An interface detection system and method for a blood component separation system
By employing a multi-channel photoelectric detection architecture and red light illuminance control in the blood component separation system, the problems of misjudgment and signal distortion in single-point optical detection devices were solved, achieving high precision and stability in the blood component separation process.
Patent Information
- Application Number
- CN202611320667.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-28
- Publication Date
- 2026-11-03
AI Technical Summary
In existing blood component separation systems, single-point optical detection devices are susceptible to transient fluid disturbances, leading to logical misjudgments. Furthermore, the lack of quantitative constraints on light source illuminance causes signal extraction to fall into nonlinear distortion, affecting separation efficiency and purity.
Multiple detection units are arranged in a circular array along the outer periphery of the centrifuge mechanism to control the red light illuminance within the range of 800 lux to 1300 lux. Combined with multi-channel initial reference ratio calculation and adjacent bit logic and verification algorithm, interference signals are filtered out to accurately extract the interface parameters of red blood cells and plasma.
It improves the closed-loop control fault tolerance and anti-interference accuracy of the blood component separation process, ensuring the purity and efficiency of the separated products.
Smart Images

Figure CN123028993A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of automated control and photoelectric detection technology for medical devices, specifically to an interface detection system and method for blood component separation systems. Background Technology
[0002] In blood component separation procedures within the medical device field, centrifugation technology utilizes the density differences among different blood components to separate whole blood into a red blood cell layer, a white blood cell layer, and a plasma layer. To ensure collection purity and separation efficiency, the system relies on an interface detection device to monitor the red blood cell and plasma separation interface in real time, outputting feedback signals to adjust the operating parameters of the external fluid pump. Existing optical detection devices face the following two main technical challenges under actual clinical operating conditions:
[0003] First, single-point detection architectures are highly susceptible to logical misinterpretations due to transient fluid disturbances. Existing interface detection systems generally employ a topology of a single-point light source and a single-channel photodetector. Under the centrifugal force field of the high-speed rotation of the separation chamber, tiny free bubbles often precipitate inside the fluid pipeline, or fibrin agglomerates form in localized areas. When the physical optical path of the single-point detection beam is briefly blocked by bubbles or agglomerates, the light intensity signal at the receiver experiences a precipitous drop not caused by interface movement. The control system misinterprets this minute local disturbance as an overall rise in the red blood cell interface, leading to abnormal pump speed adjustment commands, disrupting the established fluid steady state, and causing cross-contamination of the separated products.
[0004] Second, the lack of quantitative constraints on light source illuminance leads to nonlinear distortion in signal extraction. Hemoglobin inside red blood cells has high absorption characteristics for specific wavelengths of red light. Current technology does not define the light intensity threshold range of the irradiated interface. If the light source output illuminance is set too low, the residual energy of red light photons after penetrating the high-concentration red blood cell layer and reaching the photodetector is often lower than the dark current noise floor of the sensor, causing the useful signal to be completely submerged by the background noise, and the system cannot extract interface change features. If the light source output illuminance is set too high, high-energy red light photons undergo multiple forward scattering inside the cell layer and penetrate the cell medium directly without being fully absorbed, resulting in a physical "absorption saturation" phenomenon. Under absorption saturation, the contrast of reflected light intensity between the red blood cell layer and the cell-free plasma layer narrows sharply, and the system experiences sensitivity degradation and interface misidentification when processing highly concentrated red blood cell samples. Summary of the Invention
[0005] To achieve the above objectives, the present invention provides the following technical solution:
[0006] This invention provides an interface detection method for a blood component separation system, comprising the following steps: the blood component separation system includes multiple detection units uniformly distributed in a circular array along the outer periphery of a centrifuge mechanism, and includes the following steps:
[0007] S100, Obtain initial reference value: Control the intensity of red light emitted by each detection unit to be within the range of 800 lux to 1300 lux, so that the red light passes through the transparent colorless medium located in the centrifuge mechanism and is reflected by the reflective surface, and collect the intensity of the reflected light received by each detection unit as the initial reference value corresponding to each detection unit.
[0008] S200, Real-time Reflection Signal Acquisition: In the working state of blood component separation, the intensity of the red light emitted by each detection unit is controlled within the range of 800 lux to 1300 lux. The red light passes through the separated blood components and is reflected by the reflective surface. The intensity of the reflected light received by each detection unit in the working state is acquired in real time.
[0009] S300, Calculate light intensity characteristic parameters: For each of the detection units, calculate the ratio or difference between the real-time collected reflected light intensity of the working state and the corresponding initial reference value;
[0010] S400, Spatial Array Interface Determination: Continuously monitor the ratio or difference of each of the detection units, and determine that the interface between red blood cells and plasma passes through the location of the adjacent detection unit if and only if a mutation satisfying a preset drop condition is detected simultaneously in the ratio or difference of at least two consecutive adjacent detection units.
[0011] Furthermore, the detection unit includes a light-emitting diode;
[0012] The step of controlling the emission intensity of red light from each of the detection units to be within the range of 800 lux to 1300 lux specifically includes:
[0013] The light-emitting diode is controlled to emit light so that the illuminance of the red light with a center wavelength of 620nm to 750nm at the illumination window of the centrifugal mechanism is kept constant within the range of 800 lux to 1300 lux.
[0014] Furthermore, the step of obtaining the initial reference value is performed during the pre-rinsing stage of the pipeline of the blood component separation system, and the transparent colorless medium is physiological saline pre-injected into the centrifugation mechanism;
[0015] The reflective surface is disposed on the inner wall of the cavity of the centrifuge mechanism, and the reflective surface is a white diffuse reflective surface with a reflectivity greater than or equal to 90%.
[0016] Furthermore, after the step of real-time acquisition of the intensity of reflected light received by each of the detection units in the working state, the method further includes a signal preprocessing step:
[0017] The intensity of reflected light in the working state collected in real time by each of the detection units is independently processed by moving average filtering to remove high-frequency spike pulses caused by transient disturbances in blood components;
[0018] The smoothed light intensity value obtained after the moving average filtering process is used as the reflected light intensity of the working state adopted in step S300.
[0019] Furthermore, the step of real-time acquisition of the intensity of reflected light received by each of the detection units in the working state specifically refers to: during the entire working cycle of the centrifuge mechanism rotating to perform centrifugal separation, each of the detection units continuously performs synchronous sampling at the same preset sampling frequency to obtain the time-series dynamic data stream corresponding to each of the detection units; wherein, the preset sampling frequency is greater than or equal to 100Hz.
[0020] Furthermore, for each of the detection units, the step of calculating the ratio or difference between the intensity of the reflected light in the working state acquired in real time and the corresponding initial reference value specifically employs a ratio calculation method, including:
[0021] Divide the intensity of the reflected light in the working state collected in real time by the corresponding initial reference value to obtain the independent normalization ratio value of each detection unit;
[0022] The normalized ratio is used to characterize the change in red blood cell concentration within the corresponding irradiation window, so as to adaptively compensate for the photoaging of the light-emitting diodes of each detection unit and the individual optical differences of the centrifugation mechanism.
[0023] Furthermore, after calculating the ratio or difference of each of the detection units, and before executing step S400, the method further includes a dynamic feature extraction step: within a preset time window, the time change rate of the ratio or difference of each of the detection units is calculated in real time to obtain the dynamic data of the first derivative corresponding to each of the detection units; the time change rate is used in step S400 to assist in determining whether the mutation is a real interface movement.
[0024] Furthermore, the number of the plurality of detection units is seven, and they are distributed in a circumferential array with equal intervals of 51.4 degrees along the outer periphery of the centrifugation mechanism; the step of determining that the interface between red blood cells and plasma passes through the location of the adjacent detection unit when a sudden change satisfying a preset drop condition is detected simultaneously in the ratio or difference of at least two consecutive adjacent detection units is detected, specifically includes the following determination logic: a preset interface jump threshold is set, and the sudden change is confirmed and an interface position signal is output when the ratio of at least two consecutive adjacent detection units is lower than the interface jump threshold; if only a single detection unit is detected to have a ratio lower than the interface jump threshold, the system determines that it is a local disturbance caused by bubbles or fibrin clots, and filters and ignores the abnormal signal.
[0025] Furthermore, after determining the location of the red blood cell-plasma interface passing through the adjacent detection unit, the method further includes an automated closed-loop control step:
[0026] Based on the real-time acquired interface position feedback, closed-loop control commands are continuously generated and output; the closed-loop control commands are used to dynamically adjust the operating parameters of the relevant drive pumps in the blood component separation system to maintain the interface within the preset safe separation tolerance range.
[0027] On the other hand, an interface detection system for a blood component separation system includes: a plurality of detection units, each of which is uniformly distributed in a circular array along the outer periphery of the centrifuge mechanism of the blood component separation system; each of the detection units includes a light-emitting diode that emits red light and a photodetector for receiving reflected light signals.
[0028] The main control unit is communicatively connected to each of the aforementioned detection units;
[0029] The main control unit is configured to execute the following control logic:
[0030] Drive and light intensity control: Drive the light-emitting diodes of each of the detection units to emit light, and keep the illuminance of the red light at the illumination window of the centrifugal mechanism constant within the range of 800 lux to 1300 lux;
[0031] Signal acquisition: The initial reference value obtained by the red light passing through the transparent colorless medium in the centrifuge mechanism and being reflected is acquired respectively, and the working state reflected light intensity obtained by the red light passing through the separated blood components and being reflected in real time is acquired.
[0032] Feature calculation: For each of the detection units, calculate the ratio or difference between the reflected light intensity of the corresponding working state and the initial reference value;
[0033] Array space voting: continuously monitor the ratio or difference of each of the detection units, and determine the location of the red blood cell-plasma interface passing through the adjacent detection unit only when a mutation satisfying a preset drop condition is detected simultaneously in the ratio or difference of at least two consecutive adjacent detection units.
[0034] Beneficial effects
[0035] This invention constructs a multi-channel photoelectric detection architecture with a circular array distribution and strictly limits the red light illuminance of the irradiated surface to maintain within the optically unsaturated high-contrast linear response range of 800 lux to 1300 lux. Combined with multi-channel initial reference ratio calculation and adjacent bit logic and verification algorithm, it can accurately extract the physical displacement parameters of the red blood cell and plasma interface, effectively filter out single-point random optical occlusion interference caused by free air bubbles or local coagulation in the pipeline, and improve the fault tolerance and anti-interference accuracy of closed-loop control in the automated separation of blood components. Attached Figure Description
[0036] Figure 1 This is a flowchart of the control method of the present invention;
[0037] Figure 2 This is a nonlinear response curve of red light illuminance and light intensity signal contrast according to the present invention.
[0038] Figure 3 This is a graph showing the time relationship between the interface state and the intensity of reflected light in this invention.
[0039] Figure 4 The initial reference values for the 7 channels under the transparent medium (physiological saline) conditions of this invention. Sampling data charts;
[0040] Figure 5 This is a data table showing the mapping between the sampling signal and the filtering process of channel #1 under the working conditions of this invention;
[0041] Figure 6 This is a mapping data table for calculating the light intensity characteristic parameters of channel #1 in this invention;
[0042] Figure 7 This is a mapping data table for spatial voting and disturbance rejection fault-tolerant logic operations of the 7-channel array of the present invention. Detailed Implementation
[0043] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.
[0044] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but includes other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0045] The present invention will now be described in further detail with reference to the accompanying drawings:
[0046] Example:
[0047] like Figures 1 to 7 As shown, an interface detection method for a blood component separation system is described.
[0048] The blood component separation system includes multiple detection units evenly distributed in a circular array along the outer periphery of the centrifuge mechanism, and includes the following steps:
[0049] S100, Obtain initial reference value: Control the intensity of red light emitted by each detection unit within the range of 800 lux to 1300 lux, so that the red light passes through the transparent colorless medium located in the centrifuge mechanism and is reflected by the reflective surface. Collect the intensity of the reflected light received by each detection unit as the initial reference value for each detection unit.
[0050] S200, real-time reflection signal acquisition: In the working state of blood component separation, the intensity of red light emitted by each detection unit is controlled within the range of 800 lux to 1300 lux. The red light passes through the separated blood components and is reflected by the reflective surface. The intensity of the reflected light received by each detection unit in the working state is acquired in real time.
[0051] S300, calculate light intensity characteristic parameters: for each detection unit, calculate the ratio or difference between the real-time collected working state reflected light intensity and the corresponding initial reference value;
[0052] S400, Spatial Array Interface Determination: Continuously monitor the ratio or difference of each detection unit, and determine the location of the red blood cell and plasma interface through the adjacent detection unit only when a mutation that satisfies the preset drop condition is detected simultaneously in the ratio or difference of at least two consecutive adjacent detection units.
[0053] Furthermore, the specific implementation process of step S100 is as follows:
[0054] Step S100 includes the specific operation process of light source driving, optical path transmission, and photoelectric signal acquisition and storage.
[0055] The system's main control unit is in pipeline pre-flushing mode. The main control unit outputs digital pulse width modulation (PWM) commands to the underlying constant current source drive circuit via its internal microprocessor, driving the LEDs inside multiple detection units arranged in a circular array along the outer periphery of the centrifugal mechanism to illuminate. The LEDs are gallium arsenide polymer light emitters with a center wavelength limited to the range of 620nm to 750nm. The main control unit adjusts the output duty cycle of the constant current source drive circuit to precisely control the forward current applied across the LEDs within a preset range, thus controlling the LEDs to emit red light.
[0056] In the red light emission state, the main control unit monitors the output power of the light-emitting diode in real time through the photoelectric feedback loop, and strictly controls the physical quantity of the actual illuminance at the outer surface of the irradiation window of the centrifugal mechanism to be within the range of 800 lux to 1300 lux.
[0057] During step S100, the separation chamber inside the centrifuge mechanism is completely filled with a pre-injected transparent and colorless medium, specifically a 0.9% sodium chloride saline solution. After being emitted from the LED, the red light beam first penetrates the polycarbonate transparent shell of the irradiation window and then enters the separation chamber filled with saline. The beam propagates along a straight optical path in the saline solution and is projected onto a reflective surface located on the inner wall of the centrifuge mechanism's chamber. The reflective surface is made of a material with a polytetrafluoroethylene diffuse reflection coating, and its solid reflectivity is greater than or equal to 90%. After the red light beam strikes the reflective surface, diffuse reflection occurs. The diffusely reflected beam returns along the reverse path of the original optical path, penetrates the saline solution and the irradiation window a second time, and is projected into the receiving end of the detection unit.
[0058] The receiver of the detection unit is equipped with a silicon photodiode. Upon receiving red photons that have penetrated the transparent, colorless medium and returned, the silicon photodiode undergoes the photoelectric effect, outputting a primary analog current signal at the microampere level. This primary analog current signal is input to a transimpedance amplifier (TIA) inside the detection unit, where it is converted into a millivolt-level analog voltage signal. An analog-to-digital converter (ADC) chip discretizes the analog voltage signal at a set sampling frequency, outputting the corresponding digital quantized value.
[0059] The main control unit reads the digital quantization values corresponding to each probe unit via a serial communication bus. For a probe unit array containing probe units numbered from 1 to N, the main control unit sequentially reads and locks the digital quantization value returned by each probe unit in a saline-filled state, defining it as its independent initial reference value, denoted as $N$. The main control unit writes the acquired initial reference values into the specific physical address range of its internal non-volatile electrically erasable read-only memory (EEPROM) to complete the static calibration of the reference data.
[0060] Specific experimental scenario: In the experimental verification of the pre-rinsing stage of the simulated centrifugal separation pipeline in vitro, the system operating environment temperature was set to 25 degrees Celsius. The main shaft of the centrifuge mechanism was in a non-rotationally locked state, and the nominal volume of 0.9% physiological saline was injected into the separation chamber to purge air. The control unit instructed the multi-channel constant current source to output a constant drive current of 20 mA to seven detection units equidistantly distributed along the 360-degree outer circumference. The red light illuminance was measured at the surface of the illumination window at a distance of 3 mm using a high-precision illuminance meter, and maintained at a constant value of 1050 lux.
[0061] The optical signal is reflected back by a white diffuse reflective surface with 92% reflectivity, received by a silicon photodetector with a response wavelength covering 660nm inside the detection unit, and then sent to a 12-bit ADC (with a reference voltage set to 3.3V) for digital conversion. After continuously sampling for 100 cycles and taking the arithmetic average, the experimental data of the initial reference values of the 7 channels actually entered into the main control unit register are as follows: Figure 4 As shown.
[0062] The main control unit will Figure 4 The discrete values in the range of 2790 to 2866, which are quantized, are allocated and locked in the internal flash memory address block from 0x00A0 to 0x00AC, and are provided for use by the feature calculation module during subsequent blood centrifugation.
[0063] Furthermore, the specific implementation process of step S200 is as follows:
[0064] The blood component separation system enters the centrifugal separation working state. The main shaft of the centrifuge mechanism accelerates and maintains a constant high speed range according to the set operating instructions. The main control unit continuously outputs control commands to the underlying constant current source drive circuit to maintain the output electrical parameters of the multi-channel constant current source constant, ensuring that the light-emitting diodes inside all detection units continuously emit red light with a center wavelength in the range of 620nm to 750nm, and strictly constraining the actual illuminance at the surface of the illumination window to be maintained within the range of 800 lux to 1300 lux.
[0065] External whole blood is pumped into a high-speed rotating separation chamber. Under the influence of a centrifugal force, the whole blood separates into plasma, a white blood cell layer, and a red blood cell layer. A red light beam emitted by a self-emitting diode penetrates the illumination window and is projected into the separation chamber. At this point, the optical path is completely filled with the separated blood components. The red light beam propagates in a straight line within the blood component medium. Hemoglobin molecules inside the red blood cells absorb a large amount of the specific wavelength photon energy carried by the penetrating beam, resulting in physical attenuation of light intensity. The red light beam, having undergone energy attenuation, strikes the diffuse reflection surface on the inner wall of the separation chamber, causing reflection and refraction. The reflected beam then penetrates the blood component medium and the illumination window a second time, projecting into the receiving end of the detection unit.
[0066] The silicon photodiode inside the detection unit receives the reflected light beam, undergoes photoelectric conversion, and outputs a secondary analog current signal carrying information about blood component concentration. A transimpedance amplifier synchronously converts the secondary analog current signal into a secondary analog voltage signal. The main control unit, equipped with an internal analog-to-digital converter (ADC), performs high-frequency analog-to-digital conversion on the continuously input secondary analog voltage signal at a fixed sampling frequency greater than or equal to 100Hz, outputting a discrete time-series dynamic digital stream.
[0067] The main control unit periodically reads the time-series dynamic digital streams independently generated by each probe unit through a hardware interrupt mechanism. For the read raw digital stream data, the main control unit calls the internal arithmetic logic unit (ALU) to perform a moving average filtering operation in the digital domain. The main control unit allocates a first-in-first-out (FIFO) data buffer queue with a depth of 10 for each probe unit numbered 1 to N in the random access memory (RAM).
[0068] When the clock ticks trigger the writing of a new digital quantized value, the first-in-first-out (FIFO) data buffer queue automatically pops the oldest value from the beginning of the queue and pushes in the newly acquired value from the end. The arithmetic logic unit immediately performs an accumulation operation on the 10 digital quantized values currently in the buffer queue, divides the sum by 10, and outputs a single arithmetic average. The main control unit directly defines and stores the arithmetic average output after the moving average filtering operation as the real-time reflected light intensity of the corresponding detection unit, denoted as... .
[0069] Specific experimental scenario: In an in vitro whole blood centrifugation experiment, human whole blood with an initial hematocrit (HCT) of 42% was pumped into the system. The centrifuge spindle speed was kept constant at 3000 rpm. The main control unit was configured with an ADC sampling frequency of 200 Hz (i.e., a sampling period of 5 milliseconds). The LED output illuminance of the 7 detection units was maintained at a constant value of 1050 lux as calibrated in step S100.
[0070] During the physical time window from 1200 ms to 1245 ms of continuous centrifugation, due to the pressure-induced precipitation of tiny free bubbles in the fluid inside the centrifuge tubing, a transient disturbance without cell obstruction briefly occurred within the physical optical path of channel #1. The main control unit recorded the original ADC sampled digital stream of channel #1 during this time window, as well as the reflected light intensity after synchronously performing a 10-point moving average filtering operation. Data, detailed mapping relationships as follows Figure 5 As shown.
[0071] The main control unit will Figure 5 The rightmost column, after undergoing filtering and smoothing, displays the discrete numerical stream (1514, 1514, 1513, 1607...), representing the real-time reflected light intensity of channel #1. It is then transferred to the next level of feature calculation memory address space.
[0072] Furthermore, the specific implementation process of step S300 is as follows:
[0073] Step S300 includes the specific operation process of calculating the normalized ratio of light intensity characteristic parameters and extracting the first derivative in the time domain.
[0074] The arithmetic logic unit (ALU) inside the main control unit receives the reflected light intensity data stream of each detection unit's real-time operating status from the output of step S200. The main control unit issues a memory read command to access the initial reference value data of the corresponding detection unit locked in the non-volatile electrically erasable read-only memory (EEPROM). For physical channel number... (in The detection unit and the main control unit record the intensity of the reflected light during its transient operating state as follows: Record its initial baseline value as .
[0075] The arithmetic logic unit (ALU) is internally configured with a floating-point divider logic array. The main control unit will read... Numerical and The value is pushed into the arithmetic register. The floating-point divider executes the division instruction, and... Divide by Output dimensionless normalized ratio data, denoted as The specific algebraic expression is as follows:
[0076] Normalized ratio data The mapping corresponds to the physical volume ratio information of red blood cells and plasma within the irradiation window. The main control unit outputs normalized ratio data. Stored in a contiguous block of physical addresses within the internal cache.
[0077] During the synchronous clock cycle for performing the normalized ratio calculation, the differential arithmetic module configured inside the main control unit reads the normalized ratio data under consecutive timestamps from the cache. The adjacent sampling time interval is set to... Current sampling time The ratio is Previous sampling time The ratio is The differential operator performs the difference derivative operation and outputs the time rate of change (first derivative) of the normalized ratio of the corresponding channel, denoted as... The specific algebraic expression is as follows:
[0078] The main control unit will use the time change rate data Stored in a separate data register, it is provided to step S400 as an auxiliary logic judgment parameter for identifying the movement characteristics of the physical interface.
[0079] Specific experimental implementation scenario:
[0080] In the previously established in vitro whole blood centrifugation separation experimental model, step S100 is called. Figure 4 Initial reference value of channel #1 selected The microprocessor clock speed of the main control unit is set to 72MHz, and the floating-point unit (FPU) is enabled. Sampling time interval is set. millisecond.
[0081] During the steady-state separation phase of continuous centrifugation (within the physical time window of 3000 ms to 3025 ms after the centrifuge is powered on), the physical interface of the red blood cell layer inside the centrifuge chamber moves upward in a controlled manner, and the red blood cell layer gradually covers the physical area of the irradiation window corresponding to channel #1. The main control unit synchronously captures the reflected light intensity of the working status input in real time from channel #1. The normalized ratio is continuously output by the floating-point arithmetic unit. With time rate of change The computational mapping data synchronously recorded and printed by the microprocessor's internal Universal Asynchronous Receiver / Transmitter (UART) port is as follows: Figure 6 As shown ).
[0082] The main control unit will Figure 6 The continuous normalized ratio data (in the range of 0.998 to 0.531) obtained by medium quantization and the corresponding time change rate data are sent to the core memory area of the system logic judgment, triggering the array space voting comparison instruction of the subsequent step S400.
[0083] Furthermore, the specific implementation process of step S400 is as follows:
[0084] Step S400 includes the specific operation process of multi-channel spatial topology logic voting, single-point physical anti-disturbance filtering, and closed-loop control command output.
[0085] The main control unit is internally configured with a Boolean logic comparator and a status register array. The main control unit synchronously reads all data with the same absolute timestamp from its internal cache. Normalized ratio data of each detection unit to and the corresponding time rate of change data to The main control unit pre-writes a light intensity drop threshold defining the physical boundary of the red blood cell layer into non-volatile memory, setting the value to 0.60. Simultaneously, a negative mutation rate threshold characterizing the transient ramp-up slope of the physical interface is preset, setting the value to... .
[0086] The Boolean logic comparator concurrently executes threshold comparison instructions. For physical channel number... ( It is a positive integer, and The detection unit of ) if and only if the input is The value is less than the light intensity drop threshold, and the input is synchronized. When the value is less than the negative mutation rate threshold, the Boolean logic comparator writes a logic high-level identifier "1" to the address bit of the corresponding channel's status register; if neither of the above two conditions is met simultaneously, the Boolean logic comparator writes a logic low-level identifier "0" to the address bit of the corresponding channel's status register. The status identifiers of each channel are reassembled in memory into a single identifier of length [length missing]. A sequence of real-time state vectors for each bit.
[0087] The microprocessor inside the main control unit reads the real-time state vector sequence and performs an adjacent bit logical AND check operation based on a ring physical topology. Based on the physical structure characteristics of the detection units arranged in a 360-degree closed equiangular ring along the outer periphery of the centrifugal mechanism, the microprocessor, during program pointer addressing, will... The first and second positions are defined as physically adjacent spatial positions.
[0088] If the microprocessor detects only a single isolated bit with a logic high level identifier "1" in the real-time state vector sequence, the microprocessor determines that random optical occlusion has occurred on the surface of the single-point illumination window at that physical space coordinate due to tiny free air bubbles in the pipeline or physical agglomerates of fibrin. The main control unit will trigger a single-point anomaly fault-tolerant filtering interrupt, forcibly overwriting the logic low level identifier "0" to the single isolated bit, intercepting and discarding the abnormal trigger data, and maintaining the original centrifugal pump drive level constant.
[0089] The microprocessor confirms that the horizontally distributed, ring-shaped physical interface between red blood cells and plasma within the separation chamber has substantially crossed the concentric horizontal height coordinates of the aforementioned adjacent detection units if and only if the microprocessor retrieves at least two consecutive adjacent bits in the real-time state vector sequence that simultaneously exhibit a logic high level identifier "1". The microprocessor then sends a valid interface position feedback flag to the external bus.
[0090] After receiving the valid interface position feedback flag, the digital-to-analog converter chip outputs a speed-regulating analog voltage. This voltage is transmitted via a wiring harness to the pins of the whole blood and plasma drive pumps in the centrifuge mechanism, triggering a downward adjustment of the current commutation frequency in the stator coils of the drive pumps. This reduces the fluid inflow rate and maintains the blood separation interface within the safe separation tolerance range.
[0091] Specific experimental implementation scenario:
[0092] Continuing the in vitro whole blood centrifugation separation experimental environment in step S300, the system is equipped with a total of 7 detection units. Within the physical time window of 4500 ms to 4520 ms, the physical conditions inside the centrifuge tubing exhibit highly dynamic changes. At absolute timestamp 4505 ms, a free air bubble is adsorbed at the irradiation window of channel #4; at absolute timestamp 4515 ms, the actual red blood cell layer interface rises with the centrifugal pressure of the fluid, simultaneously covering the irradiation windows of channels #1 and #2.
[0093] During this period, the main control unit synchronously captures the feature parameters of 7 channels and generates a real-time state vector through a Boolean logic comparator. The result of the microprocessor performing the ring space topology logic verification is as follows: Figure 7 As shown:
[0094] microprocessor according to Figure 7 The data status at 4505 milliseconds successfully removed the single-point light intensity drop caused by the bubble in channel #4 from the closed-loop control link; and at 4515 milliseconds, it accurately captured the real interface position signal of the synchronous drop of channels #1 and #2, and immediately sent a PWM speed regulation control word with reduced duty cycle to the plasma drive pump end, realizing precise interception of the physical interface and automated closed-loop control.
[0095] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. An interface detection method for a blood component separation system, characterized in that, The blood component separation system includes multiple detection units evenly distributed in a circular array along the outer periphery of the centrifuge mechanism, characterized by the following steps: S100, Obtain initial reference value: Control the intensity of red light emitted by each detection unit to be within the range of 800 lux to 1300 lux, so that the red light passes through the transparent colorless medium located in the centrifuge mechanism and is reflected by the reflective surface, and collect the intensity of the reflected light received by each detection unit as the initial reference value corresponding to each detection unit. S200, Real-time Reflection Signal Acquisition: In the working state of blood component separation, the intensity of the red light emitted by each detection unit is controlled within the range of 800 lux to 1300 lux. The red light passes through the separated blood components and is reflected by the reflective surface. The intensity of the reflected light received by each detection unit in the working state is acquired in real time. S300, Calculate light intensity characteristic parameters: For each of the detection units, calculate the ratio or difference between the real-time collected reflected light intensity of the working state and the corresponding initial reference value; S400, Spatial Array Interface Determination: Continuously monitor the ratio or difference of each of the detection units, and determine that the interface between red blood cells and plasma passes through the location of the adjacent detection unit if and only if a mutation satisfying a preset drop condition is detected simultaneously in the ratio or difference of at least two consecutive adjacent detection units.
2. The interface detection method for a blood component separation system according to claim 1, characterized in that, The detection unit includes a light-emitting diode; The step of controlling the emission intensity of red light from each of the detection units to be within the range of 800 lux to 1300 lux specifically includes: The light-emitting diode is controlled to emit light so that the illuminance of the red light with a center wavelength of 620nm to 750nm at the illumination window of the centrifugal mechanism is kept constant within the range of 800 lux to 1300 lux.
3. The interface detection method for a blood component separation system according to claim 2, characterized in that, The step of obtaining the initial reference value is performed during the pre-flushing stage of the pipeline of the blood component separation system, and the transparent colorless medium is physiological saline pre-injected into the centrifugation mechanism. The reflective surface is disposed on the inner wall of the cavity of the centrifuge mechanism, and the reflective surface is a white diffuse reflective surface with a reflectivity greater than or equal to 90%.
4. The interface detection method for a blood component separation system according to claim 3, characterized in that, After the step of real-time acquisition of the intensity of reflected light from the working state received by each of the detection units, the method further includes a signal preprocessing step: The intensity of reflected light in the working state collected in real time by each of the detection units is independently processed by moving average filtering to remove high-frequency spike pulses caused by transient disturbances in blood components; The smoothed light intensity value obtained after the moving average filtering process is used as the reflected light intensity of the working state adopted in step S300.
5. The interface detection method for a blood component separation system according to claim 4, characterized in that, The step of real-time acquisition of the reflected light intensity of the working state received by each of the detection units specifically refers to: during the entire working cycle of the centrifuge mechanism rotating to perform centrifugal separation, each of the detection units continuously performs synchronous sampling at the same preset sampling frequency to obtain the time series dynamic data stream corresponding to each of the detection units; wherein, the preset sampling frequency is greater than or equal to 100Hz.
6. The interface detection method for a blood component separation system according to claim 5, characterized in that, For each of the aforementioned detection units, the step of calculating the ratio or difference between the intensity of the reflected light in the working state acquired in real time and the corresponding initial reference value is specifically implemented using a ratio calculation method, including: Divide the intensity of the reflected light in the working state collected in real time by the corresponding initial reference value to obtain the independent normalization ratio value of each detection unit; The normalized ratio is used to characterize the change in red blood cell concentration within the corresponding irradiation window, so as to adaptively compensate for the photoaging of the light-emitting diodes of each detection unit and the individual optical differences of the centrifugation mechanism.
7. The interface detection method for a blood component separation system according to claim 6, characterized in that, After calculating the ratio or difference of each of the detection units, and before executing step S400, the method further includes a dynamic feature extraction step: within a preset time window, the time change rate of the ratio or difference of each of the detection units is calculated in real time to obtain the dynamic data of the first derivative corresponding to each of the detection units; the time change rate is used in step S400 to assist in determining whether the mutation is a real interface movement.
8. The interface detection method for a blood component separation system according to claim 7, characterized in that, The number of the plurality of detection units is seven, and they are distributed in a circumferential array with equal intervals of 51.4 degrees along the outer periphery of the centrifugation mechanism. The step of determining that the interface between red blood cells and plasma has passed the location of the adjacent detection unit when a sudden change that satisfies a preset drop condition is detected simultaneously in the ratio or difference of at least two consecutive adjacent detection units is detected, specifically includes the following determination logic: a preset interface jump threshold is set, and the sudden change is confirmed and an interface position signal is output when the ratio of at least two consecutive adjacent detection units is lower than the interface jump threshold. If only a single detection unit is detected to have a ratio lower than the interface jump threshold, the system determines that it is a local disturbance caused by bubbles or fibrin clots, and filters and ignores the abnormal signal.
9. The interface detection method for a blood component separation system according to claim 8, characterized in that, After determining the location of the red blood cell-plasma interface passing through the adjacent detection unit, the method further includes an automated closed-loop control step: Based on the real-time acquired interface position feedback, closed-loop control commands are continuously generated and output; the closed-loop control commands are used to dynamically adjust the operating parameters of the relevant drive pumps in the blood component separation system to maintain the interface within the preset safe separation tolerance range.
10. An interface detection system for a blood component separation system, applied to the interface detection method for a blood component separation system according to any one of claims 1-9, characterized in that, include: Multiple detection units are uniformly distributed in a circular array along the outer periphery of the centrifuge mechanism of the blood component separation system; each detection unit includes a light-emitting diode that emits red light and a photodetector for receiving reflected light signals. The main control unit is communicatively connected to each of the aforementioned detection units; The main control unit is configured to execute the following control logic: Drive and light intensity control: Drive the light-emitting diodes of each of the detection units to emit light, and keep the illuminance of the red light at the illumination window of the centrifugal mechanism constant within the range of 800 lux to 1300 lux; Signal acquisition: The initial reference value obtained by the red light passing through the transparent colorless medium in the centrifuge mechanism and being reflected is acquired respectively, and the working state reflected light intensity obtained by the red light passing through the separated blood components and being reflected in real time is acquired. Feature calculation: For each of the detection units, calculate the ratio or difference between the reflected light intensity of the corresponding working state and the initial reference value; Array space voting: continuously monitor the ratio or difference of each of the detection units, and determine the location of the red blood cell-plasma interface passing through the adjacent detection unit only when a mutation satisfying a preset drop condition is detected simultaneously in the ratio or difference of at least two consecutive adjacent detection units.
