Sample bank whole-process scheduling method based on genetic algorithm and biological stability modeling

Abstract

The application provides a sample bank full-process scheduling method based on a genetic algorithm and biological stability modeling, and relates to the technical field of biological sample bank management. In view of the problems of the long cold ischemic time of samples from ex vivo to cryopreservation, the waste caused by uneven sample resource allocation, and the new compliance requirements such as temperature fluctuation of the biological sample bank and ethical withdrawal which are difficult to be embedded into the scheduling process, the human resources, processing equipment and storage space in the biological sample bank are coded as binary chromosomes, a multi-objective fitness function is constructed, and the genetic algorithm is used for dynamic optimization in the full process. Through the multi-objective optimization scheduling driven by the genetic algorithm, the sample activity preservation rate, the equipment utilization rate and the thermodynamic stability are improved, and the defects of the original scheduling method are solved. Through the closed-loop architecture of the physical layer, the perception layer, the decision layer and the execution layer, the collaborative optimization of the sample activity, the resource efficiency and the ethical compliance is realized, and the sample failure risk is reduced.

Description

Technical Field

[0001] This invention relates to the field of biobank management technology, and in particular to a biobank end-to-end scheduling method based on genetic algorithms and biological stability modeling. Background Technology

[0002] Currently, biobanks often face numerous challenges in their operation. For example, they have extremely high timeliness requirements; the cold ischemia time from in vitro to deep cryopreservation directly impacts research quality, and traditional manual scheduling struggles to ensure samples remain on the optimal processing path during surges in sample volume. Resource allocation in biobanks is uneven, with a lack of coordinated optimization between factors such as researcher fatigue, equipment occupancy, and storage space fragmentation. Sample safety and compliance in biobanks cannot be guaranteed; frequent refrigerator opening and removal operations cause temperature fluctuations, and compliance requirements such as ethical retractions are difficult to integrate into the real-time scheduling process. Biobanks lack predictability, failing to incorporate front-end data such as hospital surgical schedules for forward-looking resource allocation.

[0003] Biobanks primarily employ standardized methods for collecting, processing, storing, and applying samples of biological macromolecules, cells, tissues, and organs from organisms that resist or treat diseases. They facilitate the rapid commercialization of numerous important scientific research findings and their clinical application, serving as a crucial guarantee for translational medicine. Therefore, the management and allocation of biobank samples play a vital role. However, currently, no biobank can evenly distribute resources across multiple samples to ensure their optimal condition. This failure to match the actual conditions at the front end with the biobank's capabilities leads to resource waste and negatively impacts research quality.

[0004] To address the aforementioned issues, a forward-looking management method is urgently needed that integrates front-end data such as personnel and processing equipment with biobanks. This method should ensure the compliance and security of samples within the biobank, prevent disruption to temperature fluctuations, and allow for real-time embedding of compliance requirements, including ethical retractions, into the scheduling process. Furthermore, it should predict resource allocation within the biobank, reduce errors, improve the accuracy of human judgment in system evaluations, and enhance the resource preservation efficiency of the biobank. Summary of the Invention

[0005] To address the shortcomings of existing technologies in protecting samples from in vitro to cryogenic storage due to insufficient cold ischemia time, uneven distribution of sample resources leading to waste, and the difficulty in integrating new compliance requirements such as temperature fluctuations in biobanks and ethical retractions into the scheduling process, this invention first proposes a full-process scheduling method for biobanks based on genetic algorithms and biological stability modeling. This method aims to improve sample preservation efficiency, avoid impacting research quality, and allows for forward-looking resource allocation by incorporating front-end data. In this method, human resources, processing equipment, and storage space in the biobank are encoded as binary chromosomes, a multi-objective fitness function is constructed, and a genetic algorithm is used for dynamic optimization in all four stages of the process. The method specifically includes the following steps: Step S1: Construct a genetic algorithm scheduling engine driven by a multi-objective fitness function, encode human resources, processing equipment, and storage space in the sample library as binary chromosomes, and set the multi-objective fitness function with the indicators of sample activity retention rate, equipment and human resource utilization rate, and thermodynamic fluctuation minimization. Step S2: Establish a sample bioactivity decay model, collect the temperature of the transport box, sample type and ex-situ time in real time, calculate the remaining golden processing time for each batch of samples, and dynamically adjust the sample processing priority based on the remaining golden processing time, and feed the multi-objective fitness function back to the sample bioactivity decay model. Step S3: Construct a laboratory digital twin pressure balancing system, link it to the hospital's surgical scheduling system, dynamically adjust the emergency buffer ratio of automated equipment according to the expected sample output, monitor the continuous operation time of the laboratory technicians and the error rate prediction, and automatically schedule high-precision samples to automated robots when it is determined that the manual accuracy has decreased. Step S4: Introduce a thermodynamic stability sensing algorithm to partition and store samples according to their outbound frequency. High-frequency samples are stored in the front area of ​​the refrigerator, and low-frequency samples are stored in the deep area of ​​the refrigerator. During periods of low system load, a fragmentation sorting scheme is automatically generated by a genetic algorithm. The fragmentation sorting scheme is subject to freeze-thaw constraints to ensure that the sample exposure time during the sorting process is below a preset threshold. Step S5: Use a clustering algorithm to cluster and optimize multiple outbound applications according to their storage locations, so as to complete multiple outbound tasks in a single door opening, and link to the informed consent management system in real time to lock the gene location of samples that have been withdrawn due to ethical reasons.

[0006] Furthermore, in step S1, the sample viability retention rate is calculated based on the sample bioactivity decay model.

[0007] Furthermore, in step S1, the expression for the multi-objective fitness function is: ; Where Q is the sample activity retention rate, calculated based on the sample biological activity decay model, which calculates the predicted activity value of each batch of samples at the end of the scheduling period and takes the average value; E is the equipment and manpower utilization rate, which includes the weighted sum of equipment occupancy rate and the workload balance of the experimenters; C is the thermodynamic fluctuation minimization index, calculated based on the number of times the refrigerator door is opened, the duration of the door opening, and the integral of the temperature fluctuation during the door opening; w1, w2, and w3 are the corresponding weight coefficients.

[0008] Furthermore, in step S2, dynamically adjusting the sample processing priority includes: when it is detected that the rate of biological activity decay of a sample exceeds a preset threshold due to abnormal transport, the genetic algorithm scheduling engine performs priority site mutation on the chromosome corresponding to the sample through mutation operators, so that its processing priority can transcend all regular samples in the current queue and enter the immediate processing channel.

[0009] Furthermore, in step S3, the digital twin pressure balancing system includes: a real-time mirror of the laboratory physical environment, including equipment operating status, operator positions, and sample queues to be processed; a hospital surgical scheduling system providing predicted data on specimen output within a preset time period; and a genetic algorithm scheduling engine adjusting the emergency buffer ratio of automated equipment in advance based on the predicted data.

[0010] Furthermore, in step S3, the method for determining the decline in human precision includes: monitoring the continuous operation time of the experimenter, and triggering a fatigue warning when the continuous operation time exceeds a first threshold; monitoring the recent operation error rate of the experimenter, and triggering a precision decline determination when the error rate exceeds a second threshold; when the fatigue warning and the precision decline determination are triggered simultaneously, the high-precision sample is automatically scheduled to the automated robot.

[0011] Furthermore, in step S4, the freeze-thaw constraint includes: in the fragment sorting scheme generated by the genetic algorithm, the cumulative exposure time of each sample from its original location to its re-entry into the target location must not exceed a preset threshold. Sorting schemes that exceed this threshold are eliminated during the iteration process of the genetic algorithm and do not proceed to subsequent optimization.

[0012] Furthermore, in step S4, the system low-load period is the period when the sample library operation frequency is lower than a preset threshold, which is determined by statistical analysis of historical operation data, including the early morning period every day. The genetic algorithm automatically triggers the generation and execution of the defragmentation scheme during the low-load period.

[0013] Furthermore, in step S5, cluster optimization includes: obtaining the sample storage location coordinates corresponding to each application in multiple outbound applications, using the K-means clustering algorithm to group samples with adjacent storage locations into the same outbound batch, planning a single door opening path to traverse all sample storage locations within the batch, and outputting the optimal door opening path sequence.

[0014] Furthermore, in step S5, the gene locus locking includes: real-time synchronization of the ethical status of the informed consent management system; when the withdrawal of the informed consent form corresponding to any sample is detected, all gene locus data of that sample in the sample bank management system are marked as inaccessible, and any form of decoding, data export and data retrieval operations are prohibited until the ethical status is restored.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: Firstly, this invention constructs a multi-objective fitness function-driven genetic algorithm scheduling engine, establishes a sample bioactivity decay model, feeds priority information back to the genetic algorithm scheduling engine to update chromosome fitness, and uses the scheduling result as a precondition for storage optimization. It introduces a thermodynamic stability-aware algorithm, and after completing sample partitioning and fragmentation, it uses a clustering algorithm to optimize multiple outbound requests by storage location, and locks the gene loci for samples withdrawn ethically. Through biostability modeling, cold ischemia time is compressed to the limit, maximizing the preservation of sample research value. Through multi-objective optimization scheduling driven by a genetic algorithm, it simultaneously improves sample bioactivity preservation rate, equipment utilization rate, and thermodynamic stability, overcoming the shortcomings of single-objective optimization in traditional scheduling methods. The introduction of a sample bioactivity decay model and dynamic priority adjustment of remaining golden processing time significantly reduces the risk of sample failure due to transport anomalies.

[0016] Secondly, by combining a digital twin pressure balancing system with precise human prediction, human-machine collaborative scheduling is achieved, ensuring high-precision task quality while improving overall processing efficiency. The fragmentation sorting scheme under freeze-thaw constraints and the clustering-optimized retrieval strategy reduce the number of refrigerator door openings and sample exposure time, extending sample shelf life. Real-time linkage between gene locus locking and informed consent meets the ethical compliance requirements of biobanks. By constructing a closed-loop architecture encompassing the physical, perception, decision-making, and execution layers, the synergistic optimization of sample viability, resource efficiency, and ethical compliance is achieved. This realizes a shift from passively receiving samples to proactively deploying them based on surgical prediction. Attached Figure Description

[0017] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 A flowchart of a sample library scheduling method based on genetic algorithms and biological stability modeling; Figure 2This is a schematic diagram of the closed-loop structure for the entire process scheduling of the sample library based on genetic algorithms and biological stability modeling. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0020] The specific embodiments of the present invention will be described below.

[0021] To address the shortcomings of existing technologies in protecting samples from in vitro to cryopreservation during the "cold ischemia time," the waste caused by uneven sample resource allocation, and the difficulty in integrating new compliance requirements such as temperature fluctuations in biobanks and ethical retractions into the scheduling process, this invention first proposes a full-process scheduling method for biobanks based on genetic algorithms and biological stability modeling. This method aims to improve sample preservation efficiency, avoid impacting research quality, and allows for forward-looking resource allocation by incorporating front-end data.

[0022] Example 1 like Figure 1 As shown, this invention proposes a full-process scheduling method for a sample library based on genetic algorithms and biological stability modeling, which specifically includes the following steps: Step S1: Construct a genetic algorithm scheduling engine driven by a multi-objective fitness function. Human resources, processing equipment, and storage space in the sample library are encoded as binary chromosomes. The multi-objective fitness function is set based on the indicators of sample viability preservation rate, equipment and human resource utilization rate, and minimization of thermodynamic fluctuations. Specifically, in this step, the temperature of the transport box, sample type, and sample ex-vivo time are collected in real time to ensure the cold ischemia time for samples from ex-vivo to deep cryopreservation. Human resources, such as the lab technician schedule, processing equipment, such as automated dispensing systems, centrifuges, and the location of refrigerator shelves in the storage space, are encoded as binary chromosomes, with each chromosome representing a scheduling scheme. Simultaneously, the sample viability preservation rate, equipment and human resource utilization rate, and thermodynamic fluctuation minimization indicators are set as multi-objective fitness functions to organize and statistically analyze all relevant data in the sample library, facilitating subsequent optimization calculations. The sample viability preservation rate is calculated based on a sample biological activity decay model. Real-time monitoring of the sample viability preservation rate avoids situations where inadequate monitoring of existing samples affects research quality.

[0023] Specifically, the expression for the multi-objective fitness function is: ; Wherein, Q represents the sample viability retention rate, calculated based on the sample bioactivity decay model, which calculates the predicted viability value of each batch of samples at the end of the scheduling period and takes the average; E represents the equipment and manpower utilization rate, including the weighted sum of equipment occupancy rate and the workload balance of the experimenters; C represents the thermodynamic fluctuation minimization index, calculated based on the number of times the refrigerator door is opened, the duration of the door opening, and the integral of temperature fluctuation during the opening period; w1, w2, and w3 are the corresponding weighting coefficients. Statistical processing of data from multiple aspects achieves even resource allocation while simultaneously ensuring sample safety and compliance, avoiding sample temperature fluctuations, and guaranteeing sample stability.

[0024] Step S2: Under the control of the genetic algorithm scheduling engine, a sample bioactivity decay model is established. The transport box temperature, sample type, and ex-vitro time are collected in real time. The remaining golden processing time for each batch of samples is calculated, and the processing priority is dynamically adjusted based on this remaining golden processing time. Priority information is fed back to the genetic algorithm scheduling engine to update chromosome fitness. In practical applications, to ensure dynamic scheduling driven by quality decay during sample collection and transport, in addition to real-time collection of transport box temperature, sample type, and ex-vitro time, the algorithm not only queues samples based on sample level but also calculates the remaining golden processing time for each batch of samples in real time. The bioactivity decay function for each sample type is pre-determined experimentally and stored in the system database. The remaining golden processing time for each batch of samples is calculated, i.e., the time from the current moment until the bioactivity drops to the threshold. When a batch of samples experiences a decay rate exceeding the preset threshold due to overheating during transport, the genetic algorithm scheduling engine uses a mutation operator to perform priority site mutation on the chromosome corresponding to that sample, making its processing priority transcend all regular samples in the current queue and enter the immediate processing channel. That is, when abnormal transport of ordinary samples leads to accelerated activity decay, priority mutation can be automatically triggered to promote the ranking of the sample above the highest priority sample, thus realizing risk-priority scheduling.

[0025] Specifically, sample processing priorities include: when a sample's bioactivity decay rate exceeds a preset threshold due to transport anomalies, the genetic algorithm scheduling engine performs priority site mutations on the chromosome corresponding to the sample using mutation operators. This prioritizes the sample, allowing it to bypass all regular samples in the current queue and enter the immediate processing channel. This prevents samples from being overlooked and forgotten by staff, and solves the problem of traditional manual scheduling struggling to guarantee the optimal processing path during sample surges. By directly mapping the dynamic quantitative index of bioactivity decay to the priority gene site mutation operation of the genetic algorithm chromosome, proactive emergency scheduling of abnormal samples is achieved, rather than passive waiting, significantly reducing the risk of high-value samples becoming invalid.

[0026] Step S3: Based on the dynamically adjusted priority results in Step S2, a laboratory digital twin pressure balancing system is constructed and linked to the hospital's surgical scheduling system. The emergency buffer ratio of automated equipment is dynamically adjusted according to the expected sample output, and the continuous operation time and error rate prediction of the laboratory technicians are monitored. When it is determined that human accuracy is declining, high-precision samples are automatically scheduled to automated robots, and the scheduling result is output as a prerequisite for storage optimization. In this embodiment, the hospital's surgical scheduling system is also incorporated into the calculation process to achieve work efficiency synergy and monitor the continuous processing time and error rate prediction of operators. When the system determines that human accuracy may be declining, it automatically schedules high-precision samples to more accurate and precise automated robots, while leaving routine tasks to manual processing. The system mirrors the laboratory physical environment in real time, including the operating status of automated equipment, the operator's position, and the queue of samples to be processed. It is linked to the hospital's surgical scheduling system to obtain the expected sample output.

[0027] Specifically, the digital twin pressure balancing system includes: real-time mirroring of the laboratory's physical environment, including equipment operating status, operator positions, and queues of samples awaiting processing. The hospital's surgical scheduling system provides predicted sample output data for a pre-set time period, and a genetic algorithm scheduling engine adjusts the emergency buffer ratio of automated equipment in advance based on this predicted data. In practical applications, the system mirrors relevant laboratory equipment, personnel locations, and samples in real time, while simultaneously organizing related predicted data to maximize the collection and processing of relevant data for dynamic priority adjustments.

[0028] The methods for determining the decline in human precision include: monitoring the continuous operation time of the experimenter, triggering a fatigue warning when the continuous operation time exceeds a first threshold; monitoring the recent operation error rate of the experimenter, triggering a precision decline judgment when the error rate exceeds a second threshold; and automatically scheduling high-precision samples to an automated robot when both the fatigue warning and the precision decline judgment are triggered simultaneously. The first threshold is the limit of the continuous operation time during the experimenter's actual operation, and the second threshold is the upper limit of the number of errors the experimenter makes. This precisely controls human precision, minimizing operational errors, while using automated robots to alleviate the processing pressure on operators. This protects the samples while improving operational efficiency. It also improves sample viability preservation, equipment utilization, and thermodynamic stability, overcoming the single-objective optimization shortcomings of traditional scheduling methods. By fusing predictive data with real-time mirror data and using digital twins for stress simulation, a leap from passive response to proactive pre-scheduling is achieved.

[0029] Step S4: Based on the scheduling results output in Step S3, a thermodynamic stability-aware algorithm is introduced. Samples are partitioned and stored according to their retrieval frequency: high-frequency samples are stored in the front area of ​​the refrigerator, and low-frequency samples are stored in the deeper areas. During periods of low system load, the genetic algorithm scheduling engine automatically generates a fragmentation plan. This plan is constrained by freeze-thaw constraints to ensure that the sample exposure time during fragmentation is below a preset threshold. This reduces the thermal shock to deeper samples caused by opening the refrigerator door. Furthermore, during periods of low system load, such as early morning, the genetic algorithm can automatically generate fragmentation plans. This strengthens the freeze-thaw constraint mechanism, ensuring that the sample exposure time is less than 30 seconds.

[0030] The freeze-thaw constraint includes the following: in the fragmentation schemes generated by the genetic algorithm, the cumulative exposure time of each sample from its original location to its relocation to the target location must not exceed 30 seconds. Schemes exceeding this threshold are eliminated during the genetic algorithm iteration and do not proceed to subsequent optimization. This optimizes sample protection, reduces sample exposure time, and avoids temperature fluctuations caused by frequent opening and closing of the freezer. By using the freeze-thaw tolerance limit of biological samples as a hard constraint elimination condition for the genetic algorithm, rather than a soft penalty term, the physical and biological feasibility of the fragmentation schemes is ensured.

[0031] Step S5: Based on the sample partitioning and fragmentation completed in Step S4, a clustering algorithm is used to optimize multiple outbound requests by storage location, enabling multiple outbound tasks to be completed in a single opening. This is also linked in real-time to the informed consent management system, and the genetic location of samples withdrawn due to ethical concerns is locked. At this point, any form of decoding and outbound processing is prohibited. Specifically, clustering optimization includes: obtaining the sample storage location coordinates for each of the multiple outbound applications; using the K-means clustering algorithm to group samples with similar storage locations into the same outbound batch; planning a single door-opening path to traverse all sample storage locations within the batch; and outputting the optimal door-opening path sequence. Gene locus locking includes: real-time synchronization of the ethical status of the informed consent management system; when an informed consent form for any sample is detected to have been withdrawn, all gene locus data for that sample in the sample bank management system are marked as inaccessible, prohibiting any form of decoding, outbound, or data export operations until the ethical status is restored.

[0032] In one specific embodiment, samples with similar storage locations are grouped into the same outbound batch. A single door-opening path is planned to traverse all sample storage locations within that batch, outputting the optimal door-opening path sequence. For example, if the system receives 5 outbound requests, they are clustered into 2 batches, with 2-3 tasks completed in a single door opening to reduce the number of door openings. Simultaneously, the ethical status of the informed consent management system is synchronized in real time. When any informed consent form corresponding to a sample is withdrawn, all genomic data for that sample in the biobank management system is marked as inaccessible, prohibiting decoding, outbound processing, and data export until the ethical status is restored. This deep integration of physical space location clustering optimization and real-time ethical status verification improves outbound efficiency while ensuring zero-touch and zero-leakage of ethically withdrawn samples, meeting the dual core requirements of biobank management.

[0033] This invention constructs a multi-objective fitness function-driven genetic algorithm scheduling engine, establishes a sample bioactivity decay model, feeds priority information back to the genetic algorithm scheduling engine to update chromosome fitness, and uses the scheduling result as a precondition for storage optimization. A thermodynamic stability-aware algorithm is introduced to complete sample partitioning and fragmentation. A clustering algorithm is then used to optimize multiple outbound requests by storage location, and gene loci are locked for samples withdrawn due to ethical reasons. Through biostability modeling, cold ischemia time is compressed to the limit, maximizing the preservation of sample research value. The multi-objective optimization scheduling driven by the genetic algorithm simultaneously improves sample bioactivity preservation rate, equipment utilization rate, and thermodynamic stability, overcoming the shortcomings of traditional single-objective optimization scheduling methods. The introduction of a sample bioactivity decay model and dynamic priority adjustment of remaining golden processing time significantly reduces the risk of sample failure due to transport anomalies.

[0034] Example 2 like Figure 2As shown, this embodiment, based on the previous embodiment, further illustrates a sample bank full-process scheduling method based on genetic algorithms and biological stability modeling. The system architecture of this invention includes an input data layer, a core decision-making layer, an execution optimization layer, a physical execution layer, and a feedback closed-loop layer. The input data layer includes a hospital surgical scheduling system, modules for acquiring transport box temperature, sample type, and ex vivo time, an environmental sensing and equipment status module, and an ethical status and informed consent management system, used to provide predictive data and real-time status data to the core decision-making layer. The core decision-making layer includes a genetic algorithm scheduling engine, which internally includes binary chromosome encoding, a multi-objective fitness function, a biological activity decay model, a remaining golden processing time calculation, and a priority dynamic adjustment module, used to achieve collaborative optimization of the sample processing, storage, and retrieval processes. The execution optimization layer includes a digital twin pressure balancing and human-machine collaboration module, a thermodynamic sensing partitioned storage and fragmentation module, and an ethical compliance and K-means clustering retrieval path optimization module, corresponding to sample processing resource scheduling, sample thermal stability maintenance, and sample retrieval and ethical status management, respectively. The physical execution layer includes intelligent freezers, automated robots, and operators, used to execute the scheduling results. The feedback loop layer uses digital twins to dynamically correct feedback and closed-loop control, and transmits the execution results and real-time status back to the genetic algorithm scheduling engine to achieve dynamic correction of scheduling parameters and full-process closed-loop control.

[0035] This embodiment forms a closed-loop system from the front-end hospital surgical scheduling system to the final sample delivery. First, a genetic algorithm scheduling engine is used to process chromosome encodings, such as human resources, processing equipment, and equipment location, as well as a multi-objective fitness function. Real-time data streams, such as surgical scheduling, environmental sensors, and an ethics database, are input, while dynamic transport scheduling, human-machine ergonomics processing, thermodynamic sensing storage, and clustering-based compliant delivery are executed. A digital twin feedback loop is also set up to correct scheduling parameters in real time. This embodiment's closed-loop system simultaneously achieves a shift from passively receiving samples to proactively deploying resources based on surgical prediction.

[0036] Specifically, the refrigerator storing samples was divided into a high-frequency zone and a low-frequency stable zone. The high-frequency zone is located near the refrigerator door, where thermal shock is high, while the low-frequency stable zone is located deep within the refrigerator. Based on this partitioning, clustered sample retrieval was planned, routing samples required for different research applications along a single path for simultaneous retrieval with a single door opening. This reduced the number of times the refrigerator door was opened and the sample exposure time, thus extending the sample shelf life.

[0037] This invention achieves human-machine collaborative scheduling through a digital twin pressure balancing system and precise human prediction, ensuring high-precision task quality while improving overall processing efficiency. A fragmentation sorting scheme under freeze-thaw constraints and a clustering-optimized retrieval strategy reduce the number of refrigerator door openings and sample exposure time, extending sample shelf life. Real-time linkage between gene locus locking and informed consent meets biobank ethical compliance requirements. By constructing a closed-loop architecture encompassing the physical, perception, decision-making, and execution layers, it achieves synergistic optimization of sample viability, resource efficiency, and ethical compliance. This realizes a shift from passively receiving samples to proactively deploying them based on surgical prediction.

[0038] Example 3 An electronic device, comprising: Processor and memory; The processor executes the steps of the sample library full-process scheduling method based on genetic algorithm and biological stability modeling, as described in any of Embodiment 1, by calling programs or instructions stored in memory.

[0039] Example 4 A computer-readable storage medium includes computer program instructions that cause a computer to perform the steps of a sample library full-process scheduling method based on genetic algorithms and biological stability modeling as described in any of Embodiment 1.

[0040] Computer-readable storage media may take the form of any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may, for example, include, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0041] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; 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 or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention.

Claims

1. A sample library full-process scheduling method based on genetic algorithm and biological stability modeling, characterized in that, Includes the following steps: Step S1: Construct a genetic algorithm scheduling engine driven by a multi-objective fitness function, encode human resources, processing equipment, and storage space in the sample library as binary chromosomes, and set the multi-objective fitness function with the indicators of sample activity retention rate, equipment and human resource utilization rate, and thermodynamic fluctuation minimization. Step S2: Under the control of the genetic algorithm scheduling engine, a sample bioactivity decay model is established, the temperature of the transport box, sample type and ex vivo time are collected in real time, the remaining golden processing time of each batch of samples is calculated, and the priority of sample processing is dynamically adjusted based on the remaining golden processing time. The priority information is fed back to the genetic algorithm scheduling engine to update chromosome fitness. Step S3: Based on the priority information dynamically adjusted in step S2, construct a laboratory digital twin pressure balancing system, associate it with the hospital surgical scheduling system, dynamically adjust the emergency buffer ratio of automated equipment according to the expected sample output, monitor the continuous operation time of the experimenter and the error rate prediction, and automatically schedule high-precision samples to automated robots when it is determined that the manual accuracy has decreased, and output the scheduling result as a prerequisite for storage optimization. Step S4: Based on the scheduling results output in step S3, a thermodynamic stability sensing algorithm is introduced to partition and store samples according to the sample outbound frequency. High-frequency samples are stored in the front area of ​​the refrigerator, and low-frequency samples are stored in the deep area of ​​the refrigerator. During periods of low system load, the genetic algorithm scheduling engine automatically generates a fragmentation sorting scheme. The fragmentation sorting scheme is subject to freeze-thaw constraints to ensure that the sample exposure time during the sorting process is lower than a preset threshold. Step S5: Based on the sample partitioning and fragmentation completed in step S4, a clustering algorithm is used to optimize multiple outbound applications by storage location, enabling multiple outbound tasks to be completed in a single opening, and real-time association with the informed consent management system to lock the gene loci of samples that have been withdrawn due to ethical reasons.

2. The sample library full-process scheduling method based on genetic algorithm and biological stability modeling according to claim 1, characterized in that, In step S1, the sample viability retention rate is calculated based on the sample bioactivity decay model.

3. The sample library full-process scheduling method based on genetic algorithm and biological stability modeling according to claim 2, characterized in that, In step S1, the expression for the multi-objective fitness function is: ； Wherein, Q is the sample activity retention rate, which is calculated based on the sample biological activity decay model and the predicted activity value of each batch of samples at the end of the scheduling period is taken as the average; E is the equipment and manpower utilization rate, which includes the weighted sum of equipment occupancy rate and the workload balance of the experimenters; C is the thermodynamic fluctuation minimization index, which is calculated based on the number of times the refrigerator door is opened, the duration of the door opening, and the integral of the temperature fluctuation during the door opening; w1, w2, and w3 are the corresponding weighting coefficients.

4. The sample library full-process scheduling method based on genetic algorithm and biological stability modeling according to claim 3, characterized in that, In step S2, the dynamic adjustment of sample processing priority includes: when it is detected that the biological activity decay rate of the sample exceeds a preset threshold due to abnormal transport, the genetic algorithm scheduling engine performs priority site mutation on the chromosome corresponding to the sample through mutation operator, so that its processing priority crosses all regular samples in the current queue and enters the immediate processing channel.

5. The sample library full-process scheduling method based on genetic algorithm and biological stability modeling according to claim 4, characterized in that, In step S3, the digital twin pressure balancing system includes: a real-time mirrored laboratory physical environment, including equipment operating status, operator positions, and a queue of samples to be processed; the hospital surgical scheduling system provides predicted data on the amount of specimens produced within a preset time period; and the genetic algorithm scheduling engine adjusts the emergency buffer ratio of automated equipment in advance based on the predicted data.

6. The sample library full-process scheduling method based on genetic algorithm and biological stability modeling according to claim 5, characterized in that, In step S3, the method for determining the decline in human precision includes: monitoring the continuous operation time of the experimenter, and triggering a fatigue warning when the continuous operation time exceeds a first threshold; monitoring the recent operation error rate of the experimenter, and triggering a precision decline determination when the error rate exceeds a second threshold; when the fatigue warning and the precision decline determination are triggered simultaneously, the high-precision sample is automatically scheduled to the automated robot.

7. The sample library full-process scheduling method based on genetic algorithm and biological stability modeling according to claim 1, characterized in that, In step S4, the freeze-thaw constraint includes: in the fragment sorting scheme generated by the genetic algorithm, the cumulative exposure time of each sample from its original location to its re-entry into the target location does not exceed a preset threshold. Sorting schemes that exceed this threshold are eliminated during the iteration of the genetic algorithm and do not proceed to subsequent optimization.

8. The sample library full-process scheduling method based on genetic algorithm and biological stability modeling according to claim 1, characterized in that, In step S4, the system low-load period is the period when the sample library operation frequency is lower than a preset threshold, which is determined by statistical analysis of historical operation data. This includes the early morning period every day. During the low-load period, the genetic algorithm automatically triggers the generation and execution of the defragmentation scheme.

9. The sample library full-process scheduling method based on genetic algorithm and biological stability modeling according to claim 8, characterized in that, In step S5, the clustering optimization includes: obtaining the sample storage location coordinates corresponding to each application in multiple outbound applications, using the K-means clustering algorithm to group samples with adjacent storage locations into the same outbound batch, planning a single door opening path to traverse all sample storage locations within the batch, and outputting the optimal door opening path sequence.

10. The sample library full-process scheduling method based on genetic algorithm and biological stability modeling according to claim 9, characterized in that, In step S5, the gene locus locking includes: real-time synchronization of the ethical status of the informed consent management system; when the withdrawal of the informed consent form corresponding to any of the samples is detected, all gene locus data of that sample in the sample bank management system are marked as inaccessible, and any form of decoding, data export and data retrieval operations are prohibited until the ethical status is restored.

Images