Sample state detection method applied to biological sample, computer device

Abstract

Embodiments of the present application disclose a sample state detection method applied to biological samples and a computer device. A specific embodiment of the method comprises: determining state detection configuration information for a target biological sample group; moving the target biological sample group to a sample storage device according to the storage configuration information to store the target biological sample; in response to reaching a sample detection time point, performing the following processing steps for the target biological sample: sucking the target biological sample to a slide making device by a pipetting device to generate a slide to be detected; controlling an electron microscope to collect electron microscope images of the slide to be detected at different image scales to obtain an electron microscope image group; generating sub-sample detection information for the target biological sample; and generating sample detection information for the target biological sample group according to the obtained sub-sample detection information group. The embodiment guarantees consistency of sample detection and greatly improves sample detection efficiency.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the field of computer technology, and in particular to a sample state detection method applied to biological samples and a computer device. BACKGROUND

[0002] For biological samples, complete verification, partial verification or cross verification are usually used to verify the performance of biological samples. Specifically, by setting up a control group and an experimental group, the performance of biological samples is observed. However, for more complex experiments, multiple groups of biological samples need to be observed manually, which is inefficient. In addition, due to the manual observation method, the observation results may be unreliable due to errors.

[0003] The above information disclosed in this BACKGROUND section is only for the purpose of enhancing the understanding of the background of the present inventive concepts, and therefore, it can contain information that is not prior art known to those of ordinary skill in the art. SUMMARY

[0004] The summary section of the present application is used to introduce the concepts in a simple form, which will be described in detail in the specific embodiments section. The summary section of the present application is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0005] Some embodiments of the present application propose a sample state detection method applied to biological samples, a computer device and a computer readable storage medium to solve one or more of the technical problems mentioned in the background section.

[0006] In a first aspect, some embodiments of the present application provide a sample state detection method applied to a biological sample, the method comprising: determining state detection configuration information for a target biological sample group, wherein the target biological sample group comprises at least three target biological samples arranged in parallel and stored in sample tubes, the target biological sample being a biological sample to be subjected to sample state detection, the state detection configuration information comprising: sample storage configuration information and sample detection configuration information, the sample detection configuration information comprising: a sample detection time point and sample detection index information; moving the target biological sample group to a sample storage device according to the storage configuration information to store the target biological samples; in response to reaching the sample detection time point, for each target biological sample in the target biological sample group, performing the following processing steps: using a pipetting device to aspirate the target biological sample to a slide making device to generate a detection slide; controlling an electron microscope to collect electron microscope images of the detection slide at different image scales to obtain an electron microscope image group; generating sub-sample detection information of the target biological sample according to the sample detection index information, the electron microscope image group, and a pre-trained sample detection model; and generating sample detection information of the target biological sample group according to the obtained sub-sample detection information group.

[0007] In a second aspect, the present application also provides a computer device, comprising a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, the method described in any of the implementations of the first aspect is implemented.

[0008] In a third aspect, the present application also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in any of the implementations of the first aspect is implemented.

[0009] The above embodiments of this application have the following beneficial effects: Through the sample state detection method for biological samples according to some embodiments of this application, a unified standard for sample storage and sample detection for biological sample groups is achieved. Specifically, firstly, state detection configuration information for a target biological sample group is determined. This target biological sample group includes at least three parallel target biological samples stored in sample tubes. The target biological samples are biological samples to be subjected to sample state detection. The state detection configuration information includes sample storage configuration information and sample detection configuration information. The sample detection configuration information includes sample detection time points and sample detection index information. Secondly, according to the storage configuration information, the target biological sample group is moved to a sample storage device for sample storage. This achieves unified storage for biological samples. Next, in response to reaching the sample detection time point, for each target biological sample in the target biological sample group, the following processing steps are performed: First, the target biological sample is aspirated into a slide preparation device using a pipette to generate a slide to be tested. This achieves automated quantitative pipetting and automated slide preparation. The second step involves controlling an electron microscope to acquire electron microscopic images of the slide under test at different image scales, resulting in an electron microscopic image set. This yields electron microscopic images from different fields of view. The third step involves generating sub-sample detection information for the target biological sample based on the sample detection index information, the electron microscopic image set, and a pre-trained sample detection model. By combining this with the sample detection model, automated sample detection based on the sample detection index information is achieved. Finally, based on the obtained sub-sample detection information set, sample detection information for the entire target biological sample group is generated. This provides an overall sample evaluation of the entire target biological sample group. This method ensures the consistency of sample detection while significantly improving sample detection efficiency. Attached Figure Description

[0010] The above and other features, advantages, and aspects of the embodiments of this application will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and elements are not necessarily drawn to scale.

[0011] Figure 1 These are flowcharts of some embodiments of a sample state detection method for biological samples according to this application;

[0012] Figure 2 This is a schematic diagram of the sample storage device;

[0013] Figure 3 This is an overall schematic diagram of the sample storage device, pipetting device, and slide preparation device;

[0014] Figure 4 This is a schematic diagram of the sample detection model structure;

[0015] Figure 5 This is a schematic diagram of the structure of a computer device suitable for implementing some embodiments of this application. Detailed Implementation

[0016] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While some embodiments of this application are shown in the drawings, it should be understood that this application can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this application. It should be understood that the drawings and embodiments of this application are for illustrative purposes only and are not intended to limit the scope of protection of this application.

[0017] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0018] It should be noted that the concepts of "first" and "second" mentioned in this application are only used to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0019] It should be noted that the terms "a" and "a plurality of" used in this application are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0020] The names of the messages or information exchanged between multiple devices in the embodiments of this application are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0021] The present application will now be described in detail with reference to the accompanying drawings and embodiments.

[0022] refer to Figure 1 The diagram illustrates a flow 100 of some embodiments of a sample state detection method for biological samples according to this application. The sample state detection method for biological samples includes the following steps:

[0023] Step 101: Determine the status detection configuration information for the target biological sample group.

[0024] In some embodiments, the execution entity (e.g., a computing device) of a sample state detection method applied to biological samples can determine state detection configuration information for a target biological sample group. The target biological sample group comprises at least three parallel target biological samples stored in sample tubes. The target biological samples are the biological samples to be subjected to sample state detection. The aforementioned state detection configuration information includes: sample storage configuration information and sample detection configuration information. The aforementioned sample detection configuration information includes: sample detection time point and sample detection index information. The sample storage configuration information characterizes the storage requirements of the target biological samples. The sample detection time point characterizes the time point at which the target biological samples begin to be detected. For example, the sample detection time point could be 48 hours after the target biological samples are configured. The sample detection index information characterizes the sample indices to be collected, such as sample activity and sample color. In practice, experimenters can configure the state detection configuration information for the target biological sample group through a graphical interface. Specifically, when the experiment is complex, there may be multiple groups of multiple target biological samples; therefore, experimenters can configure the state monitoring configuration information in batches through a graphical interface.

[0025] It should be noted that the aforementioned wireless connection methods may include, but are not limited to, 3G / 4G / 5G connections, WiFi connections, Bluetooth connections, WiMAX connections, Zigbee connections, UWB (ultra wideband) connections, and other currently known or future wireless connection methods.

[0026] It should be noted that the aforementioned computing devices can be either hardware or software. When the computing device is hardware, it can be implemented as a distributed cluster consisting of multiple servers or terminal devices, or as a single server or a single terminal device. When the computing device is software, it can be installed on the hardware devices listed above. It can be implemented as, for example, multiple software programs or software modules used to provide distributed services, or as a single software program or software module. No specific limitations are made here.

[0027] Step 102: According to the storage configuration information, move the target biological sample group to the sample storage device to store the target biological sample.

[0028] In some embodiments, the executing entity can move the target biological sample group to a sample storage device according to storage configuration information for sample storage. In practice, storage configuration information may include sample storage humidity and sample storage brightness. Therefore, the executing entity can store the target biological sample in the sample storage device according to the sample storage humidity and sample storage brightness. For example, the sample storage device may be a light-proof sample storage box.

[0029] Optionally, sample storage configuration information includes: sample storage temperature.

[0030] Optionally, the sample storage device includes a sample storage box. The sample storage box includes a first sample storage box and a second sample storage box. The first sample storage box includes a box body and a lid. The first sample storage box contains K sample tube placement chambers, a temperature sensor, and a pressure sensor. K ≥ 1. The first sample storage box is positioned above the second sample storage box. The second sample storage box is used to store a heat-conducting medium. In practice, the heat-conducting medium can be pure water. The second sample storage box includes a temperature sensor and a heating device.

[0031] As an example, see Figure 2 The schematic diagram of the sample storage device shown includes a first sample storage box and a second sample storage box 4. The first sample storage box includes a first sample storage box body 3 and a first sample storage box lid 1. The first sample storage box body contains K sample tube placement chambers 2. When the first sample storage box is in a sealed state, the first sample storage box lid 1 and the first sample storage box body 3 are fitted together and sealed. The first sample storage box body 3 is filled with inert helium gas. The second sample storage box includes a filling port 5 and a releasing port 6. Specifically, a heat-conducting medium can be added to the second sample storage box 4 through the filling port 5 and released through the releasing port 6. A temperature sensor is located inside the first sample storage box body 3. A pressure sensor is located between the first sample storage box lid 1 and the first sample storage box body 3, for example, on the inner surface of the first sample storage box lid 1, to detect whether the internal pressure of the first sample storage box is stable (whether there is leakage) when the first sample storage box is sealed. Furthermore, since the first sample storage box 3 contains an inert gas and the second sample storage box 4 contains a thermally conductive medium, and their thermal conductivity often differs, temperature sensors are installed in both the first sample storage box 3 and the second sample storage box 4 to ensure stable and precise control. Additionally, the second sample storage box 4 may also include an electric heating wire to heat the thermally conductive medium.

[0032] In some optional implementations of certain embodiments, the execution entity moves the target biological sample group to the sample storage device according to the storage configuration information to store the target biological sample, which may include the following steps:

[0033] The first step is to control the robotic arm to move the target biological sample group to the K sample tube placement chamber included in the first sample storage box.

[0034] In practice, the aforementioned executing entity can control a high-precision robotic arm to move the aforementioned target biological sample group into the K sample tube placement chamber included in the aforementioned first sample storage box.

[0035] The second step is to control the lid of the first sample storage box to close and to detect the airtightness of the box body by means of the air pressure sensor included in the first sample storage box body.

[0036] In practice, firstly, the aforementioned executing entity can control the closing of the first sample storage box lid using a high-precision robotic arm. Then, the first sample storage box is placed in a negative pressure environment, and pressure changes are detected by a pressure sensor included in the box. When the pressure signal is stable and the same as the pressure reading when not placed in a negative pressure environment, the airtightness is stable; when the pressure signal is different from the pressure reading when not placed in a negative pressure environment, the airtightness is unstable.

[0037] Third, in response to the stable airtightness of the first sample storage box, the heating device is controlled to heat the heat-conducting medium, and the temperature is monitored by the temperature sensor included in the first sample storage box and the temperature sensor included in the second sample storage box to obtain the first temperature and the second temperature.

[0038] In practice, the aforementioned executing entity can control the electric heating wire to heat the heat-conducting medium.

[0039] Fourth step: In response to the first temperature being the sample storage temperature, control the heating device to perform constant temperature heating at the second temperature.

[0040] Step 103: In response to the arrival of the sample detection time point, for each target biological sample in the target biological sample group, perform the following processing steps:

[0041] Step 1031: The target biological sample is aspirated into the slide preparation device using a pipette to generate the slide to be tested.

[0042] In some embodiments, the aforementioned entity can use a pipette to aspirate the target biological sample into a slide-making device to generate a slide for testing. In practice, the pipette can be a pipette.

[0043] Optionally, the pipetting device includes: a pipetting robotic arm, an electrically controlled pipette, and a positioning camera. The aforementioned slide preparation device includes: a slide stage and a coverslip machine. The coverslip machine is positioned above the slide stage. The coverslip machine uses a miniature vacuum suction cup to pick up coverslips.

[0044] As an example, see Figure 3The diagram shows an overall schematic of the sample storage device, pipetting device, and slide preparation device, including: a sample storage device 7, a pipetting device, and a slide preparation device 9. Specifically, a movable frame is provided above the sample storage device 7 and the slide preparation device 9. The movable frame moves vertically along a guide rail. A pipetting robotic arm 8 and a coverslip machine 13 are provided above the movable frame. The pipetting robotic arm 8 and the coverslip machine 13 are driven by a chain installed within the movable frame. An electrically controlled pipette 10 is connected below the pipetting robotic arm 8, wherein the volume of liquid aspirated is controlled electrically. A downward-facing positioning camera 11 is provided on the outer side of the pipetting robotic arm 8. The coverslip machine 13 includes a telescopic arm and a miniature vacuum suction cup 14. The telescopic arm is positioned between the coverslip machine 13 and the miniature vacuum suction cup 14. The miniature vacuum suction cup 14 holds a coverslip.

[0045] In some optional implementations of certain embodiments, the execution entity aspirates the target biological sample into a slide-making device using a pipette to generate a slide to be tested, which may include the following steps:

[0046] The first step is to control the opening of the lid of the first sample storage box.

[0047] The second step involves capturing the first image via a positioning camera in response to the opening of the box lid.

[0048] The first image is a top view of the aforementioned first sample storage box.

[0049] The third step is to determine the coordinates of the positioning point based on the first image above.

[0050] The coordinates of the aforementioned positioning points represent the position of the sample tube placement chamber corresponding to the target biological sample to be pipetted.

[0051] In practice, the aforementioned execution entity can use the Lite-Unet model, taking the first image as input and the coordinates of the positioning points as output.

[0052] The fourth step is to generate the first robotic arm's motion path based on the coordinates of the positioning points and the initial position of the pipetting robotic arm.

[0053] In practice, to ensure that the electronically controlled pipette accurately enters the sample tube and achieves the aspiration of the target biological sample, the coordinates of the positioning point can be used as the end point coordinates, and the initial position of the pipetting robot arm can be used as the starting point coordinates. Through trajectory smoothing, a micro-motion path can be generated as the first robot arm movement path.

[0054] Fifth step: Control the above-mentioned pipetting robot arm to move along the movement path of the first robot arm to the above-mentioned positioning point coordinates.

[0055] In practice, the aforementioned executing entity can control the moving frame and the pipetting robot arm to move along the movement path of the first robot arm to the aforementioned positioning point coordinates.

[0056] Step 6: In response to successful movement, the above-mentioned pipetting robotic arm controls the above-mentioned electronically controlled pipette to quantitatively aspirate the target raw sample.

[0057] Step 7: In response to successful absorption, the second robotic arm's motion path is generated based on the coordinates of the positioning point and the glass slide.

[0058] The aforementioned slide coordinates represent the coordinates of the center point of the slide on the aforementioned slide placement stage.

[0059] In practice, to ensure that the target biological sample is accurately dropped into the center of the slide, the aforementioned execution subject can use the center point coordinates of the slide placement stage as the end point coordinates and the positioning point coordinates as the starting point coordinates, and generate a micro-motion path through trajectory smoothing, which serves as the motion path for the second robotic arm.

[0060] Step 8: Control the above-mentioned pipetting robot arm to move along the movement path of the second robot arm to the coordinates of the glass slide.

[0061] The aforementioned executing entity can control the moving frame and the pipetting robot arm to move along the movement path of the second robot arm to the coordinates of the aforementioned slide.

[0062] Step 9: In response to successful transfer, control the electronic pipette to transfer the aspirated target sample onto the glass slide.

[0063] Step 10: In response to successful movement, control the coverslip machine to cover the slide with a coverslip, thus obtaining the slide to be tested.

[0064] In practice, firstly, the aforementioned executing entity can control the moving frame and coverslip machine to move above the glass slide. Then, it controls the extension of the telescopic arm to bring the glass slide into contact with the coverslip. Next, it releases the suction of the micro vacuum suction cup so that the coverslip completely covers the glass slide by gravity, thus obtaining the glass slide to be tested.

[0065] Step 1032: Control the electron microscope to acquire electron microscope images of the slide under test at different image scales to obtain an electron microscope image set.

[0066] In some embodiments, the aforementioned executing entity can control an electron microscope to acquire electron microscope images of the slide under test at different image scales, thereby obtaining a set of electron microscope images.

[0067] In practice, depending on different image acquisition needs, conventional optical microscopes can be used to acquire images of the slide under test at different image scales, resulting in an electron microscope image set. Specifically, the slide can be moved to the electron microscope's observation position manually, or a robotic arm can be used.

[0068] Step 1033: Generate sub-sample detection information for the target biological sample based on the sample detection index information, electron microscopy image set and pre-trained sample detection model.

[0069] In some embodiments, the aforementioned executing entity can generate sub-sample detection information for a target biological sample based on sample detection index information, electron microscopy image sets, and a pre-trained sample detection model.

[0070] Optionally, the sample detection model includes: an image enhancement model, a sample image encoding model, and a text encoding model.

[0071] In some optional implementations of certain embodiments, the execution entity generates sub-sample detection information for the target biological sample based on the sample detection index information, the electron microscopy image set, and the pre-trained sample detection model, which may include the following steps:

[0072] The first step is to perform image enhancement on each electron microscope image in the electron microscope image group using the image enhancement model described above, so as to generate enhanced electron microscope images and obtain the enhanced electron microscope image group.

[0073] The second step involves extracting image features from the enhanced electron microscope images in the enhanced electron microscope image group using the sample image coding model described above, in order to generate image features and obtain an image feature group.

[0074] The third step is to extract text features from the above sample detection index information using the text encoding model described above, in order to generate text features.

[0075] The fourth step is to generate sub-sample detection information for the target biological sample based on the above image feature groups and text features.

[0076] Optionally, the image enhancement model includes: an image detection head group, a fusion model, and an image enhancement head group.

[0077] In some optional implementations of certain embodiments, the execution entity performs image enhancement on each electron microscope image in the electron microscope image group using the image enhancement model to generate an enhanced electron microscope image, thereby obtaining an enhanced electron microscope image group. This may include the following steps:

[0078] The first step is to extract shallow image features from the electron microscope image group using the image detection head group described above, thereby obtaining a shallow image feature group.

[0079] The second step involves fusing the aforementioned shallow image feature groups using a fusion model to obtain fused image features.

[0080] The third step is to generate the enhanced electron microscope image group based on the fused image features and the image enhancement head group.

[0081] As an example, see Figure 4The schematic diagram of the sample detection model shown includes: an image enhancement model, a sample image encoding model 406, and a text encoding model 407. The image enhancement model includes: an image detection head group 402, a fusion model 403, and an image enhancement head group 404. The input to the text encoding model 407 is sample detection information 405. The outputs of the sample image encoding model 406 and the text encoding model 407 are sub-sample detection information 408. Specifically, the image detection head group 402 includes: image detection head A1, image detection head A2, image detection head A3, image detection head A4, image detection head A5, image detection head A6, image detection head A7, image detection head A8, and image detection head A9. The input size of image detection head A1 is N1×M1×3. The input size of image detection head A2 is N2×M2×3. The input size of image detection head A3 is N3×M3×3. The input size of image detection head A4 is N4×M4×3. The input dimensions of image detection head A5 are N5×M5×3. The input dimensions of image detection head A6 are N6×M6×3. The input dimensions of image detection head A7 are N7×M7×3. The input dimensions of image detection head A8 are N8×M8×3. The input dimensions of image detection head A9 are N9×M9×3. Where N6 < N1 < N2, N7 < N2 < N3, N8 < N3 < N4, N9 < N4 < N5, M6 < M1 < M2, M7 < M2 < M3, M8 < M3 < M4, and M9 < M4 < M5. Image detection heads A1, A2, A3, A4, A5, A6, A7, A8, and A9 all employ three serially connected convolutional layers. Specifically, taking image detection heads A1, A2, and A6 as examples, when the image size of the electron microscope image is between N1×M1×3 and N2×M2×3, the electron microscope image can be augmented. The augmented electron microscope image has an image size of N6×M6×3, and this augmented image is directly used as the input to image detection head A6. This method can improve the robustness of the model while reducing the number of invalid pixels after augmentation. Taking image detection head A1 as an example, when the image size of the electron microscope image is N1×M1×3, the electron microscope image is first input to image detection head A1, and then the output of image detection head A1 and the output of image detection head A2 are superimposed as the output of image detection head A6. The fusion model 403 includes: fusion layer B1, fusion layer B2, and fusion layer B3. Among them, fusion layer B1 is a superposition layer to superimpose the features of the outputs of image detection heads A6, A7, A8, and A9. The model structures of fusion layers B2 and B3 are symmetrical. Specifically, fusion layer B2 is a downsampling network, and fusion layer B3 is an upsampling network.Image enhancement head group 404 includes: image enhancement head C1, image enhancement head C2, image enhancement head C3, image enhancement head C4, and image enhancement head C5. Image enhancement heads C1, C2, C3, C4, and C5 are all upsampling networks, each corresponding to one of five output sizes. The output size of image enhancement head C1 < the output size of image enhancement head C2 < the output size of image enhancement head C3 < the output size of image enhancement head C4 < the output size of image enhancement head C5. Sample image encoding model 406 includes N image encoding units. Each image encoding unit includes a SelfAttention layer, an Add&Norm layer, a Feed Forward network, and an Add&Norm layer. Text encoding model 407 takes sample detection index information as input and includes N text encoding units, each text encoding unit including a SelfAttention layer, an Add&Norm layer, a SelfAttention layer, an Add&Norm layer, a CrossAttention layer, and an Add&Norm layer. In this model, a control unit is positioned between the first Add&Norm layer and the CrossAttention layer in the text encoding model 407. This control unit controls whether the output of the first Add&Norm layer is directly input into the second SelfAttention layer or the CrossAttention layer in the text encoding model 407. Specifically, when the sample detection index information is plain text, the output of the first Add&Norm layer in the text encoding model 407 is directly input into the second SelfAttention layer. When the sample detection index information is multimodal data, the output of the first Add&Norm layer in the text encoding model 407 is directly input into the CrossAttention layer.

[0082] The aforementioned sample detection model, as an inventive point of this application, firstly achieves multi-scale electron microscopy image enhancement through an image enhancement model; secondly, it enables the identification of biological samples and the automatic detection of the matching degree between the identification results and sample detection index information through a sample image encoding model and a text encoding model. Furthermore, considering the diverse data types of sample detection index information, a control unit is set up to extract text features from both unimodal and multimodal text data types, ensuring the robustness of the recognition process.

[0083] Step 104: Generate sample detection information for the target biological sample group based on the obtained subsample detection information group.

[0084] In some embodiments, the aforementioned executing entity can generate sample detection information for a target biological sample group based on the obtained sub-sample detection information group. In practice, the aforementioned executing entity can generate sample detection information for a target biological sample group based on the obtained sub-sample detection information group through a voting mechanism. For example, the sub-sample detection information group includes: sub-sample detection information A, sub-sample detection information B, sub-sample detection information C, and sub-sample detection information D. Sub-sample detection information A, sub-sample detection information B, and sub-sample detection information C represent consistent detection results; therefore, the sample detection information is any one of sub-sample detection information A, sub-sample detection information B, and sub-sample detection information C. Furthermore, when the maximum value of the voting results, i.e., the number of sub-sample detection information representing the same detection result, is less than a threshold, sample detection information representing sample anomalies is generated.

[0085] The above embodiments of this application have the following beneficial effects: Through the sample state detection method for biological samples according to some embodiments of this application, a unified standard for sample storage and sample detection for biological sample groups is achieved. Specifically, firstly, state detection configuration information for a target biological sample group is determined. This target biological sample group includes at least three parallel target biological samples stored in sample tubes. The target biological samples are biological samples to be subjected to sample state detection. The state detection configuration information includes sample storage configuration information and sample detection configuration information. The sample detection configuration information includes sample detection time points and sample detection index information. Secondly, according to the storage configuration information, the target biological sample group is moved to a sample storage device for sample storage. This achieves unified storage for biological samples. Next, in response to reaching the sample detection time point, for each target biological sample in the target biological sample group, the following processing steps are performed: First, the target biological sample is aspirated into a slide preparation device using a pipette to generate a slide to be tested. This achieves automated quantitative pipetting and automated slide preparation. The second step involves controlling an electron microscope to acquire electron microscopic images of the slide under test at different image scales, resulting in an electron microscopic image set. This yields electron microscopic images from different fields of view. The third step involves generating sub-sample detection information for the target biological sample based on the sample detection index information, the electron microscopic image set, and a pre-trained sample detection model. By combining this with the sample detection model, automated sample detection based on the sample detection index information is achieved. Finally, based on the obtained sub-sample detection information set, sample detection information for the entire target biological sample group is generated. This provides an overall sample evaluation of the entire target biological sample group. This method ensures the consistency of sample detection while significantly improving sample detection efficiency.

[0086] Figure 5This is a schematic block diagram illustrating the structure of a computer device provided in an embodiment of this application. The computer device can be a terminal.

[0087] like Figure 5 As shown, the computer device includes a processor, memory, and network interface connected via a system bus, wherein the memory may include non-volatile storage media and internal memory.

[0088] Non-volatile storage media can store operating systems and computer programs. These computer programs include program instructions that, when executed, cause the processor to perform any sample state detection method applied to biological samples.

[0089] The processor provides computing and control capabilities, supporting the operation of the entire computer device.

[0090] Internal memory provides an environment for the execution of computer programs in non-volatile storage media. When executed by a processor, the computer program enables the processor to perform any sample state detection method applied to biological samples.

[0091] This network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0092] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0093] In one embodiment, the processor is configured to run a computer program stored in a memory to perform the following steps: determining state detection configuration information for a target biological sample group, wherein the target biological sample group comprises at least three parallel target biological samples stored in sample tubes, the target biological samples being biological samples to be subjected to sample state detection, and the state detection configuration information including: sample storage configuration information and sample detection configuration information, the sample detection configuration information including: sample detection time points and sample detection index information; and moving the target biological sample group to a sample storage device according to the storage configuration information, so as to perform state detection on the target biological sample group. The target biological samples are stored. In response to the arrival of the sample detection time point, for each target biological sample in the target biological sample group, the following processing steps are performed: the target biological sample is aspirated into a slide preparation device using a pipette to generate a slide to be tested; an electron microscope is controlled to acquire electron microscope images of the slide to be tested at different image scales to obtain an electron microscope image group; based on the sample detection index information, the electron microscope image group, and a pre-trained sample detection model, sub-sample detection information for the target biological sample is generated; based on the obtained sub-sample detection information group, sample detection information for the target biological sample group is generated.

[0094] This application also provides a computer-readable storage medium storing a computer program, which includes program instructions. When the program instructions are executed, the method implemented can be referred to the various embodiments of this application for the sample state detection method of biological samples.

[0095] The aforementioned computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiments, such as a hard disk or memory of the computer device. Alternatively, the aforementioned computer-readable storage medium may be an external storage device of the computer device, such as a plug-in hard disk, SmartMediaCard (SMC), Secure Digital (SD) card, or Flash Card equipped on the computer device.

[0096] 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 system 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 system. 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 system that includes that element.

[0097] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments. The above descriptions are merely specific implementations of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A sample state detection method for biological samples, comprising: The state detection configuration information for a target biological sample group is determined, wherein the target biological sample group includes at least three target biological samples arranged in parallel and stored in sample tubes, and the target biological samples are biological samples to be subjected to sample state detection. The state detection configuration information includes: sample storage configuration information and sample detection configuration information. The sample detection configuration information includes: sample detection time point and sample detection index information. According to the storage configuration information, the target biological sample group is moved to the sample storage device for sample storage. In response to the arrival of the sample detection time point, for each target biological sample in the target biological sample group, the following processing steps are performed: The target biological sample is aspirated into the slide preparation device using a pipette to generate the slide to be tested; The electron microscope is controlled to acquire electron microscope images of the slide under test at different image scales, resulting in a group of electron microscope images. Based on the sample detection index information, the electron microscope image set, and the pre-trained sample detection model, sub-sample detection information for the target biological sample is generated. The sample detection model includes: an image enhancement model, a sample image encoding model, and a text encoding model. The image enhancement model includes: an image detection head group, a fusion model, and an image enhancement head group. The image detection head group includes: image detection head A1, image detection head A2, image detection head A3, image detection head A4, image detection head A5, image detection head A6, image detection head A7, image detection head A8, and image detection... Probe A9, wherein each image detection head contains 3 serially connected convolutional layers. The input of image detection head A6 is the superposition of the outputs of image detection head A1 and image detection head A2; the input of image detection head A7 is the superposition of the outputs of image detection head A2 and image detection head A3; the input of image detection head A8 is the superposition of the outputs of image detection head A3 and image detection head A4; and the input of image detection head A9 is the superposition of the outputs of image detection head A4 and image detection head A5. Wherein, when the image size is within the range of the input size of image detection head A1 and the image detection... When the image size is between the input sizes of image detection head A2 and A3, the electron microscope image is expanded to the input size of image detection head A6 and directly used as the input of image detection head A6. When the image size is between the input sizes of image detection head A2 and A3, the electron microscope image is expanded to the input size of image detection head A7 and directly used as the input of image detection head A7. When the image size is between the input sizes of image detection head A3 and A4, the electron microscope image is expanded to the input size of image detection head A8 and directly used as the input of image detection head A8. When the input size is between that of image detection head A4 and image detection head A5, the electron microscope image is expanded to the input size of image detection head A9 and directly used as the input of image detection head A9. The fusion model includes: fusion layer B1, fusion layer B2, and fusion layer B3. The image enhancement head group includes: image enhancement head C1, image enhancement head C2, image enhancement head C3, image enhancement head C4, and image enhancement head C5. The sample image encoding model includes: N image encoding units, each including: a SelfAttention layer, an Add&Norm layer, a Feed Forward network, and an Add&Norm layer. The text encoding model includes: N text encoding units, each including: a SelfAttention layer, an Add&Norm layer, a SelfAttention layer, an Add&Norm layer, a CrossAttention layer, and an Add&Norm layer. A control unit is set between the first Add&Norm layer and the CrossAttention layer in the text encoding model.The output of the first Add&Norm layer in the text encoding model is directly input into the second SelfAttention layer or CrossAttention layer. When the sample detection index information is plain text, the output of the first Add&Norm layer in the text encoding model is directly input into the second SelfAttention layer. When the sample detection index information is multimodal data, the output of the first Add&Norm layer in the text encoding model is directly input into the CrossAttention layer. The input size of image detection head A1 is N1×M1×3. The input dimensions of image detection head A2 are N2×M2×3, the input dimensions of image detection head A3 are N3×M3×3, the input dimensions of image detection head A4 are N4×M4×3, the input dimensions of image detection head A5 are N5×M5×3, the input dimensions of image detection head A6 are N6×M6×3, the input dimensions of image detection head A7 are N7×M7×3, the input dimensions of image detection head A8 are N8×M8×3, and the input dimensions of image detection head A9 are N9×M9×3, where N6 < N1 < N2, N7 < N2 < N3, N8 < N3 < N4, N9 < N4 < N5, M6 < M1 < M2, M7 < M2 < M3, M8 < M3 < M4, and M9 < M4 < M5. Based on the obtained subsample detection information group, sample detection information for the target biological sample group is generated.

2. The method according to claim 1, wherein, The sample storage configuration information includes: sample storage temperature; the sample storage device includes: a sample storage box; the sample storage box includes: a first sample storage box and a second sample storage box; the first sample storage box includes: a first sample storage box body and a first sample storage box lid; the first sample storage box body is provided with: K sample tube placement chambers, a humidity sensor and a pressure sensor, K≥1; the first sample storage box is disposed above the second sample storage box; the second sample storage box is used to store a heat-conducting medium; the second sample storage box includes: a temperature sensor and a heating device; and the step of moving the target biological sample group to the sample storage device according to the storage configuration information to store the target biological sample includes: The robotic arm is controlled to move the target biological sample group into the K sample tube placement chamber included in the first sample storage box; In response to successful movement, the lid of the first sample storage box is closed, and the airtightness of the box body is detected by the air pressure sensor included in the first sample storage box body. In response to the stable airtightness of the first sample storage box, the heating device is controlled to heat the heat-conducting medium, and the temperature is monitored by the temperature sensor included in the first sample storage box and the temperature sensor included in the second sample storage box to obtain the first temperature and the second temperature. In response to the first temperature being the sample storage temperature, the heating device is controlled to perform constant temperature heating at the second temperature.

3. The method according to claim 2, wherein, The pipetting device includes a pipetting robotic arm, an electrically controlled pipette, and a positioning camera. The slide preparation device includes a slide stage and a coverslip machine, the coverslip machine being positioned above the slide stage and using a miniature vacuum suction cup to pick up coverslips. The process of aspirating the target biological sample into the slide preparation device using the pipetting device to generate a slide for testing includes: Control the opening of the first sample storage box lid; In response to the opening of the box lid, a first image is captured by a positioning camera, wherein the first image is a top view image of the first sample storage box; Based on the first image, the coordinates of the positioning point are determined, wherein the coordinates of the positioning point represent the position of the sample tube placement chamber corresponding to the target biological sample to be pipetted. Based on the coordinates of the positioning point and the initial position of the pipetting robot, a first robot arm movement path is generated; Control the pipetting robot arm to move along the movement path of the first robot arm to the positioning point coordinates; In response to successful movement, the electronically controlled pipette is used by the pipetting robotic arm to quantitatively aspirate the target raw sample. In response to successful absorption, a second robotic arm movement path is generated based on the coordinates of the positioning point and the coordinates of the slide, wherein the coordinates of the slide represent the coordinates of the center point of the slide on the slide placement stage; Control the pipetting robot arm to move along the movement path of the second robot arm to the coordinates of the glass slide; In response to successful movement, control the electronic pipette to move the aspirated target sample onto the glass slide; In response to successful movement, the cover glass cover machine is controlled to cover the slide with a cover glass to obtain the slide to be tested.

4. The method according to claim 3, wherein, The step of generating sub-sample detection information for the target biological sample based on the sample detection index information, the electron microscopy image set, and the pre-trained sample detection model includes: The image enhancement model is used to enhance each electron microscope image in the electron microscope image group to generate an enhanced electron microscope image, thus obtaining the enhanced electron microscope image group. Image features are extracted from the enhanced electron microscope images in the enhanced electron microscope image group using the sample image coding model to generate image features and obtain an image feature group. The text feature extraction of the sample detection index information is performed using the text encoding model to generate text features; Based on the image feature group and the text features, sub-sample detection information for the target biological sample is generated.

5. The method according to claim 4, wherein, The step of enhancing each electron microscope image in the electron microscope image set using the image enhancement model to generate enhanced electron microscope images, resulting in an enhanced electron microscope image set, includes: The image detection head group is used to extract shallow image features from the electron microscope image group to obtain a shallow image feature group. By using a fusion model, feature fusion is performed on the shallow image feature group to obtain fused image features; The enhanced electron microscope image set is generated based on the fused image features and the image enhancement head group.

6. A computer device, wherein, The computer device includes a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, it implements the steps of the method as described in any one of claims 1-5.

7. A computer-readable storage medium, wherein, The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1-5.

Images