Intelligent blood sampling management method and management system

Abstract

The invention relates to an intelligent blood sampling management method and management system, and belongs to the technical field of blood sampling inspection, and the method comprises the steps: responding to a sampling initiation instruction, and obtaining basic information; in response to the sampling execution instruction, basic information is provided for a sampling person, and then a sampled person sample collected by the sampling person is bound with the basic information to form a binding relationship; according to the binding relation and a preset classification basis, the samples of the sampled person are sorted and conveyed to corresponding detection analysis points; according to the sampling items in the binding relationship, executing a corresponding detection and analysis process on the sampled person sample transmitted to the detection and analysis point to obtain detection data and an analysis result; and generating a detection report according to the detection data, the analysis result and the corresponding basic information, and sending the detection report to a medical care terminal which triggers a sampling initiation instruction. The method has the effects of reducing the human intervention amount of the blood sampling inspection process and improving the efficiency of the blood sampling inspection process.

Description

Technical Field

[0001] This application relates to the field of blood collection and testing technology, and in particular to an intelligent blood collection management method and management system. Background Technology

[0002] In existing medical testing processes, especially in large-scale blood collection scenarios (such as physical examination centers and outpatient departments), the management and analysis of samples often involve many manual operations, resulting in low overall efficiency and a high risk of errors.

[0003] The traditional process is roughly as follows: The doctor issues a test request form (which may be paper or electronic). The person being sampled takes the request form to the blood collection station. The nurse manually verifies the person's identity and the test items before drawing blood and attaching a barcode containing the person's information to the sample container (such as a vacuum blood collection tube). Afterward, the samples are collected centrally and manually sorted, separating samples for different tests (such as biochemistry, immunology, and complete blood count) before transporting them to different laboratory departments.

[0004] The traditional process described above has the following problems: sample sorting relies entirely on sorting personnel visually identifying labels or barcodes on test tubes, which is not only slow but also prone to errors due to fatigue during peak periods. Furthermore, sorting personnel need to possess certain medical testing knowledge to identify different items, resulting in high labor costs; therefore, improvements are needed. Summary of the Invention

[0005] To reduce human intervention in the blood collection and testing process and improve its efficiency, this application provides an intelligent blood collection management method and system.

[0006] Firstly, this application provides an intelligent blood collection management method, including: In response to a sampling initiation command triggered by a medical terminal, basic information is obtained, including at least the identity information of the sampled person and the sampling items. In response to a sampling execution command triggered by a sampling personnel, the basic information is provided to the sampling personnel, and then the sample collected by the sampling personnel is bound to the basic information to form a binding relationship; Based on the binding relationship and the preset classification criteria, the sampled individuals are sorted and transported to the corresponding detection and analysis points; At the detection and analysis point, according to the sampling items in the binding relationship, the corresponding detection and analysis process is performed on the sample of the subject delivered to the detection and analysis point to obtain detection data and analysis results; Based on the test data, analysis results, and corresponding basic information, a test report is generated and sent to the medical terminal that triggered the sampling initiation command, so that the medical staff at the medical terminal can receive the test report.

[0007] By adopting the above technical solution, and by providing basic information in response to sampling execution instructions, a basis for secondary verification is provided to sampling personnel, reducing the risk of human error. More importantly, the automatic control of sorting equipment based on binding relationships and preset rules replaces traditional manual identification and sorting. This not only increases sorting speed by orders of magnitude but also achieves near 100% accuracy, significantly reducing labor costs and the human error rate. Intelligent sorting directly delivers samples to the corresponding testing and analysis points, achieving precise allocation of testing tasks. This avoids the hassle of repeated data entry and sample searching for testing personnel, and also avoids sample backlog at incorrect testing points and repeated transportation, allowing testing equipment and analysts to focus more on the testing itself, improving overall testing throughput and efficiency. Furthermore, the automatic acquisition of testing data and analysis results, along with automatic matching and report synthesis with pre-bound basic information (such as the sampled person's ID, name, and testing items), generates a structured testing report, which is automatically pushed to the ordering medical terminal. This process requires no manual intervention, greatly shortening the reporting cycle, enabling medical personnel and sampled persons to obtain diagnostic evidence more quickly, and improving diagnostic efficiency.

[0008] Optionally, the step of sorting the sampled individuals and transporting them to the corresponding detection and analysis points according to the binding relationship and preset classification criteria includes: Based on the binding relationship and the preset classification criteria, analyze the target detection and analysis points that the sampled person's sample needs to access; If the target detection and analysis point is unique, the sampled person's sample will be sorted and transported to the corresponding target detection and analysis point; If the target detection and analysis point is not unique, then based on the preset path planning strategy, an orderly flow path is planned for the corresponding sample to access all target detection and analysis points corresponding to the sample. According to the flow path, the sample is sequentially transported to the target detection and analysis points included in the flow path.

[0009] By adopting the above technical solutions, the system's differentiated processing logic for single-item and multi-item samples was clarified. For the latter, by generating and executing a "transfer path," the fully automated and orderly transfer of the same physical sample between multiple testing devices was achieved for the first time in the field of blood collection management, solving the pain point of manual sorting, transportation, or separate management of multi-item samples in traditional processes. By introducing the concept of "transfer path," a core architectural support was provided for subsequent intelligent path planning based on real-time status, load balancing, and other strategies, enabling the entire system to move from "automated sorting" to "intelligent scheduling."

[0010] Optionally, the step of planning an ordered flow path for the corresponding sampled subject sample to access all target detection and analysis points based on a preset path planning strategy includes: The load status information of each target detection and analysis point corresponding to the sampled person is acquired in real time. The load status information includes at least the estimated waiting time for the target detection and analysis point to add new samples. With the goal of minimizing the overall completion time of the sampled subjects, the waiting time of new samples is estimated based on the target detection and analysis points, and the optimal order for the sampled subjects to access all their target detection and analysis points is determined to generate the flow path.

[0011] By adopting the above technical solution, introducing "real-time load status information" and "estimated waiting time," and specifying the optimization objective as "minimizing the overall completion time," this solution upgrades path planning from static, fixed rules to dynamic, adaptive intelligent decision-making. The system can calculate the currently globally optimal flow sequence for each multi-project sample based on the laboratory's real-time workload, significantly shortening the total turnaround time of samples in the system and improving overall throughput. It can automatically guide samples to avoid busy equipment and select idle or short-waiting-time equipment, effectively balancing the workload of each testing and analysis point, avoiding the waste of resources caused by some equipment being congested while others are idle, and optimizing resource utilization efficiency at the system level.

[0012] Optionally, the method further includes: The system records the real-time status and timestamps of the sampled object as it flows along the transfer path. Based on the timestamps of the real-time status, the transfer path, and the real-time load status information of each target detection and analysis point included in the transfer path, the system predicts the completion time of the detection report for the sampled object. It then generates status update information in real time, containing the latest real-time status of the sampled object and the latest predicted completion time of the detection report, and sends the status update information to the sampled object's terminal. The real-time status includes at least "sorted" and "arrived at the detection and analysis point."

[0013] By adopting the above technical solution, management information (sample status) is transformed into valuable information (real-time status, estimated report time) for the sampled individuals, and proactively pushed to their terminals, providing unprecedented transparency and predictability in the testing process. This not only improves efficiency but also creates a new communication and service model. Proactive status updates and accurate time predictions significantly enhance sampled individuals' satisfaction and trust in the service, possessing significant non-technical commercial value.

[0014] Optionally, the method further includes: Collect and analyze historical sampling data to identify common sampling item combinations, which refer to combinations of sampling items that appear simultaneously and multiple times in the same basic information. Based on the common sampling item combinations and their corresponding target detection and analysis points, adjustment suggestions are generated to optimize the physical layout of the detection and analysis points. The core principle of the adjustment suggestions is to arrange multiple target detection and analysis points corresponding to the common sampling item combinations in close proximity in space.

[0015] By adopting the above technical solutions and utilizing the big data generated during operation, the physical layout optimization of hardware facilities can be guided in reverse. By analyzing the combination of high-frequency co-occurring sampling items and proposing suggestions for placing their corresponding devices close together, the physical transmission distance of high-frequency combined samples can be fundamentally shortened, resulting in a permanent efficiency improvement.

[0016] Optionally, the sampling items in the basic information obtained when responding to the sampling initiation command are the original sampling items; The method further includes: Before generating the test report, the test data, analysis results and corresponding basic information are input into a preset clinical decision model, and the clinical decision model determines whether a reflexive test needs to be triggered. If it is determined that reflective detection needs to be triggered, the sampling item corresponding to the required reflective detection is taken as a secondary sampling item. A binding relationship is established between the secondary sampling item and the sampled subject to update the binding relationship of the sampled subject. Then, according to the updated binding relationship and the preset classification criteria, the sampled subject is sorted and transported to the corresponding detection and analysis point. At the detection and analysis point, the corresponding detection and analysis process is executed on the sampled subject transported to the detection and analysis point to obtain detection data and analysis results. The process of generating a test report based on the test data, analysis results, and corresponding basic information includes: The test data and analysis results corresponding to the sampled person, along with their corresponding basic information, are integrated to generate a test report; wherein, the test data and analysis results corresponding to the sampled person refer to the test data and analysis results corresponding to all sampling items that are associated with the sampled person.

[0017] By adopting the above technical solution, a "reflexive detection" mechanism is introduced, making the testing process no longer a pre-set, static list when medical orders are written, but dynamically adjustable and expandable based on intermediate results. This simulates and automates the clinical thinking process of senior physicians, enabling the testing process to "intelligently grow," laying the foundation for truly personalized and accurate testing. Through real-time interpretation and automatic triggering of additional tests by the clinical decision model, the original lengthy and discrete process of "doctor reviewing report → manual judgment → reordering → resubmission" is compressed into a seamless, automatically executed closed loop, saving valuable time for the diagnosis of critical or complex cases and possessing significant clinical value.

[0018] Optionally, before generating the test report, inputting the test data, analysis results, and corresponding basic information into a preset clinical decision-making model includes: Whenever any detection and analysis point completes the corresponding detection and analysis process and obtains detection data and analysis results, the detection data and analysis results, along with the corresponding basic information, are input into a preset clinical decision model; The step of sorting and transporting the sampled individuals to the corresponding detection and analysis points based on the updated binding relationships and preset classification criteria includes: If the primary sampling item of the sampled person is unique, the sampled person will be sorted and transported to the detection and analysis point corresponding to the currently determined secondary sampling item. If the sampled person has a corresponding flow path, the flow path of the sampled person is replanned according to the detection and analysis points corresponding to the secondary sampling items, and then the sampled person is sequentially transported to the detection and analysis points included in the flow path according to the flow path.

[0019] By adopting the above technical solution, the triggering and execution mechanism of reflective detection is refined. It emphasizes that the decision model is triggered in real time when a result is obtained at any detection point, and dynamic path insertion or replanning is performed based on the current state of the sample (whether it is in transit). This ensures that additional detections can be arranged and executed with minimal delay, achieving real-time process response and maximizing the time efficiency of this intelligent mechanism. The integration logic of reflective detection with existing single-path or multi-project transit paths is handled in detail. Whether it is a simple sample or a sample in a complex transit process, the system can handle it properly, demonstrating the high completeness and engineering feasibility of the system design, and ensuring that this advanced function can be stably integrated into the entire intelligent management process.

[0020] Secondly, this application provides an intelligent blood collection management system, including, The sampling initiation module is used to respond to the sampling initiation command triggered by the medical terminal and obtain basic information, which includes at least the identity information of the sampled person and the sampling items. The sampling execution module is used to respond to the sampling execution command triggered by the sampling personnel, provide the basic information to the sampling personnel, and then bind the sample collected by the sampling personnel with the basic information to form a binding relationship; The sorting and conveying module is used to sort the sampled person's sample and convey it to the corresponding detection and analysis point according to the binding relationship and preset classification criteria. The detection and analysis module is used to perform corresponding detection and analysis processes on the sampled person's sample delivered to the detection and analysis point according to the sampling items in the binding relationship, so as to obtain detection data and analysis results. The report feedback module is used to generate a test report based on the test data, analysis results and corresponding basic information, and send the test report to the medical terminal that triggered the sampling initiation command, so that the medical staff at the medical terminal can receive the test report.

[0021] Thirdly, this application provides an intelligent blood collection management device, including a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as described in any of the first aspects.

[0022] Fourthly, this application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed as described in any of the first aspects.

[0023] In summary, this application includes the following beneficial technical effects: By providing basic information in response to sampling execution instructions, a basis for secondary verification is provided to sampling personnel, reducing the risk of human error. More importantly, the automatic control of sorting equipment based on binding relationships and preset rules replaces traditional manual identification and sorting. This not only increases sorting speed by orders of magnitude but also achieves near 100% accuracy, significantly reducing labor costs and the human error rate. Intelligent sorting directly delivers samples to the corresponding testing and analysis points, achieving precise allocation of testing tasks. This avoids the hassle of repeated data entry and sample searching for testing personnel, as well as the backlog and repeated transportation of samples at incorrect testing points. This allows testing equipment and analysts to focus more on the testing itself, improving overall testing throughput and efficiency. Furthermore, the system automatically acquires testing data and analysis results, automatically matches and synthesizes reports with pre-bound basic information (such as the sampled person's ID, name, and testing items), generating structured testing reports that are automatically pushed to the ordering medical terminal. This process requires no manual intervention, greatly shortening the reporting cycle and enabling medical personnel and sampled individuals to obtain diagnostic evidence more quickly, improving diagnostic efficiency. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart illustrating an intelligent blood collection management method disclosed in an embodiment of this application.

[0026] Figure 2 This is a structural block diagram of an intelligent blood collection management system disclosed in an embodiment of this application.

[0027] Explanation of reference numerals in the attached diagram: 301, Sampling initiation module; 302, Sampling execution module; 303, Sorting and conveying module; 304, Detection and analysis module; 305, Report feedback module. Detailed Implementation

[0028] The following is in conjunction with the appendix Figure 1-2 This application will be described in further detail.

[0029] This application discloses an intelligent blood collection management method, the execution subject of which is an intelligent blood collection management system (hereinafter referred to as the blood collection management system), referring to... Figure 1 The detailed process steps for the blood collection management system to implement the intelligent blood collection management method are as follows.

[0030] S101, in response to the sampling initiation command triggered by the medical terminal, obtains basic information, which includes at least the identity information of the sampled person and the sampling items.

[0031] S102, in response to the sampling execution command triggered by the sampling personnel, provides basic information to the sampling personnel, and then binds the sample collected by the sampling personnel with the basic information to form a binding relationship.

[0032] S103, based on the binding relationship and preset classification criteria, sorts the samples of the sampled persons and transports them to the corresponding detection and analysis points.

[0033] S104, at the detection and analysis point, according to the sampling items in the binding relationship, the corresponding detection and analysis process is executed on the sample of the subject delivered to the detection and analysis point to obtain detection data and analysis results.

[0034] S105 generates a test report based on the test data, analysis results, and corresponding basic information, and sends the test report to the medical terminal that triggered the sampling initiation command, so that the medical staff at the medical terminal can receive the test report.

[0035] In implementation, the exemplary implementation principle for S101 is as follows: Healthcare terminals typically refer to computers, tablets, or dedicated handheld devices on doctors' or nurses' workstations. Healthcare professionals access the blood collection management system described in this application through the healthcare terminal and generate a test request form by clicking the first button (e.g., "Create Blood Collection Instance") pre-set on the access page of the blood collection management system. Healthcare professionals can manually enter the information or directly import the basic information of the sampled individual from the hospital's registration system or hospital information system (HIS). This basic information includes at least the sampled individual's identity information (e.g., name, gender, age, ID number, medical record number) and the sampling items to be performed (e.g., complete blood count, liver function tests, etc.). When the doctor completes the aforementioned operations and clicks the pre-set second button (e.g., "Confirm and Submit Request"), the healthcare terminal generates and triggers a sampling initiation command containing the aforementioned basic information.

[0036] In response to the received sampling initiation command, the blood collection management system's server first retrieves the basic information contained in the command and then generates a unique process tracking code for this sampling procedure. This process tracking code (such as a QR code or barcode) is a concrete, digital representation of the aforementioned basic information; it contains an index pointing to this basic data or directly encodes key information. The blood collection management system sends this process tracking code to the sampled individual's terminal (e.g., the sampled individual's smartphone, received via the hospital's official WeChat account or app) so that the sampled individual can present it at the sampling point.

[0037] For S102, the exemplary implementation principle is as follows: Upon arrival at the sampling point, the person being sampled presents the process tracking code displayed on their terminal to the sampling personnel. The sampling personnel scan the process tracking code using a pre-set first scanning device (such as a barcode scanner). This scanning action is considered by the blood collection management system as automatically triggering a sampling execution command. After reading the process tracking code, the scanning device parses the basic information within it and immediately transmits it to the sampling terminal (such as a computer or tablet on the blood collection table). The sampling terminal then clearly displays this information (such as the person being sampled's name and a list of items to be tested) on the screen for the sampling personnel to verify with the person being sampled on-site. Simultaneously, the sampling terminal controls the connected printer to print a paper label for the process tracking code. After collecting blood from the person being sampled, the sampling personnel affix the printed process tracking code label to the corresponding sample tube. This physical affixing action achieves a substantial binding between the collected sample and the basic information, forming a binding relationship that runs through all subsequent processes. At this point, the process tracking code on the test tube becomes the sample's digital identity card throughout the entire system.

[0038] For S103, an exemplary implementation principle is as follows: Sample tubes (i.e., the samples taken) with process tracking codes affixed are placed into a pre-set automated conveying and sorting system (which may include conveyor belts, tracks, motors, and control units, etc.). During the conveying process, a second scanning device (such as a fixed barcode scanner) fixedly installed at key nodes of the conveying path automatically scans and reads the process tracking code on each sample tube. The blood collection management system obtains the sampling items from the basic information read by the second scanning device, and determines the detection and analysis point corresponding to the sampling item of the currently scanned sample tube according to pre-stored classification criteria, such as the mapping rules between sampling items and detection and analysis points (e.g., if the sampling item includes "blood routine", then the corresponding detection and analysis point is "blood analysis room"). Then, it generates a sorting instruction with the detection and analysis point and sends the sorting instruction to the control unit of the conveying and sorting system. The control unit drives the sorting execution mechanism (such as a guide servo motor, robotic arm, or push rod controlled by a programmable logic controller (PLC)) to accurately guide or guide the sample tube into a specific path or collection container leading to the corresponding detection and analysis point when it is conveyed to a branching point. Those skilled in the art will understand that this automated conveying and sorting process draws on the principles of mature industrial automated logistics sorting technology. Its core lies in achieving the directional flow of physical entities through the coordination of identification, address resolution, and actuator control, which will not be elaborated upon here.

[0039] For S104, the exemplary implementation principle is as follows: After the sampled person's sample is delivered to the corresponding testing and analysis point (such as the biochemistry laboratory): the testing personnel scan the process tracking code of the corresponding sampled person's sample using a preset third scanning gun, or the automatic loading and unloading robot configured at the testing and analysis point grabs the sampled person's sample for the third scanning gun to scan and obtain basic information, which is then transmitted to the blood collection management system. This scanning action is regarded by the blood collection management system as automatically triggering a testing and analysis command. In response to the testing and analysis command, the blood collection management system first determines the testing and analysis process to be executed based on the sampling items in the basic information obtained from the scan. The blood collection management system has a pre-stored database of the correspondence between testing items and testing and analysis processes. For example, for the sampling item being liver function, its corresponding testing and analysis process includes a series of sub-items such as "total protein determination" and "alanine aminotransferase (ALT) determination". Subsequently, the blood collection management system controls the automated loading and unloading robot to send the scanned sample tube into the sample tray or sampling position of the testing and analysis equipment (such as a fully automated biochemistry analyzer) at the current testing and analysis point. The detection and analysis equipment automatically executes a preset detection and analysis procedure corresponding to the sampling item on each sample chamber in the sample tray or sampling position. This procedure typically includes steps such as: quantitative sample aspiration, mixing and reaction with specific reagents, and optical or electrochemical detection under specific conditions (such as specific temperature and wavelength).

[0040] After the detection and analysis equipment completes its operation, it generates a completion signal and outputs two types of information: 1. Detection data: This refers to the raw or pre-processed physical and chemical data obtained directly from the test tube sample. For example, in ALT measurement, the detection data might be the absorbance change value measured at a specific wavelength. 2. Analysis results: The preset calculation unit within the detection and analysis equipment (or the connected LIS system) interprets and calculates the detection data according to preset algorithms and reference ranges, ultimately generating clinically significant quantitative results and / or qualitative judgments. For example, substituting the absorbance change value into a standard curve calculates the ALT concentration as "40 U / L," and judging its status (e.g., "normal" or "high") based on the normal reference range for adults (e.g., 0-40 U / L). This "analysis result" is conclusive information that can be directly read by medical personnel and used for clinical diagnosis.

[0041] For S105, the exemplary implementation principle is as follows: The blood collection management system responds to the completion signal of the testing and analysis equipment, thereby acquiring the test data and analysis results output by the equipment. It then aggregates the test data and analysis results with the basic information scanned and transmitted by the third scanner. The system retrieves a pre-defined structured report template corresponding to the sampling item in the basic information. This report template predefines the fill positions for various types of information (such as patient name, test item, test data, analysis results, reference range, etc.). The blood collection management system fills the test data, analysis results, and basic information into the corresponding fill positions in the testing module, thereby generating a complete and formatted test report. Finally, based on the source of the sampling initiation command that initially triggered this process, the blood collection management system automatically sends this test report back to the medical terminal that triggered the sampling initiation command to generate the basic information, for review and diagnosis by the medical staff at that terminal.

[0042] Optionally, S103 specifically includes the following sub-steps: S1031, Based on the binding relationship and the preset classification criteria, analyze the target detection and analysis points that the sampled person needs to access.

[0043] S1032, if the target detection and analysis point is unique, the sampled person's sample will be sorted and transported to the corresponding target detection and analysis point.

[0044] S1033 If the target detection and analysis points are not unique, then based on the preset path planning strategy, an orderly flow path is planned for the corresponding sample to access all target detection and analysis points corresponding to the sample. According to the flow path, the sample is sequentially transported to the target detection and analysis points contained in the flow path.

[0045] S1033's "based on a preset path planning strategy, plan an ordered flow path for the corresponding sampled subject to access all target detection and analysis points" specifically includes the following steps: The system acquires the load status information of each target detection and analysis point corresponding to the sampled subject in real time. The load status information includes at least the estimated waiting time for new samples at the target detection and analysis point. With the goal of minimizing the overall completion time of the sampled subject, the system determines the optimal order for the sampled subject to access all its target detection and analysis points based on the estimated waiting time for new samples at the target detection and analysis point, so as to generate a flow path.

[0046] In implementation, during the execution of S103, this application further proposes that after the second scanning device reads the process tracking code of the sample tube, the blood collection management system will parse the sampling item list in the basic information corresponding to the process tracking code. If the sampling item list contains multiple sampling items (this application assumes that different sampling lists correspond to different detection and analysis points), then the corresponding sample tube is determined to be a multi-item sample. For multi-item samples, the blood collection management system needs to generate a flow path based on a preset path planning strategy. The specific path planning strategy is a dynamic load balancing rule. The blood collection management system obtains in real time the current load status information of each detection and analysis point corresponding to each sampling item in the sampling item list corresponding to the multi-item sample. The load status information includes the estimated waiting time for adding a new sample, and its calculation formula is: The estimated waiting time for a new sample = the current number of samples to be tested × the average testing time per sample + the remaining completion time of the testing and analysis process currently being executed at the testing and analysis point; if the sample tray or sampling position of the testing and analysis equipment at the current testing and analysis point contains 4 samples, these 4 samples are the current number of samples to be tested, and in addition to these 4 samples, the testing and analysis equipment is currently in working state, that is, it is executing a testing and analysis process for a sample. The time remaining until the end of the testing and analysis process is the remaining completion time. By default, the sampling management system pre-stores the time required for the testing and analysis equipment at each testing and analysis point to execute a testing and analysis process once (i.e., the average testing time per sample), and by default, the testing and analysis equipment records and uploads the start time of its execution of the testing and analysis process to the sampling management system in real time.

[0047] The blood collection management system lists a set S = {P1, P2, ..., Pn} of all detection and analysis points (hereinafter referred to as target detection and analysis points) required for multiple sample items. It calculates the estimated waiting time Tx for each target detection and analysis point Px in the set for new samples using the aforementioned formula. Then, a greedy algorithm is used to consistently select the target detection and analysis point with the shortest Tx from the set and remove it from set S. This process is repeated (each time recalculating the real-time Tx of the remaining target detection and analysis points in set S after removal) until set S is empty. All target detection and analysis points are sorted according to the order in which they were removed from set S, resulting in an ordered flow path with each target detection and analysis point as a node. For example, if multiple sample items require testing A (biochemistry), B (immunology), and C (complete blood count), and the current blood collection management system query shows: waiting time for point A is 30 minutes, for point B is 15 minutes, and for point C is 5 minutes, then the algorithm-generated flow path is: C (complete blood count) → B (immunology) → A (biochemistry). The resulting access order is the desired flow path.

[0048] The blood collection management system, based on the order of all target detection and analysis points included in the flow path, designates the first target detection and analysis point that a multi-item sample has not yet been visited as the current destination point. It generates a sorting instruction with the current destination point in mind and sends it to the control unit of the conveying and sorting system. Then, following the scheme described in S103, the conveying and sorting system transports the multi-item sample to the current destination point. Once the blood collection management system receives a completion signal from the detection and analysis device at the current destination point, it changes the status of the current destination point from "not visited" to "detection completed." At this point, the blood collection management system again uses the first target detection and analysis point that a multi-item sample has not yet visited as the current destination point, based on the order of all target detection and analysis points included in the flow path. The process involves updating the current destination point, regenerating analysis instructions with the new destination point, and controlling the automated loading / unloading robot configured at the current testing and analysis point of the multi-item sample to re-grab the sample and transfer it to the conveying and sorting system. The conveying and sorting system then uses the newly generated analysis instructions to transport the multi-item sample to the new current destination point for testing and analysis. This process continues until there are no unvisited target testing and analysis points in the flow path, indicating that the entire testing and analysis process for the multi-item sample is complete. It's important to note that whenever a testing and analysis process for a current destination point is completed, and it's necessary to re-determine the first unvisited target testing and analysis point for the multi-item sample, the blood collection management system can re-plan the planned path for all unvisited testing and analysis points required by the multi-item sample according to a dynamic complex equilibrium rule. The system selects the testing and analysis point ranked first in the planned path as the current destination point, achieving real-time updates to the planned path. This is based on a greedy algorithm, always selecting the testing and analysis point with the earliest estimated waiting time for new samples as the current destination point.

[0049] Optionally, the intelligent sampling management method may also include the following steps: The system records the real-time status and timestamps of the sampled individuals as they are transferred along the flow path. Based on the real-time status timestamps, the flow path, and the real-time load status information of each target detection and analysis point included in the flow path, the system predicts the completion time of the sampled individuals' detection reports. It then generates status update information in real time, including the latest real-time status of the sampled individuals and the latest predicted completion time of the detection reports, and sends the status update information to the sampled individuals' terminals. The real-time status includes at least the conditions that the samples have been sorted and have arrived at the detection and analysis points.

[0050] In implementation, the blood collection management system also maintains a process status log for each sample. At each key node in the sample's flow, the system automatically records its status (specifically, the content of the corresponding key node is used as the sample's status) and the timestamp at which it occurs. Key nodes include at least: Sorted: Whenever the blood collection management system generates a sorting instruction, the status of the sampled person's sample is updated to sorted, and the corresponding timestamp can be the time when the blood collection management system generates the sorting instruction.

[0051] Arrived at the testing and analysis point: Whenever the third scanning device at the testing and analysis point scans the sample of the person being sampled, the status of the sample of the person being sampled is updated to "Arrived at the testing and analysis point". The corresponding timestamp is the time when the sample of the person being sampled was scanned by the third scanning device of the corresponding testing and analysis point. Detection and analysis begins: Whenever the detection and analysis device starts to detect one of the samples, it sends a start detection signal to the blood collection management system. At this time, the status of the sample is updated to "Detection and analysis begins". The corresponding timestamp is the time when the detection and analysis device at the detection and analysis point starts to detect the sample. Detection and analysis completed: Whenever the detection and analysis device sends an execution completion signal, the status of the corresponding sample is updated to "detection and analysis completed". The corresponding timestamp is the time when the blood collection management system receives the execution completion signal from the detection and analysis device.

[0052] Furthermore, the blood collection management system dynamically predicts and updates the completion time of the test reports for the samples taken based on the following data: Time elapsed: Determine the current status of the sampled person based on the process status log.

[0053] Future path: The remaining flow path of the sample Path_remaining.

[0054] Real-time load: The latest estimated waiting time T for new samples at each detection point in the path_remaining; Standard time consumption: Standard testing time D for each sampling item and standard report generation time R.

[0055] The formula for predicting the remaining time T_remaining can be expressed as: T_remaining = Σ(T_i for P_i in Path_remaining) + Σ(D_i for P_i in Path_remaining) + R; where Σ(T_i for P_i in Path_remaining) is the sum of the waiting time (i.e., the estimated waiting time for new samples) of each of the remaining unvisited detection and analysis points (0 ≤ i ≤ the total number of remaining unvisited detection and analysis points); Σ(D_i for P_i in Path_remaining) is the sum of the detection time (i.e., the time spent executing the detection and analysis process) of the detection item itself corresponding to each of the remaining unvisited detection and analysis points (0 ≤ i ≤ the total number of remaining unvisited detection and analysis points).

[0056] For example, for a sample in the "Arrived at the Biochemistry Lab (Point A)" state, its remaining flow path is [B, C]. Then, the predicted time = T_B + T_C + D_B + D_C + R. T_B refers to the estimated waiting time for new samples at testing point B, and T_C refers to the estimated waiting time for new samples at testing point C. D_B refers to the standard testing time for the testing equipment at testing point B to complete this testing process (i.e., to complete the corresponding sampling item), and D_C refers to the standard testing time for the testing equipment at testing point C to complete this testing process (i.e., to complete the corresponding sampling item). R refers to the standard time required for the blood collection management system to integrate the testing data, analysis results, and basic information from all testing points and generate a test report.

[0057] The blood collection management system periodically (e.g., every 5 minutes) or whenever the status of a sample (i.e., a key milestone) changes, triggering an information generation program. This program combines the sample's latest real-time status (e.g., "undergoing biochemistry testing") with the latest predicted report completion time (e.g., "report expected to be available at 14:30 today") and populates it into a pre-defined, user-friendly text template to generate a status update message. The blood collection management system then proactively sends this status update message to the sampler's terminal linked to the sample via a secure interface (e.g., hospital WeChat official account template message, APP push notification, SMS gateway). An example of the push notification content is as follows: "[Test Progress Reminder] Dear Mr. / Ms. Zhang (i.e., sampler's name), your blood sample has completed biochemistry testing and is being sent to the immunology lab. The overall report is expected to be generated in 45 minutes (14:30). Please wait patiently." Optionally, the intelligent blood collection management method may also include the following steps: Collect and analyze historical sampling data to identify common sampling item combinations. Common sampling item combinations refer to combinations of sampling items that appear simultaneously and multiple times in the same basic information. Based on common sampling item combinations and their corresponding target detection and analysis points, adjustment suggestions are generated to optimize the physical layout of the detection and analysis points. The core principle of the adjustment suggestions is to arrange multiple target detection and analysis points corresponding to common sampling item combinations in close proximity in space.

[0058] In practice, the blood collection management system of this application has accumulated a massive amount of historical data on the sampling items contained in the basic information of the sampled individuals over a long period of operation. By mining this historical data, the blood collection management system can provide data-driven decision support for optimizing the physical layout of all testing and analysis points, thereby fundamentally improving the testing efficiency of the sampled individuals. Specifically: The blood collection management system automatically collects and securely stores the list of sampling items from the basic information of each sampled person each time a sampling initiation command containing basic information is received. The list of sampling items can contain one or more sampling items (such as [complete blood count, liver function, kidney function]), forming historical sample data and storing it in a preset historical database.

[0059] The blood collection management system is used to periodically (e.g., quarterly) analyze historical sample data. Specifically, it is used to count the frequency (i.e., occurrence frequency) of different sampling items appearing simultaneously in the same historical sample data. Sampling items whose occurrence frequency exceeds a preset frequency threshold (e.g., the number of occurrences accounts for more than 5% of the total number of samples in the same period) are grouped into common sampling item combinations. For example, the blood collection management system identifies {complete blood count, high-sensitivity C-reactive protein} and {complete liver function panel, complete kidney function panel} as two combinations whose occurrence frequency exceeds the preset frequency threshold.

[0060] For each identified common sampling item combination, the blood collection management system determines the corresponding target detection and analysis point by querying the preset mapping relationship between sampling items and detection and analysis points. For example, the common sampling item combination {complete blood count, high-sensitivity C-reactive protein} corresponds to [blood analysis room A, specific protein analysis room C]. Then, with the goal of minimizing the flow distance and time of common sampling item combinations, the blood collection management system generates adjustment suggestions. The adjustment suggestions can be sent to the management terminal in text form. For example, based on the preset suggestion template: for sampling item combinations (X, Y) that frequently appear in groups during the data analysis period, it is recommended to adjust the physical location of their corresponding detection and analysis points (P_X, P_Y) to adjacent or the same working area, and use common sampling item combinations and corresponding detection and analysis points to replace the fill positions in the template (i.e., (X, Y), (P_X, P_Y)).

[0061] Optionally, the sampling items in the basic information obtained when responding to the sampling initiation command are the original sampling items; Intelligent blood collection management methods also include: S201 Before generating the test report, the test data, analysis results and corresponding basic information are input into the preset clinical decision model, and the clinical decision model determines whether to trigger the reflex test.

[0062] S202, if it is determined that reflective detection needs to be triggered, the sampling item corresponding to the reflective detection to be triggered is taken as a secondary sampling item, and a binding relationship is established between the secondary sampling item and the sampled person's sample to update the binding relationship of the sampled person's sample. Then, according to the updated binding relationship and the preset classification criteria, the sampled person's sample is sorted and transported to the corresponding detection and analysis point. At the detection and analysis point, the corresponding detection and analysis process is executed on the sampled person's sample transported to the detection and analysis point to obtain detection data and analysis results.

[0063] The step of "generating a test report based on test data, analysis results, and corresponding basic information" in S105 includes the following steps: The test report is generated by integrating the test data and analysis results corresponding to the sampled person and their basic information. The test data and analysis results corresponding to the sampled person refer to the test data and analysis results corresponding to all sampling items that are associated with the sampled person.

[0064] S201 specifically includes the following sub-steps: whenever any detection and analysis point completes the corresponding detection and analysis process and obtains detection data and analysis results, the detection data and analysis results, as well as the corresponding basic information, are input into the preset clinical decision model.

[0065] The step S202, "sorting and transporting the sampled individuals to the corresponding detection and analysis points based on the updated binding relationship and preset classification criteria," specifically includes the following sub-steps: If the primary sampling item of the sampled individual is unique, the sampled individual is sorted and transported to the detection and analysis point corresponding to the currently determined secondary sampling item; if the sampled individual has a corresponding flow path, the flow path of the sampled individual is replanned based on the detection and analysis point corresponding to the secondary sampling item, and then the sampled individual is transported sequentially to the detection and analysis points included in the flow path according to the flow path.

[0066] In implementation, when any detection and analysis device (such as a fully automated blood cell analyzer) at any detection and analysis point completes the corresponding detection and analysis process and outputs detection data and analysis results (such as a blood routine report, including the values ​​of each sub-item of blood cells, red blood cells, and platelets and abnormal markers), the blood collection management system immediately retrieves all basic information bound to the sample of the sampled person from the preset database, including: the sampled person's identity information (age, gender, main diagnosis / clinical department (such as hematology, fever to be determined)), and all sampling items included in the current sampling initiation instruction (i.e., primary sampling items), as well as all secondary sampling items determined before the current time (i.e., sampling items corresponding to the triggered reflex detection before the current time). The aforementioned information and the currently output detection data and analysis results are converted into standardized numerical features that can be processed, such as: the sampled person's age is directly used as data or segmented into unique thermal codes (e.g., children [0-12], adults [13-60], elderly [61-]); the age code is a binary value (e.g., male=0, female=1). Clinical departments / primary diagnoses are converted into high-dimensional vectors using word embedding technology or by searching a pre-defined department coding table. For example, "hematology" corresponds to a specific 256-dimensional vector, representing "high probability of hematopoietic system disease"; "fever of unknown origin" corresponds to another vector, representing "possible infection or inflammation".

[0067] For each sampled item, the detection value, unit, reference range, and label (e.g., 'H' (indicating high) or 'L' (indicating low)) are used as features. For example, the feature representation of white blood cell count (WBC) is: {Value: 15.2, Unit: x10^9 / L, Label: 'H', Whether it exceeds the reference range: Yes}. For morphological abnormality labels (e.g., "abnormal lymphocytes"), they are treated as a single Boolean feature (True / False) or multiple categorical features (e.g., Abnormality type_1 = "primitive cells", Abnormality type_2 = "atypical lymphocytes").

[0068] The standardized numerical features obtained from the above processing are then input into a pre-defined clinical decision-making model. The core of this model is a deep neural network trained on massive amounts of clinical data. Internally, this network can be understood as storing tens of thousands of complex mapping patterns of "clinical patterns - test results - follow-up actions." The clinical decision-making model performs high-speed parallel matching calculations between the currently input feature vector and the tens of thousands of "clinical patterns" it has learned. This process is not a simple "if-then" rule, but rather calculates a matching score through the weights and activation functions of each layer of the neural network.

[0069] Example as follows: In the internal computation of the clinical decision model, the departmental feature vector from "Hematology" shows a high degree of matching with the clinical patterns corresponding to "hematopoietic system diseases" learned by the clinical decision model during training. This results in the feature being highly activated in the middle layer of the network, and the activation value it generates contributes significantly to the signal strength of the final sampling item "morphological re-examination is required" (e.g., contributing +0.5 points) when propagating to subsequent layers.

[0070] During the computation of the clinical decision model, when the input features "white blood cell count marker: 'H' (elevated)" and "abnormal lymphocyte marker: True" appear simultaneously, the resulting feature combination highly matches a key discrimination rule learned by the clinical decision model during the training phase, corresponding to the clinical pattern of "suspected lymphoproliferative disorder." This match leads to the strong activation of specific neurons or groups of neurons in the network responsible for identifying such hematological abnormalities. The high activation values ​​output by these neurons, when propagated deeper into the network, significantly reinforce the signaling pathway in the final decision layer that indicates the need for morphological microscopy.

[0071] The input feature "patient age: 45 years old" does not match the typical age distribution characteristics (e.g., younger age group) corresponding to another clinical pattern learned by the clinical decision model: "childhood leukemia". This mismatch leads to reduced effective stimulation received by neurons in the network sensitive to the "childhood leukemia" clinical pattern, resulting in lower activation values ​​at their output. As this lower activation value propagates through the network, it inhibits or weakens the signaling pathway that ultimately makes the sampling item "childhood leukemia-related test".

[0072] Clinical decision models do not view outliers in isolation, but rather comprehensively evaluate the combined significance of all features. For example, an elevated white blood cell count alone may correspond to an infection. However, if the combination of a significantly decreased lymphocyte percentage and the presence of abnormal lymphocytes is included, and the sampled individual is from a hematology department, the fit of the "infection" clinical pattern will decrease, while the fit of the "hematologic malignancy or disorder" clinical pattern will increase dramatically.

[0073] After forward propagation computation through multiple layers of neural networks, the information finally reaches the output layer. Each neuron in the output layer corresponds to a possible "sampled item for reflex detection." Each neuron calculates a raw score. Subsequently, the raw scores of all neurons are normalized using the Softmax function, converting the scores into a probability distribution. This probability value is the confidence level recommended by the model. For example, the neuron corresponding to "peripheral blood cell morphology microscopy" outputs a probability of 0.92, meaning the model has a 92% confidence level to recommend adding this examination.

[0074] Ultimately, the clinical decision model will output a ranking list, for example: {Sampling item: Peripheral blood cell morphology microscopy, confidence level: 0.92, reason code: 1023}; {Sampling item: Flow cytometry immunophenotyping, confidence level: 0.65, reason code: 1023}; {Sampling item: Leukemia fusion gene screening, confidence level: 0.20, reason code: 1023}; {Item: C-reactive protein, confidence level: 0.05, reason code: 1105}; where reason code: 1023 is a pattern number defined internally by the model, which can be mapped to a human-readable clinical interpretation, such as: "Elevated white blood cell count with abnormal lymphocyte morphology markers requires microscopic examination to confirm cell type and proportion, excluding lymphoma, leukemia, etc." Finally, based on the clinically valid action threshold preset by the blood collection management system (e.g., confidence level > 0.85), and with the option to set exclusive rules for the highest confidence level item (e.g., if the confidence level of the first item ranked from highest to lowest is much higher than that of the second item, only the confidence level of the first item is adopted), the system determines whether to trigger a reflexive detection. In the example above, the confidence level of "peripheral blood cell morphology microscopy" is 0.92 > 0.85, which is much higher than the second item's 0.65. Therefore, the system's final decision is to trigger a reflexive detection and add a new sampling item (i.e., a secondary sampling item) as "peripheral blood cell morphology microscopy".

[0075] The above reasoning process involves digitizing the clinical scenario → performing deep, non-linear matching with a vast amount of historical diagnostic and treatment patterns → calculating the probability of reasonableness for various subsequent examinations → selecting high-probability and clinically feasible items as action commands. In other embodiments, after the clinical decision model outputs secondary sampling items, the blood collection management system can further send the model output results to the medical terminal that triggered the sampling initiation command for review and confirmation by the medical terminal personnel. If the addition is confirmed, a reflexive detection is ultimately triggered, and the corresponding newly added sampling item is designated as a secondary sampling item. In another embodiment, the blood collection management system can also transmit the test results and analysis results output from each completed test and analysis process, along with the corresponding basic information of the sampled individual, to the medical terminal that triggered the sampling initiation command. This allows the medical terminal personnel to determine whether a new sampling item is needed. If so, a reflexive detection is triggered, and the newly added sampling item from the medical terminal's feedback is designated as a secondary sampling item.

[0076] If reflective detection needs to be triggered, then based on all sampling items (including primary sampling items and secondary sampling items generated before the current time) that have established a binding relationship with the sampled subject's sample before the current time, the sampling items determined by the currently triggered reflective detection are also treated as secondary sampling items and a binding relationship is established with the sampled subject's sample. Then, the steps described in S1031, S1032, and S1033 above are executed, that is, to determine whether the target detection analysis point (i.e., the detection analysis point that has not been visited among all the detection analysis points corresponding to all sampling items that have established a binding relationship with the sampled subject's sample) is unique. If it is unique, then the sampled subject's sample is directly sent to the corresponding target detection analysis point for detection; if it is not unique, The process involves planning a flow path based on all current target detection and analysis points, and then sequentially sending the sampled person's sample to the target detection and analysis points according to the flow path, until the detection and analysis process at all target detection and analysis points is completed. During this process, whenever any target detection and analysis point produces analysis results and detection data, the above steps are repeated (determining whether reflective detection and the generation of new secondary sampling items need to be triggered), until the detection of all target detection and analysis points is completed. Then, the analysis results, detection data, and sampled person's identity information corresponding to all sampling items that have established a binding relationship with the sampled person's sample are integrated to generate a detection report (the specific implementation plan for generating the detection report has been disclosed above and will not be repeated here).

[0077] This application also discloses an intelligent blood collection management system. (Refer to...) Figure 2 ,include The sampling initiation module 301 is used to respond to the sampling initiation command triggered by the medical terminal and obtain basic information, which includes at least the identity information of the sampled person and the sampling items. The sampling execution module 302 is used to respond to the sampling execution command triggered by the sampling personnel, provide the basic information to the sampling personnel, and then bind the sample collected by the sampling personnel with the basic information to form a binding relationship; The sorting and conveying module 303 is used to sort the sampled person's sample and convey it to the corresponding detection and analysis point according to the binding relationship and the preset classification criteria. The detection and analysis module 304 is used to perform a corresponding detection and analysis process on the sampled person delivered to the detection and analysis point according to the sampling items in the binding relationship, so as to obtain detection data and analysis results. The report feedback module 305 is used to generate a test report based on the test data, analysis results and corresponding basic information, and send the test report to the medical terminal that triggered the sampling initiation command, so that the medical staff at the medical terminal can receive the test report.

[0078] Optionally, the sorting and conveying module 303 is further configured to analyze the target detection and analysis points that the sampled person's sample needs to access based on the binding relationship and the preset classification criteria; if the target detection and analysis point is unique, the sampled person's sample is sorted and conveyed to the corresponding target detection and analysis point; if the target detection and analysis point is not unique, based on the preset path planning strategy, an orderly flow path is planned for the corresponding sampled person's sample to access all target detection and analysis points corresponding to the sampled person's sample, and the sampled person's sample is conveyed sequentially to the target detection and analysis points included in the flow path according to the flow path.

[0079] Optionally, the sorting and conveying module 303 is also used to acquire the load status information of each target detection and analysis point corresponding to the sampled person in real time. The load status information includes at least the estimated waiting time for new samples at the target detection and analysis point. With the goal of minimizing the overall completion time of the sampled person's samples, the optimal order for the sampled person's samples to access all their target detection and analysis points is determined based on the estimated waiting time for new samples at the target detection and analysis points, so as to generate the flow path.

[0080] Optionally, a sampling progress update module is also included, which is used to record the real-time status and timestamp of the sampled subject during the process of being transferred according to the transfer path. Based on the timestamp of the real-time status, the transfer path, and the real-time load status information of each target detection and analysis point included in the transfer path, the module predicts the completion time of the detection report of the sampled subject, generates status update information with the latest real-time status of the sampled subject and the latest predicted completion time of the detection report in real time, and sends the status update information to the sampled subject's terminal; wherein, the real-time status includes at least sorted and arrived at the detection and analysis point.

[0081] Optionally, it also includes an adjustment suggestion module, which is used to collect and analyze historical sampling data, identify common sampling item combinations, wherein the common sampling item combinations refer to the combination of sampling items that appear simultaneously and multiple times in the same basic information; based on the common sampling item combinations and their corresponding target detection and analysis points, it generates adjustment suggestions for optimizing the physical layout of the detection and analysis points, wherein the core principle of the adjustment suggestions is to arrange multiple target detection and analysis points corresponding to the common sampling item combinations in close proximity in space.

[0082] Optionally, a new sampling module is also included, used to input the test data, analysis results, and corresponding basic information into a preset clinical decision model before generating a test report. The clinical decision model determines whether a reflexive test needs to be triggered. If a reflexive test needs to be triggered, the sampling item corresponding to the required reflexive test is used as a secondary sampling item. A binding relationship is established between the secondary sampling item and the sampled subject's sample to update the binding relationship of the sampled subject's sample. Then, based on the updated binding relationship and preset classification criteria, the sampled subject's sample is sorted and transported to the corresponding testing and analysis point. At the testing and analysis point, the corresponding testing and analysis process is executed on the sampled subject transported to the testing and analysis point to obtain test data and analysis results.

[0083] The report feedback module 305 is also used to integrate the test data and analysis results corresponding to the sampled person and the corresponding basic information to generate a test report; wherein, the test data and analysis results corresponding to the sampled person refer to the test data and analysis results corresponding to all sampling items that are bound to the sampled person.

[0084] Optionally, the newly added sampling module is also used to input the test data and analysis results, along with the corresponding basic information, into a preset clinical decision model whenever any testing and analysis point completes the corresponding testing and analysis process and obtains test data and analysis results. It is also used to sort and transport the sample to the testing and analysis point corresponding to the currently determined secondary sampling item if the primary sampling item of the sampled subject is unique; if the sampled subject has a corresponding flow path, it re-plans the flow path of the sampled subject according to the testing and analysis point corresponding to the secondary sampling item, and then transports the sampled subject to the testing and analysis points included in the flow path sequentially according to the flow path.

[0085] This application also discloses an intelligent blood collection management device, which includes a memory and a processor. The memory stores a computer program that can be loaded by the processor and executed as described above for intelligent blood collection management.

[0086] This application also discloses a computer-readable storage medium that stores a computer program that can be loaded by a processor and executed as described above for intelligent blood collection management. The computer-readable storage medium includes, for example, various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0087] It should be noted that in this paper, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0088] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit the scope of protection of the application. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on these embodiments, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

Claims

1. An intelligent blood collection management method, characterized in that, include: In response to a sampling initiation command triggered by a medical terminal, basic information is obtained, including at least the identity information of the sampled person and the sampling items. In response to a sampling execution command triggered by a sampling personnel, the basic information is provided to the sampling personnel, and then the sample collected by the sampling personnel is bound to the basic information to form a binding relationship; Based on the binding relationship and the preset classification criteria, the sampled individuals are sorted and transported to the corresponding detection and analysis points; At the detection and analysis point, according to the sampling items in the binding relationship, the corresponding detection and analysis process is performed on the sample of the subject delivered to the detection and analysis point to obtain detection data and analysis results; Based on the test data, analysis results, and corresponding basic information, a test report is generated and sent to the medical terminal that triggered the sampling initiation command, so that the medical staff at the medical terminal can receive the test report.

2. The intelligent blood collection management method according to claim 1, characterized in that, The step of sorting and transporting the sampled individuals to the corresponding detection and analysis points according to the binding relationship and preset classification criteria includes: Based on the binding relationship and the preset classification criteria, analyze the target detection and analysis points that the sampled person's sample needs to access; If the target detection and analysis point is unique, the sampled person's sample will be sorted and transported to the corresponding target detection and analysis point; If the target detection and analysis point is not unique, then based on the preset path planning strategy, an orderly flow path is planned for the corresponding sample to access all target detection and analysis points corresponding to the sample. According to the flow path, the sample is sequentially transported to the target detection and analysis points included in the flow path.

3. The intelligent blood collection management method according to claim 2, characterized in that, The preset path planning strategy plans an ordered flow path for each sampled subject to access all target detection and analysis points corresponding to that sampled subject, including: The load status information of each target detection and analysis point corresponding to the sampled person is acquired in real time. The load status information includes at least the estimated waiting time for the target detection and analysis point to add new samples. With the goal of minimizing the overall completion time of the sampled subjects, the waiting time of new samples is estimated based on the target detection and analysis points, and the optimal order for the sampled subjects to access all their target detection and analysis points is determined to generate the flow path.

4. The intelligent blood collection management method according to claim 3, characterized in that, The method further includes: The system records the real-time status and timestamps of the sampled object as it flows along the transfer path. Based on the timestamps of the real-time status, the transfer path, and the real-time load status information of each target detection and analysis point included in the transfer path, the system predicts the completion time of the detection report for the sampled object. It then generates status update information in real time, containing the latest real-time status of the sampled object and the latest predicted completion time of the detection report, and sends the status update information to the sampled object's terminal. The real-time status includes at least "sorted" and "arrived at the detection and analysis point." 5. The intelligent blood collection management method according to claim 3, characterized in that, The method further includes: Collect and analyze historical sampling data to identify common sampling item combinations, which refer to combinations of sampling items that appear simultaneously and multiple times in the same basic information. Based on the common sampling item combinations and their corresponding target detection and analysis points, adjustment suggestions are generated to optimize the physical layout of the detection and analysis points. The core principle of the adjustment suggestions is to arrange multiple target detection and analysis points corresponding to the common sampling item combinations in close proximity in space.

6. The intelligent blood collection management method according to claim 3, characterized in that, The sampling items in the basic information obtained when responding to the sampling initiation command are the original sampling items; The method further includes: Before generating the test report, the test data, analysis results and corresponding basic information are input into a preset clinical decision model, and the clinical decision model determines whether a reflexive test needs to be triggered. If it is determined that reflective detection needs to be triggered, the sampling item corresponding to the required reflective detection is taken as a secondary sampling item. A binding relationship is established between the secondary sampling item and the sampled subject to update the binding relationship of the sampled subject. Then, according to the updated binding relationship and the preset classification criteria, the sampled subject is sorted and transported to the corresponding detection and analysis point. At the detection and analysis point, the corresponding detection and analysis process is executed on the sampled subject transported to the detection and analysis point to obtain detection data and analysis results. The process of generating a test report based on the test data, analysis results, and corresponding basic information includes: The test data and analysis results corresponding to the sampled person, along with their corresponding basic information, are integrated to generate a test report; wherein, the test data and analysis results corresponding to the sampled person refer to the test data and analysis results corresponding to all sampling items that are associated with the sampled person.

7. The intelligent blood collection management method according to claim 6, characterized in that, Before generating the test report, the test data, analysis results, and corresponding basic information are input into a preset clinical decision-making model, including: Whenever any detection and analysis point completes the corresponding detection and analysis process and obtains detection data and analysis results, the detection data and analysis results, along with the corresponding basic information, are input into a preset clinical decision model; The step of sorting and transporting the sampled individuals to the corresponding detection and analysis points based on the updated binding relationships and preset classification criteria includes: If the primary sampling item of the sampled person is unique, the sampled person will be sorted and transported to the detection and analysis point corresponding to the currently determined secondary sampling item. If the sampled person has a corresponding flow path, the flow path of the sampled person is replanned according to the detection and analysis points corresponding to the secondary sampling items, and then the sampled person is sequentially transported to the detection and analysis points included in the flow path according to the flow path.

8. An intelligent blood collection management system, characterized in that, include, The sampling initiation module (301) is used to obtain basic information in response to the sampling initiation command triggered by the medical terminal. The basic information includes at least the identity information of the sampled person and the sampling items. The sampling execution module (302) is used to respond to the sampling execution command triggered by the sampling personnel, provide the basic information to the sampling personnel, and then bind the sample of the sampled person collected by the sampling personnel with the basic information to form a binding relationship; The sorting and conveying module (303) is used to sort the sampled person's sample and convey it to the corresponding detection and analysis point according to the binding relationship and the preset classification criteria. The detection and analysis module (304) is used to perform the corresponding detection and analysis process on the sampled person delivered to the detection and analysis point according to the sampling items in the binding relationship at the detection and analysis point, so as to obtain detection data and analysis results; The report feedback module (305) is used to generate a test report based on the test data, analysis results and corresponding basic information, and send the test report to the medical terminal that triggered the sampling initiation command, so that the medical staff at the medical terminal can know the test report.

9. An intelligent blood collection management device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer program is stored that can be loaded by a processor and executed as described in any one of claims 1 to 7.

Images