelet concentration effect. Traditional methods have insufficient utilization of the bottom static pressure change signal and fail to form a combined judgment mechanism by combining mass sensing parameters, weakening the scientific nature of the export time point judgment. In the actual separation process, if the target layer does not reach a stable density structure, the separation interface judgment only relies on position parameters and is extremely prone to being disturbed and shifted, causing other cell components to be mixed into the extraction layer, reducing the functional retention rate and post-separation usability of platelets. The above factors together lead to insufficient stability, weak intelligent regulation ability, and limited guarantee ability for high-purity extraction in the existing technology when dealing with complex sedimentation environments. Summary of the Invention

[0005] The purpose of the present invention is to solve the deficiencies existing in the prior art and propose a gravity separation system based on platelet concentration.

[0006] To achieve the above purpose, the present invention adopts the following technical solution: A gravity separation system based on platelet concentration includes:

[0007] The tension perturbation extraction module obtains the tension sensing data at the channel liquid level during the static sedimentation process of blood, calculates the perturbation aggregation index of the tension change, selects the perturbation time period, and generates the blood tension perturbation section;

[0008] The viscosity response positioning module obtains the viscosity sensing data of the blood sample during the same perturbation time period in the static sedimentation process based on the blood tension perturbation section, identifies the inflection point section from it, and generates a set of viscosity change behavior inflection point sections;

[0009] The sedimentation deceleration judgment module extracts the liquid column displacement rate sequence, duration, and fluctuation range within the target time period according to the time corresponding to the inflection point section in the set of viscosity change behavior inflection point sections, and generates platelet sedimentation trend data;

[0010] The platelet export discrimination module obtains the cumulative mass sensing data and bottom static pressure change of the blood sample based on the time period information of the platelet sedimentation trend data, judges whether it enters the convergence control interval of the sedimentation height, and generates a platelet export boundary set;

[0011] The concentrated layer interface output module judges whether the pressure change is in the pressure stable interval according to the layer thickness and window position in the platelet export boundary set, and generates platelet concentrated gravity separation data.

[0012] As a further solution of the present invention, the blood tension perturbation section includes a tension jump frequency value, a mutation window time period range, a tension range statistical sequence, and an abnormal fluctuation duration. The set of viscosity change behavior inflection point sections includes a viscosity response change slope value, a viscosity trend inflection point position, a viscosity rate boundary section, and a viscosity reversal duration. The platelet sedimentation trend data includes a liquid column acceleration attenuation sequence, a sedimentation velocity stable section, a displacement change distribution interval, and a liquid column descent fluctuation amplitude. The platelet export boundary set includes an export start window range, a platelet concentrated layer thickness distribution, a synchronous convergence pressure gradient section, and an internal pressure distribution sequence at the export port. The platelet concentrated gravity separation data specifically refers to a platelet density stable interval, a separation interface cross-sectional area, a concentrated layer persistence time, and a platelet distribution concentration gradient structure.

[0013] As a further solution of the present invention, the tension perturbation extraction module includes:

[0014] The tension bandwidth extraction sub-module obtains the tension sensing data at the channel liquid level during the static sedimentation process of blood, divides the tension data sequence according to each fixed time window, extracts the maximum tension value and the minimum tension value within the corresponding time window, takes the deviation between the maximum tension value and the minimum tension value as the tension mutation bandwidth, and sequentially obtains the tension mutation bandwidth data sequence within all time windows;

[0015] Based on the mutation bandwidth in the tension mutation bandwidth distribution sequence, the abnormal mutation recognition sub-module compares each mutation bandwidth with the blood liquid surface tension fluctuation threshold, screens the window numbers and time positions where the mutation bandwidth is greater than the blood liquid surface tension fluctuation threshold, and obtains a set of mutation abnormal time periods;

[0016] The tension frequency classification sub-module obtains the set of mutation abnormal time periods, extracts the tension change frequency values in the corresponding time periods segment by segment, statistically analyzes the characteristic indexes of the frequency fluctuation amplitude, frequency fluctuation density, and frequency stability trend, and uses the formula:

[0017]

[0018] Calculate the perturbation aggregation index P of the i-th segment after normalization processing i , classify the perturbation aggregation index into multiple perturbation types, and generate a blood tension perturbation section;

[0019] Among them, f ij is the j-th tension change frequency sampling value, is the mean value of the frequency sampling values of the i-th segment, d ij represents the time interval of the j-th frequency sampling value from the central moment, T i represents the total time window length corresponding to the i-th segment, m i is the number of frequency sampling points of the i-th segment, Δf ik is the difference between the k-th frequency sampling value and the previous sampling value.

[0020] As a further solution of the present invention, the viscous response positioning module includes:

[0021] The viscous data extraction sub-module extracts the viscous sensing data of the blood sample during the same time period in the static sedimentation process based on the time period corresponding to the blood tension perturbation section, matches the start and end times of all corresponding time periods, calls the original sequence recorded by the viscous sensor and reads it in chronological order, and generates a corresponding time period viscous sequence;

[0022] The viscous curve construction sub-module sorts the time and viscous values in chronological order according to the corresponding time period viscous sequence to form a continuous viscous curve, performs sampling point interpolation and connection on the curve formed by each response segment, and generates a viscous response slope sequence;

[0023] The behavior inflection point recognition sub-module based on the viscous response slope sequence judges the slope change of the viscous change trend in each time period, screens the time periods where the trend changes from increasing to stable or decreasing, records the turning point positions before and after the change, and measures the range of the viscous change rate interval, and establishes a set of viscous change behavior inflection point sections.

[0024] As a further solution of the present invention, the sedimentation slowdown judgment module includes:

[0025] The acceleration construction sub-module extracts the start and end times, as well as the liquid column height values and the descending rate sequences according to the time periods indicated in the set of inflection point sections of the viscous change behavior, and uses the formula:

[0026]

[0027] Calculate the average descending rate of the i1-th inflection point section and the average rate change rate of the i2-th inflection point section respectively, and establish a sequence of liquid column acceleration changes;

[0028] Among them, represents the liquid column height value monitored at the start moment of the i1-th inflection point time period, represents the liquid column height value monitored at the end moment of the i1-th inflection point time period, represents the start time point of the i1-th inflection point time period, represents the end time point of the i1-th inflection point time period, g represents the total number of liquid column descending rate sampling points within the inflection point time period numbered i2, represents the liquid column descending rate corresponding to the j2-th time point within the inflection point time period numbered i2, represents the liquid column descending rate corresponding to the (j2 + 1)-th time point within the inflection point time period numbered i2, represents the time point corresponding to the j2-th rate value, represents the time point corresponding to the (j2 + 1)-th rate value;

[0029] The acceleration change trend recognition sub-module, based on the sequence of liquid column acceleration changes, identifies whether the acceleration values at consecutive time points show a monotonically decreasing trend, and whether the end acceleration value is lower than the sedimentation deceleration control threshold. If both of the above two conditions are met, then this time period is marked as the sedimentation acceleration slowdown section;

[0030] The sedimentation feature extraction sub-module processes the data within the time range corresponding to each section of the sedimentation acceleration slowdown section, extracts the liquid column displacement rate sequence, counts the time length between the start and end time points as the duration, and extracts the maximum and minimum liquid column heights, calculates the difference as the liquid column height fluctuation range, and obtains the platelet sedimentation trend data.

[0031] As a further solution of the present invention, the platelet export discrimination module includes:

[0032] Based on the time period information of the platelet sedimentation trend data, the static pressure deviation calculation sub-module obtains the cumulative mass sensing data and the bottom static pressure change sequence of the blood sample within the corresponding time period, extracts the real-time height of the liquid column at the sampling moment, calculates the theoretical static pressure height corresponding to each moment, compares it with the real-time height, determines whether the deviation is stable within the convergence interval threshold, and obtains the convergence control section;

[0033] The synchronization segment screening sub-module extracts the liquid column acceleration and the mass change rate of the convergence control section, compares whether they are both within the stable control interval range, extracts the liquid column height change amplitude, the bottom static pressure difference and the corresponding time span of each segment, and obtains the platelet export boundary set.

[0034] As a further solution of the present invention, the concentrated layer interface output module includes:

[0035] The pressure section judgment sub-module selects the pressure sensor sequence and the density distribution sensing data within the corresponding channel section range according to the layer thickness and the window position in the platelet export boundary set, compares the pressure sampling results of consecutive time periods within the channel section, determines whether it is continuously within the set pressure stable interval, and obtains the pressure stable interval paragraph;

[0036] The density gradient extraction sub-module calls the platelet density sensing distribution within the pressure stable interval paragraph, calculates the relative change of the density value in chronological order, constructs a density change gradient sequence, and measures the duration range of each stable section, and obtains the platelet density gradient and the stable duration parameter;

[0037] The structural parameter analysis sub-module calculates the effective separation cross-sectional area of the corresponding area according to the platelet density gradient and the stable duration parameter, combined with the position and layer thickness data of the export boundary layer, and makes a judgment according to the density stability degree and the structural contour change situation, and obtains the platelet concentration gravity separation data.

[0038] Compared with the prior art, the advantages and positive effects of the present invention are:

[0039] In the present invention, by introducing the real-time extraction of tension change frequency and mutation amplitude, combined with the normalization processing method of disturbance aggregation index, the mechanical instability area under small disturbance during the static sedimentation of blood is effectively portrayed, and the sensitive recognition ability of potential stratification behavior changes is enhanced. By continuously tracking the viscosity behavior curve during the disturbance period, the inflection point of the viscous resistance change inside the fluid is located, and the turning characteristics of the sedimentation behavior in the process of stratification evolution are identified, providing an accurate time anchor point for the subsequent determination of the motion state of the liquid column. Based on the microscopic changes in the height and rate of the liquid column, a multi-dimensional sedimentation trend identification mechanism is constructed through the joint judgment of the acceleration change trend and the attenuation amplitude, so that the identification of the sedimentation state no longer depends on the macro parameter threshold, but is dynamically regulated by the change trend and the fluctuation sequence. By collecting the coupling change relationship between the bottom static pressure and the mass change data, the pressure convergence interval is established, and the precise control of the export timing is achieved, which improves the stability and timeliness of the platelet concentration layer export operation. In the interface judgment stage, the spatial structural parameters and existence interval of the platelet concentration layer are determined through the joint analysis of pressure stability and density gradient sequence, which improves the accuracy of target layer identification in the separation process, and improves the overall concentration efficiency and platelet purity. The overall processing flow is driven by multi-source real-time data, opening up the full-link perception and feedback mechanism between physical disturbances, rheological behavior, acceleration evolution, pressure changes and concentration structure, so that the separation process can achieve high-precision identification, flexible adjustment and adaptive export, and enhance the adaptability to nonlinear sedimentation behavior in complex fluid systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 is a system flow chart of the present invention;

[0041] Figure 2 It is a flow chart of the tension disturbance extraction module of the present invention;

[0042] Figure 3 It is a flow chart of the viscous response positioning module of the present invention;

[0043] Figure 4 This is a flow chart of the sedimentation slowdown judgment module of the present invention;

[0044] Figure 5 A flow chart of a platelet derivation and discrimination module of the present invention;

[0045] Figure 6 It is a flow chart of the concentrated layer interface output module of the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0047] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present invention. In addition, in the description of the present invention, the meaning of "a plurality of" is two or more, unless otherwise specifically defined.

[0048] Please refer to Figure 1 , the gravity separation system based on platelet concentration includes:

[0049] The tension disturbance extraction module obtains the tension sensing data at the channel liquid level during the static sedimentation of blood, calculates the disturbance aggregation index of the tension change, selects the disturbance time period, and generates the blood tension disturbance section.

[0050] The viscous response positioning module obtains the viscous sensing data of the blood sample during the same disturbance time period in the static sedimentation process based on the blood tension disturbance section, identifies the inflection point section therefrom, and generates a set of inflection point sections of the viscous change behavior.

[0051] The sedimentation deceleration judgment module extracts the liquid column displacement rate sequence, duration, and fluctuation range within the target time period according to the time corresponding to the inflection point section in the set of inflection point sections of the viscous change behavior, and generates platelet sedimentation trend data.

[0052] The platelet export discrimination module obtains the cumulative mass sensing data and the bottom static pressure change of the blood sample based on the time period information of the platelet sedimentation trend data, judges whether it enters the convergence control interval of the sedimentation height, and generates a platelet export boundary set.

[0053] The concentrated layer interface output module judges whether the pressure change is in the pressure stable interval according to the layer thickness and window position in the platelet export boundary set, and generates platelet concentration gravity separation data.

[0054] The blood tension disturbance section includes the tension jump frequency value, the mutation window time period range, the tension range statistical sequence, and the abnormal fluctuation duration. The set of inflection point sections of the viscosity change behavior includes the viscosity response change slope value, the viscosity trend inflection point position, the viscosity rate boundary section, and the viscosity inversion duration period. The platelet sedimentation trend data includes the liquid column acceleration attenuation sequence, the sedimentation velocity stable section, the displacement change distribution interval, and the liquid column drop fluctuation amplitude. The platelet export boundary set includes the export start window range, the platelet concentration layer thickness distribution, the synchronous convergence pressure gradient section, and the internal pressure distribution sequence at the export port. The platelet concentration gravity separation data specifically refers to the platelet density stable interval, the separation interface cross-sectional area, the duration of the concentration layer, and the platelet distribution concentration gradient structure.

[0055] Please refer to Figure 2 , the tension disturbance extraction module includes:

[0056] The tension bandwidth extraction sub-module obtains the tension sensing data at the channel liquid level during the static sedimentation of blood, divides the tension data sequence according to each fixed time window, extracts the maximum tension value and the minimum tension value within the corresponding time window, and takes the deviation between the maximum tension value and the minimum tension value as the tension mutation bandwidth, and sequentially obtains the tension mutation bandwidth data sequence within all time windows;

[0057] When obtaining the tension sensing data at the channel liquid level during the static sedimentation of blood, first set a tension sensor at the channel liquid level on the experimental platform. For example, a strain gauge sensor is used, and its sampling frequency is set to 50 Hz to continuously monitor the tension change curve of blood during sedimentation. The acquisition time lasts for 600 seconds, 50 data points are acquired per second, and the total data volume reaches 30,000. Subsequently, the data is divided into time windows with a duration of 10 seconds each, that is, each window contains 500 data points. Each time window is independently numbered and data processing is performed sequentially. The maximum and minimum tension values are extracted in each time window. For example, in the 5th time window, the maximum tension value is 8.6 mN and the minimum tension value is 3.2 mN, then the tension mutation bandwidth is 5.4 mN. By traversing all time windows and repeating this calculation process, the mutation bandwidth value sequence of all time windows is obtained. During this process, each bandwidth value is identified and stored in a unified unit of mN to form a tension mutation bandwidth sequence.

[0058] The abnormal mutation recognition sub-module compares each mutation bandwidth in the tension mutation bandwidth distribution sequence with the blood liquid surface tension fluctuation threshold, and screens the window numbers and time positions where the mutation bandwidth is greater than the blood liquid surface tension fluctuation threshold to obtain the set of mutation abnormal time periods;

[0059] Call the preset blood liquid surface tension fluctuation threshold and perform a judgment operation. This threshold is set at 4.0 mN through a preliminary analysis of the background fluctuation range of the tension change during the previous blood sedimentation process. The setting basis is that the bandwidth values of most tension mutations in the non-disturbed state are lower than 3.8 mN. Therefore, 4.0 mN is used as the boundary value to identify the abnormal disturbance section. For example, in the time window numbered 11, its bandwidth value is 4.3 mN, which is greater than the set threshold of 4.0 mN. Then it is determined that this time window is an abnormal mutation window, and its start and end times are recorded. If adjacent time windows continuously meet the conditions, they are merged. Finally, a list of abnormal mutation time periods is obtained. For example, it is identified that the time windows from 11 to 13 mutate continuously, and the corresponding start and end times are 110 seconds to 140 seconds, forming an abnormal mutation period and adding it to the list. A total of 8 abnormal mutation time periods are identified during the entire sedimentation process. The judgment operation in this process is achieved through an operation that compares the current window bandwidth value with the threshold. The set value of the threshold is shown in the following calculation process. The threshold setting example is as follows: Calculate the mean and standard deviation of the bandwidth values of 20 randomly selected non-abnormal state windows. Assume the mean is 3.2 mN and the standard deviation is 0.6 mN. Then the set threshold is the mean plus 1.5 times the standard deviation, that is: Threshold = 3.2 + 1.5×0.6 = 4.1 mN. This threshold is used for the judgment of all subsequent bandwidth values to obtain a list of tension abnormal mutation time periods;

[0060] The tension frequency classification sub-module obtains the set of mutation abnormal time periods, extracts the tension change frequency values in the corresponding time periods segment by segment, and statistically analyzes the characteristic indexes of the frequency fluctuation amplitude, frequency fluctuation density, and frequency stability trend. The formula is used:

[0061]

[0062] Calculate the normalized disturbance aggregation index Pi of the i-th segment i , classify the disturbance aggregation index into multiple disturbance types, and generate a blood tension disturbance section;

[0063] Among them, f ij is the j-th tension change frequency sampling value (Hz), is the mean value (Hz) of the i-th segment of frequency sampling values, d ij represents the time interval (s) between the j-th frequency sampling value and the central moment of this segment, T i represents the total duration (s) of the time window corresponding to the i-th segment, m i is the number of frequency sampling points of the i-th segment, Δf ik is the difference (Hz) between the k-th frequency sampling value and its previous sampling value.

[0064] After obtaining the list of abnormal tension mutation time periods, for each mutation time period, extract the corresponding tension change frequency sequence. Based on a tension sampling rate of 50 times per second, use the frequency domain discretization method to extract local frequency components. For example, in the 3rd abnormal time period (120s - 140s), the frequency sequence f ij is [2.4Hz, 2.1Hz, 2.5Hz, 2.7Hz, 2.6Hz], and the average frequency within the segment is 2.46Hz, the central time is 130 seconds, and the time interval d of each data point ij is [10, 5, 0, 5, 10] seconds respectively, and the total segment duration T i is 20 seconds, the total number of frequency points m i = 5, and the adjacent frequency difference Δf ik is [-0.3Hz, 0.4Hz, 0.2Hz, -0.1Hz] respectively. Substitute the above values into the formula:

[0065]

[0066] The calculated disturbance aggregation index value is 0.0463. This result is lower than the set aggregation intensity reference value of 0.08. Therefore, it is classified into the low disturbance type section. Finally, based on the disturbance aggregation index values of all sections, classification and aggregation are carried out to establish a blood tension disturbance section;

[0067] Through the normalization processing of frequency deviation, adjacent frequency fluctuation amplitude, and frequency sampling time distance, a dimensionless index sensitive to the disturbance fluctuation intensity is constructed to accurately reflect the density and complexity of tension disturbance during blood sedimentation. This result shows that the disturbance fluctuation in this section is weak and no complex disturbance structure is formed.

[0068] Please refer to Figure 3 , the viscosity response positioning module includes:

[0069] The viscosity data extraction sub-module extracts the viscosity sensing data of the blood sample during the same time period of static sedimentation based on the time period corresponding to the blood tension disturbance section, matches the start and end times of all corresponding time periods, calls the original sequence recorded by the viscosity sensor and reads it in chronological order to generate the viscosity sequence corresponding to the time period;

[0070] First, the start time and end time information of each tension disturbance section are identified, and they are uniformly converted into a timestamp interval in seconds. For example, if the start time of a disturbance section is 185 seconds and the end time is 210 seconds, the corresponding timestamp interval is [185, 210] seconds. Then, the viscosity sensor data sequence recorded at a sampling frequency of 1 second during the entire static sedimentation process of the blood sample is retrieved. Each data point represents the viscosity reading at a certain second in Pa·s. In the above example, the range of [185, 210] seconds is used as the index interval, and the viscosity values ​​in the interval are extracted to form an array. For example, the viscosity value sequence is [1.92, 2.01, 2.03, 2.04, 2.07, 2.12, 2.10, 2.09, 2.04, 1.98, 1.95, 1.91, 1.89, 1.88, 1.85, 1.84], and a total of 16 The viscosity data points are one-to-one corresponding to the original timestamps. Then all disturbance segments are sequentially executed with this operation. An independent viscosity data subsequence is constructed in each disturbance segment. To avoid the interference of abnormal values ​​of the viscosity sensor, it is necessary to perform edge anomaly detection on the sequence of each segment after data extraction, and remove the data points whose viscosity values ​​exceed the mean ± 2.5 times the standard deviation. For example, the mean of this segment is 1.98 Pa·s and the standard deviation is 0.08 Pa·s, then the abnormal judgment interval is [1.78, 2.18]. If a data point is 2.22 Pa·s, then the point is deleted from the array. After processing, each viscosity subsequence is bound to the time period identifier of the original disturbance segment to form a structured data pair. Then, multiple structure pairs are combined to generate a response matrix, including the time period and the corresponding viscosity value sequence, and finally a corresponding time-viscosity sequence set of a continuous structure is formed to obtain the corresponding time period viscosity sequence.

[0071] The viscosity curve construction submodule arranges the time and viscosity values ​​into a continuous viscosity curve according to the viscosity sequence of the corresponding time period in the sequence time order, interpolates and connects the sampling points of the curve formed by each response segment, and generates a viscosity response slope sequence;

[0072] First, arrange the viscosity value sequence in each structure in chronological order, and construct the response relationship between the viscosity value and the time point. Assuming that the viscosity value sequence in a section is [1.88, 1.91, 1.93, 1.95, 1.94, 1.92, 1.89], the corresponding time series is [0, 1, 2, 3, 4, 5, 6] seconds, and construct a two-dimensional value pair [(0, 1.88), (1, 1.91), (2, 1.93), (3, 1.95), (4, 1.94), (5, 1.92), (6, 1.89)]. This data pair constitutes the discrete points of the viscosity response curve. By calculating the slope between adjacent points, the response slope sequence of this section is obtained. The slope calculation adopts the difference divided by the time interval. The above data points are processed item by item: For example, (1.91 - 1.88) / 1 = 0.03,

[0073] (1.93 - 1.91) / 1 = 0.02, (1.95 - 1.93) / 1 = 0.02, (1.94 - 1.95) / 1 = -0.01,

[0074] (1.92 - 1.94) / 1 = -0.02, (1.89 - 1.92) / 1 = -0.03

[0075] Finally, the slope sequence is obtained: [+0.03, +0.02, +0.02, -0.01, -0.02, -0.03], indicating the trend that the viscosity value first rises and then falls. Each slope value corresponds one-to-one with its time position. This process is carried out for the viscosity response sequences constructed for all disturbance sections to obtain the response slope sequence for each time period. Each item is the instantaneous change rate at that time point. Subsequently, it is saved in the order of time periods and organized into a structured set for subsequent judgment steps, and finally the viscosity response slope sequence is obtained.

[0076] Based on the viscosity response slope sequence, the behavior inflection point recognition sub-module judges the slope change of the viscosity change trend in each time period, screens the time periods where the trend changes from increasing to stable or decreasing, records the turning point positions before and after the change, and determines the range of the viscosity change rate, establishing a set of viscosity change behavior inflection point sections;

[0077] Call the established viscosity response slope sequence to judge point by point the viscosity change trend in each time period. The judgment process is based on the change relationship between adjacent two slope items. If the slope of the current point is positive and the slope of the next point is zero or negative, then the current point is marked as a trend inflection point. The time period that continuously satisfies this change condition is defined as an inflection point section. Still taking the aforementioned slope sequence [+0.03, +0.02, +0.02, -0.01, -0.02, -0.03] as an example, from the 3rd to the 4th point, it changes from +0.02 to -0.01, that is, from positive to negative. It is determined that the 3rd point is the starting point of the inflection point, and the start and end times are recorded as [3, 6] seconds. The viscosity value sequence for this section of viscosity value extraction is [1.95, 1.94, 1.92, 1.89], and the viscosity rate range is the difference between the maximum change point and the minimum point Δviscosity value = 1.95 - 1.89 = 0.06 Pa·s. Execute this method for each section of all viscosity response slope sequences. In each execution, eliminate the sequences that do not meet the condition of changing from rising to stable or decreasing, construct the inflection point set for all sections, record the time range and rate change interval of each section of viscosity change behavior, and finally establish a set of viscosity change behavior inflection point sections.

[0078] Please refer to Figure 4 , the sedimentation slowdown judgment module includes:

[0079] According to the time periods indicated in the set of inflection point segments of the viscous change behavior, the acceleration construction sub-module extracts the start and end times, as well as the liquid column height values and the descending rate sequence, and uses the formula:

[0080]

[0081] Calculate the average descending rate of the i1-th inflection point segment and the average rate change rate of the i2-th inflection point segment to establish a sequence of liquid column acceleration changes;

[0082] where, represents the liquid column height value monitored at the start moment of the i1-th inflection point time period, with the unit of millimeter, represents the liquid column height value monitored at the end moment of the i1-th inflection point time period, with the unit of millimeter, represents the start time point of the i1-th inflection point time period, with the unit of second, represents the end time point of the i1-th inflection point time period, with the unit of second, g represents the total number of liquid column descending rate sampling points within the inflection point time period numbered i2, taking a positive integer value and being greater than or equal to 2, j2 represents the index number of the rate sampling point within the time period, with the value range from 1 to g - 1, used to traverse the differences between rate points, represents the liquid column descending rate corresponding to the j2-th time point within the inflection point time period numbered i2, with the unit of millimeter per second, represents the liquid column descending rate corresponding to the j2 + 1-th time point within the inflection point time period numbered i2, with the unit of millimeter per second, represents the time point corresponding to the j2-th rate value, with the unit of second, represents the time point corresponding to the j2 + 1-th rate value, with the unit of second.

[0083] According to the time periods indicated in the set of inflection point segments of the viscous change behavior, extract the start and end time points of the i1 = 3, i2 = 3 inflection point segments, which are 640 seconds and 670 seconds respectively. The monitored liquid column height value within this time period is 122.6 millimeters at the start time and 119.5 millimeters at the end time. Calculate the height change value as 122.6 - 119.5 = 3.1 millimeters, and the corresponding time difference is 670 - 640 = 30 seconds. Substitute into the height change calculation formula to obtain: Obtain the liquid column descending rate sequence within this segment, with a sampling frequency of 1 Hz and the number of sampling points g = 8, to obtain the rate sequence as:

[0084] v3 = [0.112, 0.108, 0.101, 0.096, 0.091, 0.089, 0.085, 0.080] mm / s, and the corresponding time point sequence is: t3 = [640, 641, 642, 643, 644, 645, 646, 647] s. Substitute the above rate sequence into the calculation formula of the average rate change corresponding to the falling rate difference:

[0085]

[0086] Expand and calculate item by item as follows:

[0087]

[0088] Finally, the average falling rate of this section is 0.103 mm / s, and the average rate change is 0.00457 mm / s². These two values are jointly used to establish the acceleration characteristics of this section, and finally the liquid column acceleration change sequence is obtained.

[0089] Based on the liquid column acceleration change sequence, the acceleration change trend recognition sub-module identifies whether the acceleration values at consecutive time points show a monotonically decreasing trend, and whether the end acceleration value is lower than the sedimentation deceleration control threshold. If both of the above conditions are met, this time period is marked as the sedimentation acceleration slowdown section;

[0090] Based on the liquid column acceleration change sequence, extract the acceleration value sequence in chronological order as [0.0075, 0.0062, 0.0051, 0.0045, 0.0042] mm / s². Compare the differences between adjacent data in turn, and the change amplitudes are: from the first section to the second section, it decreases by 0.0013, from the second section to the third section, it decreases by 0.0011, from the third section to the fourth section, it decreases by 0.0006, and from the fourth section to the fifth section, it decreases by 0.0003. It meets the condition of continuous decrease 4 times and meets the standard of "continuous decrease of acceleration values". Call the end acceleration value of this sequence as 0.0042 mm / s², and compare it with the set sedimentation deceleration control threshold of 0.0050 mm / s². The end value is lower than the threshold, which meets the judgment condition. The threshold setting is based on the average critical level of the acceleration values when the slowdown trend is formed in the previous 100 groups of liquid column sedimentation samples, and is set in the range of 0.0048 to 0.0052. In this example, the median value of 0.0050 is taken as the threshold reference, and "the value is less than the threshold" is used as the judgment standard, not allowing equality. Finally, the above paragraph is judged as the slowdown section, the complete time range is screened out, and the sedimentation acceleration slowdown section is obtained.

[0091] The settlement feature extraction sub-module processes the data within the time range corresponding to each section of the settlement acceleration slowdown section, extracts the liquid column displacement rate sequence, counts the time length between the start and end time points as the duration, extracts the maximum and minimum values of the liquid column height, calculates the difference as the liquid column height fluctuation range, and obtains the platelet settlement trend data;

[0092] Call the obtained settlement acceleration slowdown section with a time period from 760 seconds to 785 seconds, extract the liquid column displacement rate sequence within this section, with a sampling frequency of 1 Hz, and obtain the rate values as [0.083, 0.081, 0.080, 0.079, 0.078, 0.078, 0.077, 0.075, 0.074, 0.072] mm / s. Calculate the average rate value as: (0.083 + 0.081 + 0.080 + 0.079 + 0.078 + 0.078 + 0.077 + 0.075 + 0.074 + 0.072) / 10 = 0.0777 mm / s. Then calculate the duration as 785 - 760 = 25 seconds based on the start and end time differences. Next, extract the liquid column height sequence within this section as [66.4, 66.1, 65.9, 65.8, 65.6, 65.5, 65.4, 65.2, 65.1, 65.0] mm, obtain the maximum value of this section as 66.4 mm, the minimum value as 65.0 mm, and the fluctuation range as 66.4 - 65.0 = 1.4 mm. Finally, combine the average displacement rate, duration, and fluctuation range into a data group to obtain the platelet settlement trend data.

[0093] Please refer to Figure 5 , the platelet export discrimination module includes:

[0094] The static pressure deviation calculation sub-module, based on the time period information of the platelet settlement trend data, obtains the cumulative mass sensing data of the blood sample and the bottom static pressure change sequence within the corresponding time period, extracts the real-time height of the liquid column at the sampling moment, calculates the theoretical static pressure height corresponding to each moment, compares the deviation with the real-time height, determines whether the deviation is stable within the convergence interval threshold, and obtains the convergence control section;

[0095] First, the synchronous sampling data of the liquid column height, static pressure and mass in each section are extracted, the cumulative mass sensing values ​​collected per second are aggregated according to the time axis, and the sampling cross-sectional area is uniformly set. Assuming that the cross-sectional area of ​​the blood sample container is 11.3 square centimeters, the mass values ​​collected at t = 100 seconds, t = 120 seconds, and t = 140 seconds in the corresponding time period are 4.65 grams, 4.83 grams, and 5.01 grams, respectively. The mass value at each moment is divided by the sample cross-sectional area, and the theoretical static pressure height of each point is calculated to be 41.15 mm, 42.74 mm, and 44.34 mm. The actual monitored liquid column height values ​​in this period are 40.68 mm, 42.81 mm, and 44.22 mm, respectively. The difference between the theoretical value and the actual value is calculated to obtain a deviation sequence of [0.47, -0.07, 0.12] mm. For continuous data points with an absolute deviation value of less than 1.5 mm Identify and locate continuous sections in the time series, mark the minimum and maximum deviations of the sections, and classify the sections that meet this judgment condition as suspected highly convergent sections. Then, review the absolute value range of all deviation values ​​within the highly convergent sections to determine whether they all fall within the maximum allowable deviation setting threshold. The reference basis for setting the deviation threshold is the fluctuation range analysis of the samples in the stable stage during the sedimentation process. The threshold is set to ±1.5 mm. During the setting process, take the average upper limit of the maximum and minimum deviations of 20 groups of samples in the stable period to determine the threshold value range. If all deviation data in the section are within this interval, mark the section as a section that meets the requirements for entering the convergence control interval. In the example here, the section meets the conditions, so its start time of 100 seconds and end time of 140 seconds are recorded, and the time period range is output in the "convergence control section" state to finally obtain the highly convergent control section.

[0096] The synchronization segment screening submodule extracts the liquid column acceleration and mass change rate in the convergence control section, compares whether they are in the stable control range at the same time, extracts the liquid column height change amplitude, bottom static pressure difference and corresponding time span of each section, and obtains the platelet export boundary set;

[0097] Extract the acceleration values and mass change rates in this section in sequence. The sampling frequency is uniformly set to 1 Hz. Record the liquid column velocity values in this section as [0.094, 0.091, 0.088, 0.087] mm / s and the mass values as [5.01, 5.03, 5.05, 5.07] g respectively. Calculate the change amplitude of the slope between adjacent points from the velocity value sequence, which are -0.003, -0.003, -0.001 mm / s² respectively. Take the average value to get the acceleration of -0.00233 mm / s². Compare the absolute value of this value, 0.00233 mm / s², with the stable determination acceleration interval [0.002, 0.006] mm / s². Determine that this value is within the allowable interval and the acceleration is stable. Then divide the difference between adjacent mass points by the time interval to calculate the mass rates as 0.02, 0.02, 0.02 g / s respectively, and the average value is 0.02 g / s. This value is higher than the upper limit of the set stable mass rate interval of 0.009 g / s and does not meet the stable interval condition. Therefore, the current section does not meet the synchronous stability condition. Select the next section as [180 s to 220 s]. Extract the acceleration values as [0.0043, 0.0041, 0.0040, 0.0039] mm / s², and the average value is 0.00408 mm / s². The mass values are [5.12, 5.14, 5.15, 5.16] g, and the average mass rate is (5.16 - 5.12) / 3 = 0.0133 g / s. This value is within the stable interval [0.005, 0.015]. Confirm that this section meets the synchronous stability condition. Extract the minimum value of the liquid column height in this section as 43.1 mm, the maximum value as 44.7 mm, the derived thickness as 1.6 mm, the maximum and minimum difference of the static pressure as 0.4 kPa, and the time span as 40 s. Record the above parameters and combine them into a control unit to obtain the platelet-derived boundary set.

[0098] Please refer to Figure 6 , and the concentrated layer interface output module includes:

[0099] The pressure section judgment sub-module selects the pressure sensor sequence and density distribution sensing data within the corresponding channel section range according to the layer thickness and window position in the platelet-derived boundary set, compares the pressure sampling results in the continuous time period within the channel section, judges whether it is continuously in the set pressure stable interval, and obtains the pressure stable interval paragraph;

[0100] First, clarify the window position and thickness boundary value range involved in the parameter set, obtain the liquid channel cross-sectional segment number corresponding to the window, sequentially retrieve the pressure sampling data recorded by the pressure sensors arranged in the channel segment, and map and locate the platelet density sensor nodes in the corresponding area of this segment to form an associated data set containing the pressure change sequence and density data distribution. Then, extract each pressure value point by point based on time, and obtain the change amount of the pressure value between adjacent time points. If the absolute value of the change amplitude of the pressure value within a continuous time period is less than the set pressure stability reference threshold, mark this time period as a stable state. The stability threshold is set to 0.2 kPa according to the sensor range and control accuracy requirements. Compare the change results of the pressure differences at multiple points within the time period. For example, in a certain area, the pressure values at five consecutive time points are 35.2, 35.3, 35.2, 35.4, and 35.3 kPa respectively. The maximum change value between any two adjacent points is 0.2 kPa, which is determined to meet the stable condition, and the time period of this area is included in the stable interval. Finally, output the set of all time periods within the window segment range that meet the stable judgment rules to obtain the pressure stable interval paragraph.

[0101] The density gradient extraction sub-module calls the platelet density sensing distribution within the pressure stable interval paragraph, calculates the relative change of the density value in chronological order, constructs a density change gradient sequence, and measures the duration range of each stable segment to obtain the platelet density gradient and stable duration parameters;

[0102] Call the platelet density sensing data corresponding to the pressure stable interval paragraph, obtain the density values at each sampling moment in chronological order, construct a continuous-time density value sequence, and calculate the difference between adjacent data points in this sequence to form a change gradient sequence. Use whether the absolute value of the continuous density difference is stably maintained within a density change limit value as the judgment criterion for whether it is a "density gradient extractable section". The density change limit value is set to 0.005 g / cm³. If the density values in a certain interval are 0.195, 0.198, 0.200, 0.202, and 0.204 g / cm³ respectively, then the change gradients are 0.003, 0.002, 0.002, and 0.002 g / cm³. If it is continuously less than the limit value, it is considered to meet the extraction condition. Record the density gradient of this section and measure its time span. Repeat the above process for each stable segment to obtain the density change gradients and duration values within all stable time periods, and organize and summarize the time span and gradient amplitude as a data group. Finally, obtain the platelet density gradient and stable duration parameters.

[0103] The structural parameter analysis sub-module calculates the effective separation cross-sectional area of the corresponding area based on the platelet density gradient and stable duration parameters, combined with the derived boundary layer position and layer thickness data, and makes a judgment according to the density stability degree and structural contour change situation to obtain the platelet concentration gravity separation data;

[0104] Take the derived boundary layer position given in the boundary parameter set and the corresponding layer thickness value for this region, determine the relative height range of the derived region in the fluid channel, and locate the specific vertical position section where this range is located within the current cross-section. According to the assumption that the cross-sectional structure of the channel is a rectangular channel, call the width and thickness parameters of the channel in this height section to construct the spatial structure form of the derived region in the cross-sectional direction. Further derive the projected area of this region within the current layer thickness range. Consider this thickness segment as a unit strip, and obtain the unit area by multiplying the width value by the thickness value. If the thickness is 2 mm and the corresponding channel width is 10 mm, the separation cross-sectional area is 20 mm². If the derived region also includes multiple adjacent unit segments longitudinally, sum the areas of all unit strips to form the complete separation cross-sectional area of the derived region. Subsequently, combine whether the change in platelet density gradient within this region shows a continuous stratification trend within the cross-sectional center range to determine whether the structural stability condition is met. If the density difference between the central region and the edge region is greater than 0.01 g / cm³ and the maintenance time exceeds the stability threshold of 30 seconds, record this region as a stable concentration structure. Finally, mark and summarize the separation cross-sectional area and the concentration structure state correspondingly to generate platelet concentration gravity separation data.

[0105] The above are only the preferred embodiments of the present invention and do not limit the present invention in other forms. Any person skilled in the art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A gravity separation system based on platelet concentration, characterized in that The system includes: The tension disturbance extraction module obtains the tension sensing data at the channel liquid level during the static sedimentation process of the blood, calculates the disturbance aggregation index of the tension change, selects the disturbance time period, and generates the blood tension disturbance section; The viscosity response positioning module obtains the viscosity sensing data of the blood sample during the same disturbance time period during the static sedimentation process based on the blood tension disturbance section, identifies the inflection point section therefrom, and generates a set of viscosity change behavior inflection point sections; The sedimentation deceleration judgment module extracts the liquid column displacement rate sequence, duration, and fluctuation range within the target time period according to the time corresponding to the inflection point section in the set of viscosity change behavior inflection point sections, and generates platelet sedimentation trend data; The platelet export discrimination module obtains the cumulative mass sensing data and the bottom static pressure change of the blood sample based on the time period information of the platelet sedimentation trend data, judges whether it enters the convergence control interval of the sedimentation height, and generates a platelet export boundary set; The concentrated layer interface output module judges whether the pressure change is within the pressure stable interval according to the layer thickness and window position in the platelet export boundary set, and generates platelet concentrated gravity separation data.

2. The platelet concentrate-based gravity separation system according to claim 1, wherein, The blood tension disturbance section includes the tension jump frequency value, the mutation window time period range, the tension range statistical sequence, and the abnormal fluctuation duration. The set of viscosity change behavior inflection point sections includes the viscosity response change slope value, the viscosity trend inflection point position, the viscosity rate boundary section, and the viscosity reversal duration period. The platelet sedimentation trend data includes the liquid column acceleration attenuation sequence, the sedimentation speed stable section, the displacement change distribution interval, and the liquid column descent fluctuation amplitude. The platelet export boundary set includes the export start window range, the platelet concentrated layer thickness distribution, the synchronous convergence pressure gradient section, and the internal pressure distribution sequence within the export port. The platelet concentrated gravity separation data specifically refers to the platelet density stable interval, the separation interface cross-sectional area, the duration of the concentrated layer existence, and the platelet distribution concentration gradient structure.

3. The platelet concentrate-based gravity separation system according to claim 2, wherein The tension disturbance extraction module includes: The tension bandwidth extraction sub-module obtains the tension sensing data at the channel liquid level during the static sedimentation process of the blood, divides the tension data sequence according to each fixed time window, extracts the maximum tension value and the minimum tension value within the corresponding time window, takes the deviation between the maximum tension value and the minimum tension value as the tension mutation bandwidth, and sequentially obtains the tension mutation bandwidth data sequence within all time windows; The abnormal mutation identification sub-module compares each mutation bandwidth in the tension mutation bandwidth distribution sequence with the blood liquid surface tension fluctuation threshold, screens the window numbers and time positions with mutation bandwidths greater than the blood liquid surface tension fluctuation threshold, and obtains a set of mutation abnormal time periods; The tension frequency classification sub-module obtains the set of mutation abnormal time periods, extracts the tension change frequency values within the corresponding time periods one by one, statistically analyzes the characteristic indexes of the frequency fluctuation amplitude, frequency fluctuation density, and frequency stability trend, and uses the formula: Calculate the perturbation aggregation index Pi of the i-th segment after normalization processing i , classify the perturbation aggregation index into multiple perturbation types, and generate a blood tension perturbation section; where, f ij is the j-th sampling value of the tension change frequency, is the mean value of the i-th segment of the frequency sampling values, d ij represents the time interval of the j-th frequency sampling value from the central moment, T i represents the total duration of the time window corresponding to the i-th segment, m i is the number of frequency sampling points of the i-th segment, Δf ik is the difference between the k-th frequency sampling value and the previous sampling value.

4. The platelet concentrate-based gravity separation system according to claim 3, characterized in that, The viscosity response positioning module includes: The viscous data extraction sub-module extracts the viscous sensing data of the blood sample during the same time period in the static sedimentation process based on the time period corresponding to the blood tension disturbance section, matches the start and end times of all corresponding time periods, calls the original sequence recorded by the viscous sensor and reads it in chronological order to generate a viscous sequence for the corresponding period; The viscous curve construction sub-module sorts the time and viscous values into a continuous viscous curve in the order of the sequence time according to the viscous sequence for the corresponding period, performs sampling point interpolation and connection on the curve formed by each response section to generate a viscous response slope sequence; The behavior inflection point identification sub-module judges the slope change of the viscous change trend in each time period based on the viscous response slope sequence, screens the time periods that change from an increasing trend to a stable or decreasing trend, records the positions of the turning points before and after the change, and determines the range of the viscous change rate interval to establish a set of viscous change behavior inflection point sections; 5. The platelet concentrate-based gravity separation system according to claim 4, wherein The sedimentation deceleration judgment module includes: The acceleration construction sub-module extracts the start and end times, the liquid column height value and the descent rate sequence according to the time period indicated in the set of viscous change behavior inflection point sections, and uses the formula: Calculate the average descent rate of the i1 inflection point segment and the average rate of change of the i2 inflection point segment respectively; Establish a sequence of changes in the liquid column acceleration; Wherein, represents the liquid column height value monitored at the start time of the i1-th inflection point time period, represents the liquid column height value monitored at the end time of the i1-th inflection point time period, represents the start time point of the i1-th inflection point time period, represents the end time point of the i1-th inflection point time period, and g represents the total number of liquid column descent rate sampling points within the inflection point time period numbered i2, represents the liquid column descent rate corresponding to the j2-th time point within the inflection point time period numbered i2, represents the liquid column descent rate corresponding to the (j2 + 1)-th time point within the inflection point time period numbered i2, represents the time point corresponding to the j2-th rate value, represents the time point corresponding to the (j2 + 1)-th rate value; The acceleration change trend identification sub-module identifies whether the acceleration values at consecutive time points show a monotonically decreasing trend and whether the end acceleration value is lower than the sedimentation deceleration control threshold based on the liquid column acceleration change sequence. If both of the above conditions are met, the time period is marked as a sedimentation acceleration slowdown section; The sedimentation feature extraction sub-module processes the data within the time range corresponding to each section of the sedimentation acceleration slowdown section, extracts the liquid column displacement rate sequence, counts the time length between the start and end time points as the duration, and extracts the maximum and minimum values of the liquid column height, calculates the difference as the liquid column height fluctuation range, and obtains the platelet sedimentation trend data.

6. The platelet concentrate-based gravity separation system according to claim 5, characterized in that, The platelet export discrimination module includes: The static pressure deviation calculation sub-module obtains the cumulative mass sensing data and the bottom static pressure change sequence of the blood sample within the corresponding time period based on the time period information of the platelet sedimentation trend data, extracts the real-time height of the liquid column at the sampling moment, calculates the theoretical static pressure height corresponding to each moment, compares it with the real-time height, judges whether the deviation is stable within the convergence interval threshold, and obtains the convergence control section; The synchronization section screening sub-module extracts the liquid column acceleration and the mass change rate of the convergence control section, compares whether they are both within the stable control interval range, extracts the liquid column height change amplitude, the bottom static pressure difference and the corresponding time span of each section, and obtains the platelet export boundary set.

7. The platelet concentrate-based gravity separation system according to claim 6, characterized in that, The concentrated layer interface output module includes: The pressure section judgment sub-module selects the pressure sensor sequence and the density distribution sensing data within the corresponding channel section range according to the layer thickness and the window position in the platelet export boundary set, compares the pressure sampling results of consecutive time periods within the channel section, and judges whether it is continuously within the set pressure stable interval to obtain the pressure stable interval paragraph; The density gradient extraction sub-module calls the platelet density sensing distribution within the pressure stable interval paragraph, calculates the relative change of the density value in chronological order, constructs a density change gradient sequence, and measures the duration range of each stable segment to obtain the platelet density gradient and stable duration parameters; The structural parameter analysis sub-module calculates the effective separation cross-sectional area of the corresponding region based on the platelet density gradient and stable duration parameters, combined with the derived boundary layer position and layer thickness data, and makes a judgment based on the density stability degree and the change of the structural contour to obtain the platelet concentration gravity separation data.

Images