An organoid biological sample live bank management system based on artificial intelligence

Abstract

The application relates to the field of organoids, in particular to an organoid biological sample live bank management system based on artificial intelligence, which comprises an organoid preparation module for monitoring prepared organoid samples; a label module for generating a first management label according to morphological scores and auxiliary scores in a preparation process; a quality evaluation module for performing quality scoring on finally prepared organoid samples by adopting preset organoid sample evaluation rules, and generating a second management label for the organoid samples according to the quality scores; and an organoid management module for generating a risk label according to the first management label and the second management label. The system provided by the application can support whole life cycle information management of organoids, including sample collection, culture, subculture, cryopreservation, recovery, quality evaluation and the like, so that standardized preservation and efficient utilization of organoid resources are realized, and the deep development of precise medical treatment and scientific research transformation is supported.

Description

[0001] This application claims priority to Chinese patent application filed on June 16, 2025 [Application No.: 2025108004545, Invention Title: An Artificial Intelligence-Based Organoid Biobank Management System], which is incorporated herein by reference in its entirety. Technical Field

[0002] This invention relates to the field of organoids, and more specifically to an artificial intelligence-based organoid biobank management system. Background Technology

[0003] With the development of precision medicine and personalized treatment, the application of organoid technology in disease modeling, drug screening, and regenerative medicine is rapidly expanding. An organoid is a miniature tissue structure derived from stem cells or tissue cells, capable of self-organizing and forming under three-dimensional in vitro culture conditions, possessing the corresponding organ structure and function. Compared to traditional two-dimensional cell models, organoids better preserve the genetic information and pathological characteristics of the original tissue, and are therefore widely used in tumor research, genetic disease modeling, and personalized medicine evaluation.

[0004] As organoid research expands, research and medical institutions have accumulated a large number of organoid samples from different individuals, tissue sources, and disease states. These samples exhibit high individual variability and clinical application potential; however, an excessive number of samples can lead to increased difficulty in sample management and greater sample loss due to poor management.

[0005] Existing organoid sample banks suffer from the following technical deficiencies: 1. Insufficient storage and viability maintenance: Traditional cryopreservation leads to a decline in organoid viability, and lacks a dynamic culture environment to simulate real physiological conditions. 2. Quality assessment relies on manual methods: The morphological and functional assessment of organoids largely depends on microscopic observation, which is highly subjective and inefficient.

[0006] For example, patent application CN202211573548.6 discloses a management system and method for constructing and storing organoid live bank samples, including: a surgery list information module, used to select surgical patients to be sampled based on the current surgery list, and generate a sample collection data package based on the surgical patient's identity; a sample collection workstation module, used to obtain the surgical patient's basic information and surgical information based on the sample collection data package, generate a sample collection number, print the corresponding sample collection instruction label, and affix it to the sample tube; a recording module; a work order distribution module; a tissue organoid preparation module; a blood organoid preparation module; a drug sensitivity test information management module; an HE staining quality control management module; and an immunohistochemistry test information management module.

[0007] However, the applicant noted that due to the unique nature of organoid culture, this method of constructing and storing live bank samples would face high implementation costs, making it difficult to promote and apply. Summary of the Invention

[0008] This invention provides an organoid biobank management system, comprising:

[0009] An organoid preparation module for monitoring prepared organoid samples, wherein the organoid preparation module includes:

[0010] A monitoring unit is configured to monitor the organoid sample using a first monitoring data range, the first monitoring data range including: image information and at least one feature information, and the at least one feature information including: environmental information and / or operating procedure information, the environmental information including at least one of the following types: temperature, pH, oxygen concentration;

[0011] A morphological scoring unit is used to generate a morphological score based on the image information and the expected morphological standard of the organoid sample, and the morphological score is used to define at least one of the following morphological indicators: proliferation rate, degree of organoid morphological abnormality, and degree of cell morphological abnormality.

[0012] An auxiliary scoring unit is used to generate an auxiliary score based on the feature information and the expected preparation criteria of the organoid sample;

[0013] A monitoring and adjustment unit is configured to generate a suggested monitoring scheme based on the morphological score and the auxiliary score; wherein, the monitoring and adjustment unit is configured to:

[0014] When the morphological score is within a first morphological range and the auxiliary score is within a first auxiliary range, a first suggested solution is generated, which suggests using a first monitoring data range for monitoring.

[0015] When the morphological score is in the second morphological range and the auxiliary score is in the first auxiliary range, or when the morphological score is in the first morphological range and the auxiliary score is in the second auxiliary range, a second suggested solution is generated. The second suggested solution suggests expanding the first monitoring data range to a first degree. The expanded monitoring range includes: a type of biological information, and the type of biological information includes at least one of the following: nutrient consumption and metabolic waste accumulation.

[0016] A tagging module is used to generate a first management tag based on the morphological score and the auxiliary score during the preparation process;

[0017] The quality assessment module is used to score the quality of the finally prepared organoid samples using preset organoid sample assessment rules, and generate a second management label for the organoid samples based on the quality score.

[0018] The organoid management module is used to generate a risk label based on the first management label and the second management label.

[0019] In some embodiments, the monitoring and adjustment unit is used for:

[0020] When the morphological score is in the second morphological range and the auxiliary score is in the second auxiliary range, a third suggested solution is generated. The third suggested solution proposes to expand the first monitoring data range to a second degree, and the expanded monitoring range includes: Class I biological information and Class II biological information, and the Class II biological information includes: gene testing information; wherein, the second degree is greater than the first degree.

[0021] In some embodiments, the organoid samples are further associated with sample category tags, and the sample category tags record the following information: sample quantity and / or sample passage number; correspondingly, the system further includes:

[0022] The dynamic monitoring module is used to generate a suggested attention level based on the sample quantity and / or the number of sample passages. When the sample quantity is lower than a preset sample quantity threshold, or when the number of sample passages is greater than a preset number of sample passages, the suggested attention level is increased. When the suggested attention level is greater than a preset attention level, the expansion level in the suggested scheme is updated.

[0023] In some embodiments, the system further includes: a condition update module, configured to monitor the organoid sample using a second monitoring data range during a second period; generate a biochemical score based on the first type of biological information and the first reference biological information; and generate preparation suggestions based on the biochemical score, morphological score, and auxiliary score.

[0024] In some embodiments, the risk labeling includes: a suggested risk period; correspondingly, the organoid management module includes:

[0025] The period recognition unit is used to mark the period as a morphologically abnormal period when the morphological score in a period exceeds the first morphological range, and / or to mark the period as a feature abnormal period when the auxiliary score in a period exceeds the first auxiliary range.

[0026] A term generation unit is used to generate a corresponding expected risk term based on the number of the morphological abnormality cycles and / or the number of the characteristic abnormality cycles.

[0027] The term correction unit is used to correct the expected risk term based on the second management label to obtain the suggested risk term.

[0028] In some embodiments, it also includes:

[0029] A storage module is used to generate a storage plan for the organoid sample based on the risk label, and the storage plan includes: a suggested risk period and a storage area.

[0030] In some embodiments, the storage module further includes:

[0031] The first storage unit is used to mark the organoid samples whose risk marker is greater than a preset first risk threshold as a type of risk sample, and the type of risk sample is recommended to be allocated to the first storage space.

[0032] The second storage unit is used to mark organoid samples whose risk marker is less than or equal to the first risk threshold as Class II risk samples, and the Class II risk samples are recommended to be allocated to the second storage space.

[0033] In some embodiments, the organoid samples are labeled with a sample category tag, and the sample category tag records the following information: sample quantity and / or sample passage number;

[0034] Correspondingly, the storage module further includes:

[0035] A category identification unit is used to identify the sample category label of the risk sample.

[0036] The third storage unit is used to mark the organoid samples as three types of risk samples when the number of samples is lower than a preset sample number threshold, or when the number of sample passages is greater than a preset number of sample passages, and the three types of risk samples are recommended to be allocated to the third storage space.

[0037] In some embodiments, it also includes:

[0038] The invocation module is used to obtain the invocation request input by the user, the invocation request including: invocation time and experiment category; generate recommended organoid samples based on the invocation request and the risk markers; generate recommended resuscitation procedures based on the risk markers; and generate suggested invocation schemes based on the recommended organoid samples and the recommended resuscitation procedures.

[0039] In some embodiments, it also includes:

[0040] The scheme output module is used to record the performance capability of a sample when a user conducts a sample experiment using at least one organoid sample. The performance capability is defined by at least one of the following indicators: recovery rate, survival rate, and resuscitation activity. Based on the performance capability, an updated risk period for the organoid sample is generated. When the difference between the updated risk period and the original recommended risk period is greater than a period threshold, the recommended risk period is updated to the updated risk period.

[0041] Beneficial technical effects:

[0042] In this embodiment, a monitoring scheme based on a restricted dynamic expansion mechanism is provided to address the large-scale needs for organoid sample culture and supervision. This scheme works in conjunction with conventional organoid sample evaluation methods during the monitoring and storage phases to provide a risk-based classification management system for organoid samples. This risk-based classification management can, on the one hand, improve the pass rate of organoid samples to a certain extent (i.e., avoid discarding some low-quality samples, resulting in excessive losses), and on the other hand, by monitoring the risks of organoid samples, it can provide reasonable usage plans for the application process, so as to ensure that organoid samples are used effectively as much as possible.

[0043] In particular, the present invention employs a restrictive dynamic expansion mechanism for monitoring. Specifically, it classifies data based on data type and detection method to limit the scope of monitoring data. Furthermore, when the health status of the cultured organoid samples shows a downward trend or signs, the scope of monitoring data is gradually expanded in a restrictive manner.

[0044] From another perspective, this invention, through a dual monitoring mechanism of the preparation process and the end of preparation, can comprehensively score the latent and manifest defects of organoid samples.

[0045] Furthermore, this invention integrates the storage of organoid samples in partitions based on occult and visible defects to reduce the risk of cross-contamination of samples.

[0046] Furthermore, the present invention can also monitor experimental data during the application of organoids to dynamically adjust risk management.

[0047] In other words, the present invention provides a more refined recommendation for organoid storage by integrating fluctuation factors in the organoid preparation process (such as abnormal cycles) with organoid sample evaluation rules (i.e., industry standard specifications for organoids). This refined recommendation can comprehensively assess observable and unobservable potential problems in organoids through fluctuation factors and evaluation rules (for example, evaluation rules can often assess relatively intuitive parameters such as activity, while fluctuation factors are beneficial for indirectly scoring some difficult-to-predict or observable problems), thereby improving the tolerability of organoid storage processes or organoids to a certain extent.

[0048] Here, "tolerance" can be understood as the degree to which the tolerance for organoid quality can be increased to a certain extent. For example, even if the quality level of the organoid is on the edge of the acceptable range (it can be called a marginal sample), a suitable application scheme can still be recommended based on a comprehensive evaluation of the organoid and its preparation stage. That is, this invention can improve the utilization rate of marginal samples to a certain extent.

[0049] From another perspective, this invention can effectively improve the utilization rate of organoid samples and reduce losses (such as avoiding the premature discarding of low-quality / marginal samples). In other words, by introducing fluctuations in the preparation process to provide a more refined evaluation of the quality of organoid samples, this invention can, to some extent, reduce organoid losses and lower organoid application costs by optimizing organoid storage and application processes. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. The elements or parts in the drawings are not necessarily drawn to scale. Obviously, the drawings described below are some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0051] Figure 1 This is a schematic diagram of the architecture of the organoid biobank management system in Example 1;

[0052] Figure 2 This is a schematic diagram of the system architecture in an exemplary embodiment of the present invention;

[0053] Figure 3 Microscopic photographs showing bacterial contamination in a culture medium;

[0054] Figure 4 A photograph showing a turbid culture medium with clearly visible colonies;

[0055] Figure 5 The diagram illustrates the observations of adherent growth, cell flattening and growth arrest, organoid disintegration into single cells, and necrosis in the central region of the organoid.

[0056] Figure 6 The diagrams show cell morphology under the conditions of being too small, cell senescence, apoptosis, and organoid disintegration-structure loss.

[0057] Figure 7 This is a schematic diagram showing the comparison of initial sample quality and morphology during the cultivation process.

[0058] Figure 8 These are schematic diagrams illustrating the morphology under normal conditions, abnormal budding, and irregular shapes.

[0059] Figure 9 The diagram shows the morphological characteristics of cells cultured at different passage numbers;

[0060] Figure 10 A schematic diagram showing the comparison of proliferation rates of organoids of different generations is presented;

[0061] Figure 11 This demonstrates the impact of cryopreservation on organoid culture;

[0062] Figure 12 This diagram illustrates the changes in organoid area over different resuscitation days. Detailed Implementation

[0063] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0064] In this document, suffixes such as "module," "part," or "unit" used to denote elements are used only for the purpose of illustrative purposes and have no specific meaning in themselves. Therefore, "module," "part," or "unit" may be used interchangeably.

[0065] In this document, the terms "upper," "lower," "inner," "outer," "front," "rear," "one end," and "the other end," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the present invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the present invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0066] In this document, unless otherwise explicitly specified and limited, the terms "installed," "equipped with," "connected," etc., should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, a direct connection, or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0067] In this document, "and / or" includes any and all combinations of one or more of the listed related items.

[0068] In this article, "multiple" means two or more, that is, it includes two, three, four, five, etc.

[0069] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0070] As used in this specification, the term "about" typically means + / -5% of the value, more typically + / -4% of the value, more typically + / -3% of the value, more typically + / -2% of the value, even more typically + / -1% of the value, and even more typically + / -0.5% of the value.

[0071] In this specification, certain embodiments may be disclosed in a range-bound format. It should be understood that this "range-bound" description is merely for convenience and brevity and should not be construed as a rigid limitation on the disclosed range. Therefore, the description of a range should be considered as having specifically disclosed all possible subranges and the individual numerical values ​​within those ranges. For example, a description of the range 1-6 should be considered as having specifically disclosed subranges such as from 1 to 3, from 1 to 4, from 1 to 5, from 2 to 4, from 2 to 6, from 3 to 6, etc., and the individual numbers within those ranges, such as 1, 2, 3, 4, 5, and 6. This rule applies regardless of the breadth of the range.

[0072] Organoid biosamples (also referred to as organoids or organoids in this article): These are miniature simplified models with similar structure and function to real organs, constructed using in vitro three-dimensional culture technology from stem cells (including pluripotent stem cells or adult stem cells) or cells derived from specific tissues. These samples can self-organize and differentiate into various cell types, mimicking key features of the source organ (such as the brain, liver, intestine, kidney, etc.) in terms of cellular composition, spatial structure, and some physiological functions.

[0073] Example 1

[0074] See Figure 2 As shown, the present invention provides an organoid biobank management system 200, comprising the following steps:

[0075] Organoid preparation module 202, used for monitoring the prepared organoid samples, wherein the organoid preparation module includes:

[0076] The monitoring unit 2020 is used to monitor the organoid sample using a first monitoring data range, the first monitoring data range including: image information and at least one feature information, and the at least one feature information including: environmental information and / or operating procedure information, the environmental information including at least one of the following types: temperature, pH, oxygen concentration;

[0077] During the culture of organoid samples, the physiological state of the organoid samples can be continuously monitored.

[0078] For example, in some embodiments, one or more photographs of the sample can be captured using an imaging device. For instance, photographs of organoid samples can be acquired using microscopic imaging devices such as confocal microscopes, light-sheet fluorescence microscopes, phase-contrast microscopes, etc.

[0079] Preferably, the culture dish of the organoid sample can be photographed at regular intervals. Furthermore, partial sampling of the organoid sample can be performed for microscopic observation to observe the morphology of individual cells.

[0080] The morphological scoring unit 2022 is used to generate a morphological score based on the image information and the expected morphological standard of the organoid sample, and the morphological score is used to define at least one of the following morphological indicators: proliferation rate, degree of organoid morphological abnormality, and degree of cell morphological abnormality.

[0081] For example, in some embodiments, for a specific organoid sample, the standard morphology (i.e., the desired morphology) of the organoid sample at a specific culture time point and in a healthy state can be pre-stored. For example, the standard morphology usually has a certain standard range in terms of shape and size.

[0082] For example, in some embodiments, for a specific organoid sample, its overall shape (i.e., organoid morphology) should generally have a desired morphology (such as shape or size) at a specific culture time point. By using an AI model to identify the actual photographs, the difference between the organoid sample and the desired morphology can be identified and analyzed.

[0083] For example, in some embodiments, when organoid samples exhibit blurred edges, it may indicate necrosis of internal cells within the organoid sample.

[0084] The auxiliary scoring unit 2024 is used to generate an auxiliary score based on the feature information and the expected preparation criteria of the organoid sample;

[0085] For example, in some embodiments, the characteristic information is used as environmental information, such as temperature, pH, and oxygen concentration. Correspondingly, during the actual culture process, the values ​​of temperature, pH, and oxygen concentration in the culture environment can be monitored periodically. When the actual values ​​differ significantly from the values ​​set by the user (usually the experimental operator), a lower auxiliary score is generated. Conversely, when the actual values ​​differ slightly from the set values, the auxiliary score is relatively higher.

[0086] For example, in some embodiments, the feature information can be operational specification information, specifically the collected user operation parameters (such as whether organoid samples are observed regularly, or whether digestion time is strictly controlled (e.g., when using digestion solutions such as TryplE, the action time needs to be monitored), or whether the culture medium is changed regularly, etc.). By comparing with standard operating procedures, the user's experimental operation standardization can also be scored to obtain the auxiliary score.

[0087] Understandably, the feature information focuses on scoring the standardization of the environment or human intervention in the cultivation process, and the specific scoring mechanism can be set by the user according to their cultivation needs.

[0088] The monitoring and adjustment unit 2026 is used to generate a suggested monitoring scheme based on the morphological score and the auxiliary score; wherein, the monitoring and adjustment unit is used for:

[0089] When the morphological score is within a first morphological range and the auxiliary score is within a first auxiliary range, a first suggested solution is generated, which suggests using a first monitoring data range for monitoring.

[0090] When the morphological score is in the second morphological range and the auxiliary score is in the first auxiliary range, or when the morphological score is in the first morphological range and the auxiliary score is in the second auxiliary range, a second suggested solution is generated. The second suggested solution suggests expanding the first monitoring data range to a first degree. The expanded monitoring range includes: a type of biological information, and the type of biological information includes at least one of the following: nutrient consumption and metabolic waste accumulation.

[0091] In some embodiments, the first morphology range can be [90-100], the second morphology range can be [80-89], the first auxiliary range can be [90-100], and the second auxiliary range can be [80-89].

[0092] For example, in some embodiments, the organoid sample can be monitored using a first monitoring data range in the initial state, and when the morphological score ∈ [90-100] and the auxiliary score ∈ [90-100], it can preferably be kept within the first monitoring data range.

[0093] Furthermore, if the morphological score ∈ [90-100] and the auxiliary score ∈ [80-89], or if the morphological score ∈ [80-89] and the auxiliary score ∈ [90-100], it may be suggested to expand the current monitoring data range, such as expanding the first monitoring data range to a first degree (i.e., obtaining the second monitoring data range), and optionally the expanded range can be at least one of the following: nutrient consumption and metabolic waste accumulation.

[0094] Tag module 204 is used to generate a first management tag based on the morphological score and the auxiliary score during the preparation process;

[0095] For example, in some embodiments, the first management label can be the weighted sum of the morphological score and the auxiliary score, such as first management label = a * morphological score + b * auxiliary score, where a and b are the set first weight and second weight, respectively.

[0096] The quality assessment module 206 is used to score the quality of the finally prepared organoid sample using preset organoid sample assessment rules, and generate a second management label for the organoid sample based on the quality score.

[0097] Preferably, the organoid sample evaluation rules in this embodiment adopt the general testing standards in the field of organoid culture, which can be set by industry experts. Furthermore, the calculation weights in the scoring process can be freely adjusted by the user based on culture or storage requirements.

[0098] For example, the identification methods used may use or refer to the standards published in the group standard "Guidelines for the Preparation, Cryopreservation, Resuscitation and Identification of Human Breast and Breast Cancer Organoids".

[0099] Organoid management module 208 is used to generate a risk label based on the first management label and the second management label.

[0100] It should be noted that in this embodiment, the organoid preparation module 202 focuses on using limited factors to dynamically monitor the organoid samples throughout the entire organoid culture process, so as to make timely adjustments to the culture process of the organoid samples and improve the culture success rate.

[0101] Furthermore, in this embodiment, the organoid preparation module will play a crucial role in the preparation and storage processes:

[0102] 1) During the preparation process, a restrictive dynamic expansion mechanism is used for monitoring. Specifically, the data is classified from the perspective of data type and detection method to limit the range of monitoring data. When the health status of the organoid sample shows a downward trend or signs, the range of monitoring data is gradually expanded in a restrictive manner.

[0103] When organoid samples are in a relatively healthy culture state, AI image recognition technology is used to evaluate the overall morphology and cell morphology of the organoid samples, and to quickly detect indicators such as oxygen, pH, and temperature. This allows for comprehensive verification of the culture state from both physical morphology and biochemical parameters. At the same time, the range of selected monitoring data requires relatively little workload for monitoring, which is beneficial for controlling monitoring costs.

[0104] 2) Evaluate abnormal states during the cultivation process based on limited monitoring data (i.e., identify abnormal periods) and conduct a comprehensive evaluation of subsequent storage schemes based on the abnormal states.

[0105] Meanwhile, the quality assessment module 206 focuses on assessing the quality of the final, fully formed organoid sample to determine whether the organoid sample meets the requirements for subsequent drug screening tests.

[0106] It is understandable that in the actual cultivation process, especially for manufacturers selling organoid samples, the types and quantities of organoid samples involved are very large. Therefore, if problems arise in storage or management during the specific cultivation process, it will lead to serious economic losses.

[0107] To address this, this embodiment provides a monitoring scheme based on a restricted dynamic expansion mechanism to meet the needs of large-scale organoid sample culture and monitoring. This scheme works in conjunction with conventional organoid sample evaluation methods during the monitoring and storage phases to provide a risk-based classification management system for organoid samples. This risk-based classification management can, on the one hand, improve the pass rate of organoid samples to some extent (i.e., avoid discarding low-quality samples and causing excessive losses), and on the other hand, by monitoring the risks of organoid samples, it can provide reasonable usage plans during application, thereby ensuring that organoid samples are utilized effectively as much as possible.

[0108] From another perspective, this invention, through a dual monitoring mechanism of the preparation process and the end of preparation, can comprehensively score the latent and manifest defects of organoid samples.

[0109] Among them, obvious defects refer to problems that can be visually detected through conventional sample evaluation rules (such as low cell activity, low cell number, etc.), while latent defects refer to some hidden dangers that may be caused by accidental or non-standard operations during the preparation process (and these accidental or non-standard operations may be minor and difficult to detect or assess the degree of impact). For example, the organoid sample may appear to be in a relatively healthy state under the detection of sample evaluation rules, but it is prone to developing hidden dangers under long-term storage.

[0110] In some embodiments, the proliferation rate refers to the number of live cells in an organoid within a set time period, where live cells in an organoid refer to all surviving single cells that make up the organoid.

[0111] In some embodiments, the proliferation rate refers to the number of qualified organoids that grow within a set time period. Qualified organoids are those whose morphology conforms to their initial morphology.

[0112] Alternatively, in some embodiments, the proliferation rate may refer to the increase in volume of the organoid over a set period of time.

[0113] Preferably, in this invention, the first monitoring data range is based on the morphological information obtained from the image (such as morphological proliferation rate, degree of organoid morphological abnormality, degree of cell morphological abnormality, etc.) and limited biochemical parameters that are easy to detect quickly, such as temperature, pH, and oxygen concentration.

[0114] For example, in some embodiments, unsuitable pH values ​​or oxygen gradients (hypoxia or hyperxia) can affect the normal physiological functions and differentiation of cells, such as causing abnormal organoid morphology, blurred edges, necrotic areas inside, and decreased cell viability.

[0115] For example, in some embodiments, temperature fluctuations may cause organoids to grow slowly or exhibit abnormal morphology.

[0116] In some exemplary embodiments, the application process of AI image recognition technology is as follows:

[0117] First, organoid samples can be fluorescently labeled. For example, Calcein-AM can be used, which emits green fluorescence after being hydrolyzed by live cell esterases (active labeling). Another example is propidium iodide (PI), which only penetrates the cell membrane of dead cells and binds to DNA (dead labeling), thus marking dead cells in red. Yet another example is Hoechst 33342 (blue), which can be used to label all cell nuclei (and can be used to count cell numbers).

[0118] Correspondingly, multi-channel fluorescence microscopy images of organoid samples (such as a combined bright field / green / red / blue four-channel image) can be obtained and output to the AI ​​image recognition model.

[0119] In some embodiments, the photo can be denoised before being input into the AI ​​image recognition model, such as by using Gaussian filtering or wavelet transform to eliminate fluorescent background noise.

[0120] In some embodiments, the method further includes the step of standardizing the photographs, such as normalizing fluorescence intensity, to avoid differences in illumination between batches and improve the accuracy of AI recognition results.

[0121] The AI ​​image recognition model can be an existing image recognition model or it can be trained by the user. This invention does not impose any restrictions on this.

[0122] For example, in some embodiments, the image recognition model can employ a U-Net or Mask R-CNN network. The training process can be as follows: A large number of sample images (e.g., photos containing different culture states, such as dead, healthy, bacterial contamination, etc.) are acquired; the user manually labels the sample images (e.g., labeling healthy cells, dead cells, foreign objects, etc.); the sample photos are then input into the network for training to obtain the final image recognition model. Furthermore, in some embodiments, abnormal substances (e.g., colonies, hyphae) can also be identified based on the image information. If the content / area of ​​the identified abnormal substance exceeds a preset abnormality threshold, the auxiliary score can be corrected, such as by reducing the auxiliary score.

[0123] Below, with Figures 3-12 As shown, the application of image data-based data monitoring in this invention will be illustrated by example:

[0124] Figure 3 Images (a) and (b) show microscopic photographs of bacterial contamination in the culture medium, while images (c) and (d) show microscopic photographs of fungal contamination, respectively. It can be seen that when foreign matter such as bacteria or fungi is present, the contamination status can be analyzed through the geometric morphology of the foreign matter.

[0125] Figure 4 The contaminated culture medium appears turbid and shows obvious bacterial colonies.

[0126] Figure 5 (a), (b), (c), and (d) show the following conditions: organoid adherent growth, cell flattening and growth arrest, organoid disintegration into single cells, and necrosis in the central region of the organoid, respectively. When the AI ​​image model identifies a type of morphology or a similar tendency in the photo, a morphological score can be given in combination with the abnormal conditions.

[0127] Figure 6 (a), (b), (c), and (d) show schematic diagrams of cell morphology under conditions such as small size (DAY10, cultured to day 10), cell senescence, apoptosis, and structural loss due to organoid disintegration, respectively. Blue represents senescence staining, and green represents apoptosis staining.

[0128] Figure 7 (a) and (b) show the comparison of the initial sample quality during the culture process. In (a), the cell clusters are uniform, the cell activity is good, and the refractive index is good, indicating that the initial sample quality is better. In contrast, in (b), the cell clusters are not uniform, the cell activity is poor, and the refractive index is poor, indicating that the initial sample quality is worse.

[0129] Figure 7 (c) and (d) show the morphological changes. In (c), the organoid is a regular vacuolar shape with thin edges, which is the initial morphology in a healthy state. In (d), the organoid wall is thickened and a necrotic area appears in the center, which indicates that the morphological variation may be caused by the preparation process, resulting in reduced cell activity or even necrosis.

[0130] Figure 8 Figures (a), (b), and (c) show the morphological diagrams under normal, abnormal budding, and irregular shapes, respectively. In the initial state in (a), the organoids are uniform in size, have complete structure, and good refractive index. In (b), the organoids are irregular in shape and have abnormal budding. In (c), the organoids are irregular in shape and tend to grow in a planar manner.

[0131] Figure 9 The results show the effects of multiple passages on cell culture. (a) shows the culture results of P1 generation on day 7, which shows that the low-generation organoids have complete morphology and structure and grow normally. (b) shows the culture results of P20 generation on day 7, which shows that the high-generation organoids grow slowly and cannot form organoids.

[0132] Figure 10 The proliferation rates of organoids at different generations are shown, and it can be seen that higher-generation organoids may be prone to growth arrest.

[0133] Figure 11 The effects of cryopreservation on organoid culture are shown. (a) and (b) show the organoid culture status before cryopreservation and after thawing, respectively. In (a), the organoids are uniform in size, have complete structure, and good refractive index. In (b), it is shown that after thawing, the organoids are smaller in size and their morphology has changed.

[0134] Figure 12 The diagram shows the changes in organoid area under different resuscitation days, and it can be seen that there is a trend of slower proliferation rate after resuscitation.

[0135] In some embodiments, nutrient consumption refers to the consumption of essential nutrients such as glucose and amino acids in the culture medium.

[0136] For example, in some embodiments, the amount of metabolic waste accumulation may refer to the accumulation of metabolic waste such as lactic acid, which can lead to changes in pH, organoid growth arrest, and decreased cell viability.

[0137] In some embodiments, the monitoring and adjustment unit 2026 is used for:

[0138] When the morphological score is in the second morphological range and the auxiliary score is in the second auxiliary range, a third suggested solution is generated. The third suggested solution proposes to expand the first monitoring data range to a second degree, and the expanded monitoring range includes: Class I biological information and Class II biological information, and the Class II biological information includes: gene testing information, histopathological information, and specific biomarker information; wherein, the second degree is greater than the first degree.

[0139] For example, in some embodiments, "second degree greater than first degree" means that the number of categories of monitoring data expanded under the second degree is greater.

[0140] For example, in some embodiments, the second degree being greater than the first degree means that the reliability of the information dimensions of the expanded monitoring data under the second degree is higher, that is, the detection method will adopt a more comprehensive, refined or in-depth solution with more complete data dimensions, such as gene testing.

[0141] For example, in some embodiments, gene testing information may be the accumulation of mutations shown by gene sequencing, specifically single-cell transcriptomics information.

[0142] For example, in some embodiments, histopathological information may be histological morphology as shown by H&E staining or pathological marker identification as shown by IHC staining.

[0143] For example, in some embodiments, the specific marker may be the expression and localization of specific cell markers as shown by immunofluorescence staining.

[0144] In some embodiments, the organoid samples are further associated with sample category tags, and the sample category tags record the following information: sample quantity and / or sample passage number; correspondingly, the system further includes:

[0145] The dynamic monitoring module is used to generate a suggested attention level based on the sample quantity and / or the number of sample passages. When the sample quantity is lower than a preset sample quantity threshold, or when the number of sample passages is greater than a preset number of sample passages, the suggested attention level is increased. When the suggested attention level is greater than a preset attention level, the expansion level in the suggested scheme is updated.

[0146] For example, in some embodiments, updating the degree of expansion refers to increasing the degree of expansion.

[0147] For example, in some embodiments, a sample count below a preset sample count threshold may be considered a rare sample. In this embodiment, for rare samples and samples with a high number of generations, the selected monitoring data category will be appropriately expanded within an extended range to allow for focused monitoring of specific samples.

[0148] In some embodiments, the system further includes: a condition update module, configured to monitor the organoid sample using a second monitoring data range during a second period; generate a biochemical score based on the first type of biological information and the first reference biological information; and generate preparation suggestions based on the biochemical score, morphological score, and auxiliary score.

[0149] In other words, in this embodiment, when a type of biological information detection is initiated, preparation suggestions (such as increasing nutrient solution, adjusting the ratio of key cytokines in the culture medium, etc.) can be automatically generated based on the current monitoring data.

[0150] Preferably, in some embodiments, recommended preparation suggestions are pre-defined for different biochemical scores, morphological scores, and auxiliary score ranges. Specifically, these preparation suggestions are generated based on experimental operation data accumulated through long-term monitoring.

[0151] In some embodiments, the risk label includes: a suggested risk period; correspondingly, the organoid management module 208 includes:

[0152] The period recognition unit is used to mark the period as a morphologically abnormal period when the morphological score in a period exceeds the first morphological range, and / or to mark the period as a feature abnormal period when the auxiliary score in a period exceeds the first auxiliary range.

[0153] The term generation unit is used to generate a corresponding expected risk term based on the number of morphological abnormality cycles and / or the number of characteristic abnormality cycles; preferably, for a specific organoid sample, different expected risk terms are pre-corresponding to different numbers of morphological abnormality cycles and / or the number of characteristic abnormality cycles.

[0154] For example, the greater the number of morphological abnormality cycles and / or characteristic abnormality cycles, the closer the expected risk period, i.e., the shorter the recommended storage time. Conversely, when abnormalities are rare in organoid samples during preparation, the recommended storage time can be appropriately extended.

[0155] The term correction unit is used to correct the expected risk term based on the second management label to obtain the suggested risk term.

[0156] For example, in some embodiments, abnormal situations during the preparation process and the quality score of the final organoid sample can be used to jointly calculate the recommended risk period (i.e., storage time) from implicit and explicit perspectives, respectively.

[0157] For example, in some embodiments, the second management label can be used to define the risk level of organoid samples, such as high (superior quality), medium (moderate quality), and low (just acceptable quality, applicable to some less demanding drug screening tests). When the risk level of the second management label is greater than the preset first risk level, it can be suggested to increase the expected risk period; when the risk level of the second management label is less than the preset second risk level, it can be suggested to decrease the expected risk period.

[0158] In some embodiments, it also includes:

[0159] Storage module 210 is used to generate a storage scheme for the organoid sample based on the risk label, and the storage scheme includes: a suggested risk period and a storage area.

[0160] In some embodiments, the storage module further includes:

[0161] The first storage unit is used to mark the organoid samples whose risk marker is greater than a preset first risk threshold as a type of risk sample, and the type of risk sample is recommended to be allocated to the first storage space.

[0162] The second storage unit is used to mark organoid samples whose risk marker is less than or equal to the first risk threshold as Class II risk samples, and the Class II risk samples are recommended to be allocated to the second storage space.

[0163] It is understood that in this embodiment, risk markers can be used to comprehensively define the risk level of organoid samples. The higher the risk marker value, the higher the risk, and the more quickly the sample needs to be stored to shorten the preservation period.

[0164] In this embodiment, organoid samples in the same risk range are preferably divided into the same storage space. For example, risk samples of type I and risk samples of type II are placed in different storage cabinets to reduce the risk of cross-infection to a certain extent.

[0165] In some embodiments, the organoid samples are labeled with a sample category tag, and the sample category tag records the following information: sample quantity and / or sample passage number;

[0166] Correspondingly, the storage module 210 further includes:

[0167] A category identification unit is used to identify the sample category label of the risk sample.

[0168] The third storage unit is used to mark the organoid samples as three types of risk samples when the number of samples is lower than a preset sample number threshold, or when the number of sample passages is greater than a preset number of sample passages, and the three types of risk samples are recommended to be allocated to the third storage space.

[0169] Furthermore, for rare or high-risk samples that have been passaged many times, they will be further classified and stored to distinguish them from other regular samples, thereby further reducing the risk of cross-infection and controlling sample loss.

[0170] In some embodiments, it also includes:

[0171] The calling module 212 is used to obtain the calling requirements input by the user, the calling requirements including: calling time and experiment category; generate recommended organoid samples based on the calling requirements and the risk markers; generate recommended resuscitation processes (e.g., the type of cryosol and resuscitation fluid) based on the risk markers; and generate a suggested calling scheme based on the recommended organoid samples and the recommended resuscitation processes.

[0172] It is important to understand that organoids with low viability and poor condition are more susceptible to ice crystal damage during cryopreservation and thawing, resulting in significantly reduced survival rates and recovery speeds after thawing. Therefore, for these organoid samples, high-quality cryosols or thawing solutions are preferred.

[0173] For example, in some embodiments, the higher the risk value of the risk label, the higher the recommended level of recovery process, specifically the higher the recommended level of cryosol or recovery fluid.

[0174] In some embodiments, the user pre-rates the recovery performance of the cryocooler or revival fluid to classify them into different grades of cryocooler or revival fluid.

[0175] In some embodiments, it also includes:

[0176] The scheme output module 214 is used to record the performance capability of a sample when a user conducts a sample experiment using at least one organoid sample. The performance capability is defined by at least one of the following indicators: recovery rate, survival rate, and resuscitation activity. Based on the performance capability, an updated risk period for the organoid sample is generated. When the difference between the updated risk period and the original recommended risk period is greater than a period threshold, the recommended risk period is updated to the updated risk period.

[0177] In this embodiment, the risk markers or risk periods for organoid samples can also be dynamically updated.

[0178] From another perspective, the present invention also provides a data management and reuse module for integrating and analyzing the full life cycle data of the organoid biological samples and generating high-value application suggestions.

[0179] The data management and reuse module includes:

[0180] The data integration unit is used to acquire and structure the full lifecycle data of the organoid samples, the full lifecycle data including:

[0181] The image information, feature information, morphological score, and auxiliary score generated by the organoid preparation module;

[0182] The quality score generated by the quality assessment module;

[0183] The risk marker generated by the organoid management module;

[0184] And externally input historical experimental data, which includes at least one of the following: drug sensitivity test results, gene sequencing data, proteomics data, and metabolomics data.

[0185] The intelligent analysis unit is used to perform in-depth analysis and pattern mining on the structured full life cycle data using artificial intelligence models, in order to establish a correlation model between the biological characteristics, preparation process parameters, quality status and experimental performance of the organoid samples.

[0186] An application generation unit is used to receive analysis requirements input by the user and generate at least one application output based on the association model.

[0187] The analytical requirements include at least: experimental objectives or research directions;

[0188] The application output includes at least one of the following:

[0189] Suggested experimental protocol: This can be the original protocol output module, including sample recommendations, candidate drug recommendations, drug concentration ranges, drug action time and other specific experimental parameters recommendations, experimental cycle prediction, etc.

[0190] Recommended sample cohort: Based on the experimental objective, select and recommend one or more groups of organoid samples with specific biological characteristics or historical data performance;

[0191] Potential biomarkers or drug targets: Identify potential therapeutic targets or diagnostic biomarkers by analyzing the association between multi-omics data and drug response;

[0192] Drug sensitivity prediction: For a specific drug, predict the sensitivity or resistance of a specified organoid sample and provide a confidence level assessment;

[0193] Mechanism of action inference: Based on the sample's response pattern to the drug and its multi-omics characteristics, the possible mechanism of action of the drug is inferred.

[0194] It should be noted that the organoid biobank management system of the present invention, based on the monitoring of the entire life cycle of organoids (such as relevant data collected during the preparation and experimental stages), can provide a comprehensive data management and reuse solution. Alternatively, this embodiment provides an intelligent experimental assistance method based on an organoid biobank management system.

[0195] The following example, using the screening of a novel targeted drug (denoted as "Drug-X"), illustrates the data management and reuse process. This drug theoretically targets colorectal cancer carrying specific gene mutations (e.g., KRAS). The most suitable organoid samples need to be selected from a live organoid library for drug sensitivity testing, with the aim of predicting the drug's effectiveness.

[0196] Step 1: Integration of full lifecycle data (corresponding to the "Data Integration Unit")

[0197] A data integration unit was used to collect and structure data from all organoid samples derived from colorectal cancer in the live bank. Taking one sample, organoid numbered "CRC-088," as an example:

[0198] Preparation and quality control data acquisition:

[0199] From the organoid preparation module: The system retrieved all data from the preparation and amplification process of CRC-088. This includes continuous microscopic images recording its morphology as it gradually grew from a 3D spherical structure and began to sprout; environmental monitoring logs (characteristic information) showing that the temperature was consistently maintained at 37±0.1°C and the pH value between 7.2 and 7.4 during the culture period; and operational standardization logs showing that it fully complied with the Standard Operating Procedures (SOPs). Its morphological score averaged 95 points (out of 100) throughout the entire culture period, with an auxiliary score of 98 points, indicating that its preparation process was highly standardized and its growth status was good.

[0200] From the quality assessment module: After the final preparation of CRC-088, it underwent H&E staining and immunohistochemistry (IHC) verification, and its quality score was "Grade A" (the highest level).

[0201] From the organoid management module: The system has assigned a risk label of "low risk" to it and recommended a risk period of 12 months.

[0202] Historical experimental data acquisition:

[0203] The system further integrated historical experimental data from external inputs associated with CRC-088. The database showed that the sample had undergone whole-exome sequencing, confirming it carried a KRAS mutation, along with a TP53 mutation. Furthermore, it had been used in a drug sensitivity test for a standard chemotherapy drug (e.g., irinotecan), showing moderate sensitivity (IC50 value of 25 μM).

[0204] The data integration unit links all the above data—image features, culture parameters, scores, labels, risk markers, genotypes, and historical drug sensitivity results—to a unique sample ID, "CRC-088," forming a comprehensive, multi-dimensional data archive. This process is performed synchronously on all samples in the database.

[0205] Step Two: Analysis and Mining of Artificial Intelligence Models (corresponding to "Intelligent Analysis Unit")

[0206] The intelligent analysis unit deploys a pre-trained artificial intelligence model (e.g., a deep learning model combining graph neural networks and multi-task learning). This model has learned the full lifecycle data of thousands of organoid samples in the library.

[0207] Model training: During the training phase, the model learns the complex nonlinear relationship between "preparation process parameters (such as morphological score fluctuations)," "genomic features (such as specific mutations)," and "historical drug responses (IC50 value)."

[0208] Building a correlation model: Through training, a high-dimensional correlation model was successfully constructed. This model is able to: identify samples carrying KRAS mutations whose proliferation rates change in a specific pattern under certain culture conditions (e.g., slight fluctuations in oxygen concentration); discover that samples carrying KRAS and concurrently with TP53 mutations are generally more sensitive to drugs targeting the PI3K pathway; and establish correlations between sample passage number, risk markers, and their recovery rate and survival rate in drug trials.

[0209] Step 3: Application Generation and Decision Support (corresponding to the "Application Generation Unit")

[0210] Input the experimental objective and requirements into the application generation unit, and the application generation unit will output a report.

[0211] Experimental objective: "To test the efficacy of the novel targeted drug Drug-X against KRAS G12D-mutant colorectal cancer."

[0212] Demand types: "Recommended sample cohort" and "Drug sensitivity prediction".

[0213] After receiving the request, the application generation unit calls the pre-built association model and performs the following operations to ultimately generate an interactive report as the application output:

[0214] Generate a queue of recommendation samples:

[0215] Experimental Group: The system screened out three most suitable samples: CRC-088: Recommendation reason - KRAS and TP53 co-mutation, quality score A, risk marker low, historical data shows moderate sensitivity to standard chemotherapy, making it an ideal model for studying the efficacy of new drugs in the context of drug resistance. CRC-102: Recommendation reason - Carries only KRAS mutation, without other common driver gene mutations, can be used as a model for studying the specificity of this target. CRC-075: Recommendation reason - KRAS mutation, but its morphological score fluctuated slightly during preparation, risk marker moderate, can be used to assess the robustness of drugs on non-ideal samples.

[0216] Control group: The system automatically recommended two control samples: CRC-091: KRAS wild-type, used to verify the target specificity of the drug. NTC-005: organoid derived from normal colon tissue, used to assess the toxicity of the drug to normal cells.

[0217] Generate drug sensitivity predictions:

[0218] The system predicts the sensitivity of Drug-X based on its molecular structure and known target information, combined with the genomics and historical data of the recommended samples.

[0219] The prediction results show that CRC-088 and CRC-102 have predicted IC50 values ​​of 5 μM and 8 μM (high sensitivity) for Drug-X, respectively. However, CRC-091 (wild type) has a predicted IC50 value of 150 μM (resistance).

[0220] Generation of potential targets and mechanism inference:

[0221] The report includes a mechanism inference generated by the model: model analysis shows that the high sensitivity prediction of CRC-088 is strongly correlated with its TP53 co-mutation status. This may mean that the efficacy of Drug-X is affected by the integrity of the apoptosis pathway. It is recommended to monitor the expression changes of apoptosis-related proteins (such as Bcl-2 and Caspase-3) in subsequent experiments to verify this mechanism of action.

[0222] See below Figure 1 As shown, another set of module labeling systems is used to explain the architecture of the management system in an exemplary embodiment:

[0223] like Figure 1 As shown, this embodiment provides an organoid biobank management system 100, which includes:

[0224] The organoid preparation module 101 is configured to construct organoids to prepare organoid biological samples;

[0225] In some embodiments, the organoid biosamples are prepared by digesting normal or tumor tissue from the subject and then mixing it with matrix gel to culture.

[0226] The organoid management module 102 is configured to manage organoid biological samples prepared by the organoid preparation module 101;

[0227] In some embodiments, the organoid management module 102 may include:

[0228] The information storage unit is configured to store first data, second data, and third data of the organoid biological sample.

[0229] In some embodiments, the first data may include one or more of sample source, culture parameters, single-cell transcriptomics information, and metabolomics information.

[0230] In some embodiments, the organoid management module 102 may further include:

[0231] A live bank storage unit is configured for storing the organoid biological samples in a live bank.

[0232] In some embodiments, the storage method of the live storage unit may be liquid nitrogen storage.

[0233] In some embodiments, the organoid management module 102 may further include:

[0234] The resuscitation unit is configured to resuscitate organoid biological samples stored in the live bank storage unit.

[0235] In some embodiments, the organoid management module 102 may further include:

[0236] The first adjustment unit is configured to predict the timing of cryopreservation / thawing of the organoid biological sample in order to increase the survival rate of the organoid biological sample and reduce ice crystal damage, etc.

[0237] In some embodiments, the organoid management module 102 can be implemented in an automated manner.

[0238] In some embodiments, the system 100 may further include:

[0239] The dynamic monitoring module 104 is configured to dynamically monitor the status of organoid biological samples being constructed in the organoid preparation module 101.

[0240] In some embodiments, the dynamic monitoring module 104 may include:

[0241] An image acquisition unit is configured to acquire image data of the organoid biological sample;

[0242] In some embodiments, the data collection is periodic, such as every 12 hours or every 24 hours.

[0243] In some embodiments, the resolution of the image data may be 2048×2048.

[0244] In some embodiments, the dynamic monitoring module 104 may further include:

[0245] The image recognition unit is configured to analyze the image data using an image recognition algorithm to obtain second data of the organoid biological sample.

[0246] In some embodiments, the second data may include one or more of the following: organoid species, morphology, proliferation status, and viability.

[0247] In some embodiments, improved U-Net networks, convolutional neural networks (CNNs), superpixel segmentation algorithms, and other methods can be used to perform real-time analysis of the image data for automatic classification (e.g., intestinal / lung organoids) and abnormal state identification (e.g., identifying necrotic areas through changes in vitality).

[0248] In some embodiments, the dynamic monitoring module 104 may further include:

[0249] The culture system monitoring unit is configured to monitor the dynamic changes of the corresponding culture system of the organoid biological sample constructed in the organoid preparation module 101, and obtain the third data of the organoid biological sample.

[0250] In some embodiments, the third data may include one or more of the following: the pH of the culture medium, the composition of cytokines in the culture medium, metabolites (e.g., glucose, lactate), cell damage markers (e.g., LDH), inflammatory factors (e.g., TNF-α), and microorganisms.

[0251] In some embodiments, the dynamic monitoring module 104 may further include:

[0252] The early warning unit is configured to adjust the real-time status of the organoid biological sample according to a preset early warning mechanism.

[0253] In some embodiments, the early warning mechanism includes a first early warning mechanism.

[0254] In some embodiments, the first warning mechanism includes: when at least one of the second data and / or the third data deviates from a first preset value (e.g., necrotic area > 10%), adjusting the organoid biological sample in real time by adding a regulating substance to the culture system.

[0255] In some embodiments, the regulatory substance may include cytokines, culture medium, etc.

[0256] In some embodiments, the early warning mechanism may further include a second early warning mechanism.

[0257] In some embodiments, the second early warning mechanism includes: triggering an alarm mechanism to remind researchers to check the status of the organoid biological sample when at least one of the second data and / or the third data deviates from a second preset value (e.g., necrotic area > 20%).

[0258] In some embodiments, the dynamic monitoring module 104 may further include:

[0259] The culture system conditioning unit is configured to add conditioning substances to the culture system according to conditioning requirements.

[0260] In some embodiments, the adjustment requirement may be based on the first early warning mechanism.

[0261] In some embodiments, the regulation requirements may be based on the culture requirements of the organoid biological sample, such as organoid functional maturity.

[0262] In some embodiments, the system 100 may further include:

[0263] The quality assessment module 106 is configured to assess the quality of the organoid biological samples prepared in the organoid preparation module 101.

[0264] In some embodiments, the quality assessment module 106 is configured to assess the similarity between the organoid biological sample and a real sample based on a constructed organoid fidelity assessment model.

[0265] In some embodiments, the organoid fidelity assessment model can calculate a similarity score with a real sample based on one or more of the histological identification results, genetic identification results, and marker staining results of the organoid biological sample (e.g., ASC organoids need to have a >90% adult mapping rate).

[0266] In some embodiments, the system 100 may further include:

[0267] The microenvironment simulation module 108 is configured to dynamically regulate one or more of temperature, oxygen gradient, and fluid shear force in the organoid preparation module 101 to better simulate the tumor microenvironment or organ development conditions.

[0268] In some embodiments, the temperature control range may be 37±0.5℃.

[0269] In some embodiments, the oxygen regulation range may be 0.5%-21%.

[0270] In some embodiments, the range of fluid shear force adjustment can be 0.1-1 kPa.

[0271] In some embodiments, the organoid preparation module 101 may be disposed in a microfluidic chip.

[0272] In some embodiments, the microenvironment simulation module 108 can be implemented using a bioreactor.

[0273] It should be understood that the specific modules and units mentioned above can be implemented based on trained artificial intelligence models.

[0274] This invention provides an organoid biobank management system by combining AI image recognition, multimodal data fusion, and dynamic culture technology. The system supports the full lifecycle information management of organoids, including sample collection, culture, passage, cryopreservation, resuscitation, and quality assessment, thereby achieving standardized preservation and efficient utilization of organoid resources and supporting the in-depth development of precision medicine and scientific research translation.

[0275] The system provided by this invention avoids, to some extent, the limitations of traditional sample banks, such as insufficient storage and activity maintenance, low storage efficiency, reliance on manual and subjective quality assessment, scattered and difficult-to-integrate data, difficulty in data traceability, and low level of intelligence. At the same time, the system provided by this invention can effectively prepare and manage patient-specific organoid models, significantly shortening the drug development cycle and promoting personalized medicine.

[0276] Furthermore, this invention also complies with the ISO 20387 biobank certification standard, supports seamless integration with organ-on-a-chip and organoid databases (such as HEOCA), and has broad application scenarios in promoting disease modeling, drug screening, and regenerative medicine.

[0277] Example 2

[0278] Understandably, different monitoring data ranges or evaluation rules can be selected for different organoid samples (such as different cells or different drug screening needs). For example, the main monitoring data involved in the construction of a colorectal cancer organoid living bank and drug testing are as follows:

[0279] 1. Epithelial cells were isolated from intraoperative samples from patients to construct organoids with a diameter of 200 μm, with an initial survival rate of 92%.

[0280] 2. The image recognition unit detects a 3% necrotic area and adds a regulating substance to the culture system to regulate the organoid biological sample.

[0281] 3. After 7 days of dynamic culture, the fidelity score was 85% (adult colon mapping).

[0282] 4. After testing oxaliplatin (50μM), the system generated a drug sensitivity test report: apoptosis rate 35%, IC50 = 12.4μM.

[0283] Example 3

[0284] This invention also provides a method for managing organoid biobank samples, comprising the following steps:

[0285] S101, preparing organoid samples, wherein S101 includes the following steps:

[0286] S1011, the organoid sample is monitored using a first monitoring data range, the first monitoring data range including: image information and at least one feature information, and the at least one feature information includes: environmental information and / or operating procedure information, the environmental information including at least one of the following types: temperature, pH, oxygen concentration;

[0287] S1012, a morphological score is generated based on the image information and the expected morphological standard of the organoid sample, and the morphological score is used to define at least one of the following morphological indicators: proliferation rate, degree of organoid morphological abnormality, and degree of cell morphological abnormality.

[0288] S1013, Generate an auxiliary score based on the feature information and the expected preparation criteria of the organoid sample;

[0289] S1014, generate a suggested monitoring scheme based on the morphological score and the auxiliary score; wherein, S1014 includes the following steps:

[0290] When the morphological score is within a first morphological range and the auxiliary score is within a first auxiliary range, a first suggested solution is generated, which suggests using a first monitoring data range for monitoring.

[0291] When the morphological score is in the second morphological range and the auxiliary score is in the first auxiliary range, or when the morphological score is in the first morphological range and the auxiliary score is in the second auxiliary range, a second suggested solution is generated. The second suggested solution suggests expanding the first monitoring data range to a first degree. The expanded monitoring range includes: a type of biological information, and the type of biological information includes at least one of the following: nutrient consumption and metabolic waste accumulation.

[0292] S102, Generate a first management label based on the morphological score and the auxiliary score during the preparation process;

[0293] S103, The quality score of the finally prepared organoid sample is evaluated using a preset organoid sample evaluation rule, and a second management label is generated for the organoid sample based on the quality score.

[0294] S104, generate a risk label based on the first management label and the second management label.

[0295] In some embodiments, S1014 includes the step of:

[0296] When the morphological score is in the second morphological range and the auxiliary score is in the second auxiliary range, a third suggested solution is generated. The third suggested solution proposes to expand the first monitoring data range to a second degree, and the expanded monitoring range includes: Class I biological information and Class II biological information, and the Class II biological information includes: gene testing information; wherein, the second degree is greater than the first degree.

[0297] In some embodiments, it also includes:

[0298] The organoid samples are also associated with sample category tags, and the sample category tags record the following information: sample quantity and / or sample passage number;

[0299] A suggested attention level is generated based on the number of samples and / or the number of sample passages, wherein the suggested attention level is increased when the number of samples is lower than a preset sample number threshold or when the number of sample passages is greater than a preset number of sample passages.

[0300] If the suggested attention level is greater than the preset attention level, the expansion level in the suggested scheme will be updated.

[0301] In some embodiments, S101 further includes the step of:

[0302] During the second cycle, the organoid samples are monitored using a second monitoring data range;

[0303] A biochemical score is generated based on the aforementioned type of biological information and the first reference biological information;

[0304] Preparation suggestions are generated based on the biochemical score, morphological score, and auxiliary score.

[0305] In some embodiments, the risk label includes: a suggested risk period, and correspondingly, S104 includes:

[0306] If the morphological score in a period exceeds the first morphological range, the period is marked as a morphological abnormality period, and / or, if the auxiliary score in a period exceeds the first auxiliary range, the period is marked as a feature abnormality period.

[0307] The expected risk period is generated based on the number of the morphological abnormality cycles and / or the number of the characteristic abnormality cycles.

[0308] The expected risk period is adjusted based on the second management label to obtain the suggested risk period.

[0309] In some embodiments, the steps further include:

[0310] A storage scheme is generated for the organoid sample based on the risk labeling, and the storage scheme includes: a suggested risk period and a storage area.

[0311] In some embodiments, generating a storage scheme for the organoid samples based on the risk label includes the following steps:

[0312] The organoid samples whose risk marker is greater than a preset first risk threshold are marked as a type of risk sample, and the type of risk sample is recommended to be allocated to the first storage space.

[0313] Organoid samples whose risk marker is less than or equal to the first risk threshold are marked as Class II risk samples, and the Class II risk samples are recommended to be allocated to the second storage space.

[0314] In some embodiments, the organoid samples are labeled with a sample category tag, and the sample category tag records the following information: sample quantity and / or sample passage number;

[0315] Correspondingly, the method for generating and storing organoid samples based on the risk labeling further includes the following steps:

[0316] Identify the sample category label for the aforementioned type of risk sample;

[0317] When the number of samples is lower than a preset sample number threshold, or when the number of sample passages is greater than a preset number of sample passages, the organoid samples are marked as Class III risk samples, and the Class III risk samples are recommended to be allocated to the third storage space.

[0318] In some embodiments, it also includes:

[0319] Obtain the user's input request, which includes: request time and experiment category;

[0320] Based on the invocation requirements and the risk markers, a recommended organoid sample is generated;

[0321] A recommended recovery process is generated based on the risk markers;

[0322] A suggested invocation scheme is generated based on the recommended organoid samples and the recommended resuscitation process.

[0323] In some embodiments, the method further includes: when a user conducts a sample experiment using at least one organoid sample, recording the performance capability of the sample, wherein the performance capability is defined using at least one of the following indicators: recovery rate, survival rate, and resuscitation activity;

[0324] The update risk period for the organoid samples is generated based on the performance capability.

[0325] If the difference between the updated risk period and the original recommended risk period is greater than the period threshold, the recommended risk period will be updated to the updated risk period.

[0326] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a computer terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0327] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A management system for a live organoid biobank, characterized in that, include: An organoid preparation module for monitoring prepared organoid samples, wherein the organoid preparation module includes: A monitoring unit is configured to monitor the organoid sample using a first monitoring data range, the first monitoring data range including: image information and at least one feature information, and the at least one feature information including: environmental information and / or operating procedure information, the environmental information including at least one of the following types: temperature, pH, oxygen concentration; A morphological scoring unit is used to generate a morphological score based on the image information and the expected morphological standard of the organoid sample, and the morphological score is used to define at least one of the following morphological indicators: proliferation rate, degree of organoid morphological abnormality, and degree of cell morphological abnormality. An auxiliary scoring unit is used to generate an auxiliary score based on the feature information and the expected preparation criteria of the organoid sample; A monitoring and adjustment unit is configured to generate a suggested monitoring scheme based on the morphological score and the auxiliary score; wherein, the monitoring and adjustment unit is configured to: When the morphological score is within a first morphological range and the auxiliary score is within a first auxiliary range, a first suggested solution is generated, which suggests using a first monitoring data range for monitoring. When the morphological score is in the second morphological range and the auxiliary score is in the first auxiliary range, or when the morphological score is in the first morphological range and the auxiliary score is in the second auxiliary range, a second suggested solution is generated. The second suggested solution suggests expanding the first monitoring data range to a first degree. The expanded monitoring range includes: a type of biological information, and the type of biological information includes at least one of the following: nutrient consumption and metabolic waste accumulation. A tagging module is used to generate a first management tag based on the morphological score and the auxiliary score during the preparation process; The quality assessment module is used to score the quality of the finally prepared organoid samples using preset organoid sample assessment rules, and generate a second management label for the organoid samples based on the quality score. The organoid management module is used to generate a risk label based on the first management label and the second management label.

2. The system according to claim 1, characterized in that, The monitoring and adjustment unit is used for: When the morphological score is in the second morphological range and the auxiliary score is in the second auxiliary range, a third suggested solution is generated. The third suggested solution proposes to expand the first monitoring data range to a second degree, and the expanded monitoring range includes: Class I biological information and Class II biological information, and the Class II biological information includes: gene testing information, histopathological information, and specific biomarker information; wherein, the second degree is greater than the first degree.

3. The system according to claim 1, characterized in that, The organoid samples are also associated with sample category tags, and the sample category tags record the following information: sample quantity and / or sample passage number; correspondingly, the system also includes: The dynamic monitoring module is used to generate a suggested attention level based on the sample quantity and / or the number of sample passages. When the sample quantity is lower than a preset sample quantity threshold, or when the number of sample passages is greater than a preset number of sample passages, the suggested attention level is increased. When the suggested attention level is greater than a preset attention level, the expansion level in the suggested scheme is updated.

4. The system according to claim 1, characterized in that, Also includes: The condition update module is used to monitor the organoid samples using a second monitoring data range during the second period; A biochemical score is generated based on the aforementioned type of biological information and the first reference biological information; Preparation suggestions are generated based on the biochemical score, morphological score, and auxiliary score.

5. The system according to claim 1, characterized in that, The risk marker includes: a suggested risk period; correspondingly, the organoid management module includes: The period recognition unit is used to mark the period as a morphologically abnormal period when the morphological score in a period exceeds the first morphological range, and / or to mark the period as a feature abnormal period when the auxiliary score in a period exceeds the first auxiliary range. A term generation unit is used to generate a corresponding expected risk term based on the number of the morphological abnormality cycles and / or the number of the characteristic abnormality cycles. The term correction unit is used to correct the expected risk term based on the second management label to obtain the suggested risk term.

6. The system according to claim 5, characterized in that, Also includes: A storage module is used to generate a storage plan for the organoid sample based on the risk label, and the storage plan includes: a suggested risk period and a storage area.

7. The system according to claim 6, characterized in that, The storage module also includes: The first storage unit is used to mark the organoid samples whose risk marker is greater than a preset first risk threshold as a type of risk sample, and the type of risk sample is recommended to be allocated to the first storage space. The second storage unit is used to mark organoid samples whose risk marker is less than or equal to the first risk threshold as Class II risk samples, and the Class II risk samples are recommended to be allocated to the second storage space.

8. The system according to claim 7, characterized in that, The organoid samples are labeled with sample category tags, and the sample category tags record the following information: sample quantity and / or sample passage number; Correspondingly, the storage module further includes: A category identification unit is used to identify the sample category label of the risk sample. The third storage unit is used to mark the organoid samples as three types of risk samples when the number of samples is lower than a preset sample number threshold, or when the number of sample passages is greater than a preset number of sample passages, and the three types of risk samples are recommended to be allocated to the third storage space.

9. The system according to any one of claims 6-8, characterized in that, Also includes: The invocation module is used to obtain the invocation request input by the user, the invocation request including: invocation time and experiment category; generate recommended organoid samples based on the invocation request and the risk markers; generate recommended resuscitation procedures based on the risk markers; and generate suggested invocation schemes based on the recommended organoid samples and the recommended resuscitation procedures.

10. The system according to any one of claims 6-8, characterized in that, Also includes: The scheme output module is used to record the performance capability of a sample when a user conducts a sample experiment using at least one organoid sample. The performance capability is defined by at least one of the following indicators: recovery rate, survival rate, and resuscitation activity. Based on the performance capability, the module generates an update risk period for the organoid sample. If the difference between the updated risk period and the original recommended risk period is greater than the period threshold, the recommended risk period will be updated to the updated risk period.

Images