Methods and devices for detecting the activity of biological samples

Abstract

Embodiments of the present disclosure disclose a biological sample activity detection method and device. A specific embodiment of the method comprises: in response to successful reading of biological sample information table characterization and loading of a target slide at an observation position of a fluorescence microscope, collecting a microscope image set; performing fluorescence noise reduction on each microscope image in the microscope image set to generate a microscope noise-reduced image, obtaining a microscope noise-reduced image set; determining a target focal length according to the microscope noise-reduced image set; collecting a biological sample image at the target focal length; and in response to determining that a detection mode is a first detection mode, performing full-amount biological activity detection on the biological sample image to generate biological activity detection information, wherein the detection mode comprises: the first detection mode and a second detection mode, and the first detection mode is a full-amount activity detection mode. This embodiment realizes automatic biological sample activity detection and improves detection efficiency.

Description

Technical Field

[0001] The embodiments disclosed herein relate to the field of biological sample activity detection, specifically to methods and apparatus for biological sample activity detection. Background Technology

[0002] Biological sample activity assay refers to a technique for detecting the number of healthy, live cells within a sample. Currently, the common method for biological sample activity assay is to manually determine the number of live cells. However, this method often suffers from the following technical problems: manual detection of live cell counts is inefficient. Summary of the Invention

[0003] The summary portion of this disclosure is intended to provide a brief overview of the concepts, which will be described in detail in the detailed description portion. This summary portion 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.

[0004] Some embodiments of this disclosure provide methods, apparatus, electronic devices, and computer-readable media for detecting the activity of biological samples to address the technical problems mentioned in the background section above.

[0005] In a first aspect, some embodiments of this disclosure provide a method for detecting the bioactivity of a biological sample. The method includes: reading a biological sample tube label using a label reading device to obtain biological sample information, wherein the biological sample tube label is a label located at the bottom of the biological sample tube, and the biological sample tube contains a sample to be tested for bioactivity; in response to successful reading of the biological sample information and a target slide being mounted at the observation position of a fluorescence microscope, acquiring a set of microscope images, wherein the target slide contains the biological sample from the biological sample tube and a corresponding cell dye, and the microscope images in the set are acquired at different focal lengths and are localized. Images are obtained from the global field of view; fluorescence denoising is performed on each microscopic image in the aforementioned microscopic image set to generate a denoised microscopic image, resulting in a denoised microscopic image set; the target focal length is determined based on the aforementioned denoised microscopic image set; images of biological samples are acquired at the aforementioned target focal length, wherein the aforementioned biological sample images are images from the global field of view; in response to determining the detection mode as the first detection mode, full bioactivity detection is performed on the aforementioned biological sample images to generate bioactivity detection information, wherein the aforementioned detection mode includes: a first detection mode and a second detection mode, the aforementioned first detection mode being a full bioactivity detection mode and the aforementioned second detection mode being a non-full bioactivity detection mode. The determination of the target focal length includes: randomly generating candidate regions, wherein the candidate regions are square regions and the region size of the candidate regions is smaller than the image size of the denoised microscope image; for each denoised microscope image in the denoised microscope image set, performing the following processing steps: cropping the region image corresponding to the candidate region to obtain a candidate image; performing fluorescence source segmentation on the candidate image to generate fluorescence source information; generating a focal length score for the denoised microscope image based on the fluorescence source information; and taking the image focal length corresponding to the denoised microscope image in the denoised microscope image set that meets the focal length score condition as the target focal length.

[0006] Secondly, some embodiments of this disclosure provide a biological sample activity detection device, comprising: a reading unit configured to read a biological sample tube label code via a label code reading device to obtain biological sample information, wherein the biological sample tube label code is a sample tube label code set at the bottom of the biological sample tube, and the biological sample tube contains a sample to be tested for biological activity; a first acquisition unit configured to acquire a set of microscope images in response to the successful reading of the biological sample information and the observation position of the fluorescence microscope being loaded with a target slide, wherein the target slide contains the biological sample and corresponding cell dye in the biological sample tube, and the microscope images in the set of microscope images are images acquired at different focal lengths and under local fields of view; a noise reduction unit configured to perform fluorescence noise reduction on each microscope image in the set of microscope images to generate a denoised microscope image, thereby obtaining a set of denoised microscope images; a determination unit configured to determine a target focal length based on the set of denoised microscope images; the determination unit is further configured to: randomly generate candidate regions, wherein... The candidate region is a square region, and the size of the candidate region is smaller than the image size of the denoised microscope image. For each denoised microscope image in the denoised microscope image set, the following processing steps are performed: cropping the region image corresponding to the candidate region to obtain a candidate image; performing fluorescence source segmentation on the candidate image to generate fluorescence source information; generating a focal length score for the denoised microscope image based on the fluorescence source information; taking the image focal length corresponding to the denoised microscope image that meets the focal length score condition in the denoised microscope image set as the target focal length; a second acquisition unit is configured to acquire biological sample images at the target focal length, wherein the biological sample images are images under the global field of view; a detection unit is configured to perform full-scale bioactivity detection on the biological sample images in response to determining the detection mode as a first detection mode to generate bioactivity detection information, wherein the detection mode includes: a first detection mode and a second detection mode, wherein the first detection mode is a full-scale bioactivity detection mode and the second detection mode is a non-full-scale bioactivity detection mode.

[0007] Thirdly, some embodiments of this disclosure provide an electronic device, including: one or more processors; and a storage device having one or more programs stored thereon, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation of the first aspect above.

[0008] Fourthly, some embodiments of this disclosure provide a computer-readable medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method described in any of the implementations of the first aspect above.

[0009] The above-described embodiments of this disclosure have the following beneficial effects: Through the biological sample activity detection method of some embodiments of this disclosure, firstly, the biological sample tube label code is read using a label code reading device to obtain biological sample information. The biological sample tube label code is located at the bottom of the biological sample tube, which contains the sample to be tested for biological activity. By reading the label code, the relevant description of the biological sample within the biological sample tube (e.g., biological sample type, biological sample source, precautions for biological sample use, etc.) is determined. Secondly, in response to the successful reading of the biological sample information and the presence of a target slide at the observation position of the fluorescence microscope, a set of microscope images is acquired. The target slide contains the biological sample from the biological sample tube and the corresponding cell dye, and the microscope images in the set are images acquired at different focal lengths and under local fields of view. Therefore, a slide needs to be prepared for subsequent microscopic observation before detection. Next, fluorescence denoising is performed on each microscope image in the set to generate a denoised microscope image, resulting in a denoised microscope image set. Here, fluorescent dyes are typically used to stain nucleic acids for subsequent activity detection. However, the shot noise generated by fluorescence affects image clarity, thus impacting subsequent activity detection; therefore, fluorescence noise reduction is necessary. Next, based on the aforementioned set of denoised microscope images, the target focal length is determined. Due to the significant difference between cell size and image size, focal length adjustment is required during activity detection to ensure clear cell observation; thus, the optimal focal length needs to be determined. Finally, images of the biological sample at the target focal length are acquired. These images represent the global field of view. In response to determining the detection mode as the first detection mode, full bioactivity detection is performed on the biological sample images to generate bioactivity detection information. The detection modes include a first detection mode and a second detection mode; the first detection mode is the full bioactivity detection mode, and the second detection mode is the partial bioactivity detection mode. Different detection modes correspond to different cell counts; therefore, different detection modes are designed to reduce counting costs. This method achieves automated biological sample activity detection, significantly improving detection efficiency compared to manual methods. Attached Figure Description

[0010] The above and other features, advantages, and aspects of the embodiments of this disclosure 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 1This is a flowchart of some embodiments of the biological sample activity detection method according to the present disclosure;

[0012] Figure 2 This is a schematic diagram of a microscope's structure.

[0013] Figure 3 These are schematic diagrams of some embodiments of the biological sample activity detection device according to this disclosure;

[0014] Figure 4 This is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. Detailed Implementation

[0015] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure 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 disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0016] 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 in this disclosure can be combined with each other.

[0017] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only 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.

[0018] It should be noted that the terms "a" and "a plurality of" used in this disclosure 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".

[0019] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0020] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0021] Figure 1 This is a flowchart 100 of some embodiments of a biological sample activity detection method according to some embodiments of this disclosure. The biological sample activity detection method includes the following steps:

[0022] Step 101: Read the biological sample tube label code using the label code reading device to obtain the biological sample information.

[0023] In some embodiments, the execution entity (e.g., a computing device) of the biological sample activity detection method can read the biological sample tube label code using a label code reading device to obtain biological sample information. The biological sample tube label code is located at the bottom of the biological sample tube, which contains the sample to be tested for biological activity. The label code can be a QR code. That is, the label code reading device can be a QR code scanning device. The target sample tube contains the biological sample to be tested for biological activity. In practice, firstly, the execution entity can acquire an image containing the biological sample tube label code. Then, it scans the biological sample tube label code using a label code reading device to obtain the biological sample information. In practice, the biological sample information may include: biological sample name, biological sample number, biological sample source, and precautions for using the biological sample.

[0024] 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.

[0025] Step 102: In response to the successful reading of the above biological sample information characterization and the fact that the target slide is mounted at the observation position of the fluorescence microscope, a set of microscope images is acquired.

[0026] In some embodiments, the execution entity may acquire a set of microscope images in response to successful reading of the biological sample information and the presence of a target slide at the observation position of the fluorescence microscope. The target slide contains the biological sample and corresponding cell dye from the biological sample tube, and the microscope images in the set are images acquired at different focal lengths and from local fields of view.

[0027] In practice, the target slide may include: a glass slide, a coverslip, and biological samples and cell dyes placed between the glass slide and the coverslip. Specifically, the target slide can be prepared by the experimenter. The fluorescence microscope can be an inverted fluorescence microscope. The observation position can be the location used to place the target slide. The microscope images in the microscope image set are microscope images acquired at different focal lengths, showing a local field of view. In practice, the aforementioned execution entity can control the focal length of the fluorescence microscope to obtain microscope images at different focal lengths. In practice, as the focal length changes, especially with an increase in focal length, the size of the image corresponding to the entire biological sample within the target slide also increases, and the volume of the resulting image data also increases. Based on this, by determining images at different focal lengths within the same field of view as the first microscope image, the amount of data processing can be reduced without affecting the determination of the optimal observation focal length (target focal length).

[0028] like Figure 2 The example illustrates a structural scene of a microscope, including: a fluorescence microscope 1, a beam assembly 2, a cell dye assembly 3, a coverslip aspirator 4, a coverslip 5, and a glass slide 6. In operation, the cell dye assembly 3 and the coverslip aspirator 4 move horizontally along the beam assembly 2, driven by a chain within the beam assembly 2. A dye pipette is connected below the cell dye assembly 3, and the dye volume is electrically controlled. The coverslip aspirator 4 includes a telescopic arm and a miniature vacuum suction cup. The miniature vacuum suction cup holds the coverslip 5. In practice, the cell dye assembly 3 can be controlled to drop cell dye onto the sample on the glass slide 6, as needed. Then, the coverslip aspirator 4 is controlled to place the coverslip 5 onto the glass slide 6 for image acquisition. Image acquisition can be achieved using a camera built into the microscope.

[0029] Step 103: Perform fluorescence denoising on each microscope image in the above microscope image set to generate a denoised microscope image, thus obtaining a denoised microscope image set.

[0030] In some embodiments, the execution entity may perform fluorescence denoising on each microscope image in the microscope image set to generate a denoised microscope image, thus obtaining a denoised microscope image set. Cellular dyes often exhibit fluorescence properties. For example, different colored nucleic acid dyes and different colored dead cell dyes. Fluorescence can produce shot noise that affects image clarity; therefore, this disclosure performs fluorescence denoising on the microscope images. Specifically, the execution entity may employ a Gaussian filtering algorithm to perform fluorescence denoising on the microscope images to obtain denoised microscope images.

[0031] In practice, the aforementioned implementing entity can perform fluorescence denoising on each microscope image in the aforementioned microscope image set through the following steps to generate a denoised microscope image:

[0032] The first step involves extracting features from the aforementioned microscope images at different scales using an image feature extraction model to generate a first image feature set. This image feature extraction model includes three preprocessing models: a first-scale image feature preprocessing model, a second-scale image feature preprocessing model, and a third-scale image feature preprocessing model. All three models are based on a Transformer architecture. Specifically, each of these models includes a decoder and an encoder. The decoder includes a multi-head attention mechanism layer, a normalization layer, a forward propagation network, and a normalization layer. Alternatively, it may include a mask-based multi-head attention mechanism layer, a normalization layer, a forward propagation network, and a normalization layer. The feature processing scales of the first-scale, second-scale, and third-scale image feature preprocessing models differ. The feature scale of the first-scale model is larger than that of the second-scale model. The feature scale of the second-scale image feature preprocessing model is larger than that of the third-scale image feature preprocessing model. In practice, before inputting the microscope image into the image feature preprocessing model, the aforementioned execution entity can divide the microscope image into uniformly sized image patches, perform a linear transformation on each image patch to reduce the feature dimension, and embed the position information of the image patch as input to the image feature extraction model.

[0033] The second step involves generating a denoised microscope image corresponding to the aforementioned microscope image using the fluorescence denoising model and the first image feature set. The fluorescence denoising model employs a U-shaped network structure. In practice, the U-net model can be used as the base model for this fluorescence denoising model. Specifically, before the U-net model, the fluorescence denoising model includes a 3-layer downsampling network and a feature fusion layer to fuse features from the first image at different scales, serving as input to the U-net model.

[0034] Step 104: Determine the target focal length based on the above-mentioned set of denoised microscope images.

[0035] In some embodiments, the execution entity can determine the target focal length based on the aforementioned set of denoised microscope images. The target focal length can be the optimal observation focal length for detecting the activity of biological samples. In practice, firstly, the execution entity selects the denoised microscope image with the highest clarity from the set of denoised microscope images as the first candidate denoised microscope image. Then, the image acquisition focal length corresponding to the first candidate denoised microscope image is determined as the target focal length.

[0036] In practice, the aforementioned implementing entity can determine the target focal length through the following steps:

[0037] The first step is to randomly generate candidate regions. These candidate regions are square regions, and their size is smaller than the image size of the denoised microscope image.

[0038] The second step involves performing the following processing steps on each of the denoised microscope images in the aforementioned set:

[0039] 1. Crop the region image corresponding to the above candidate regions to obtain candidate images.

[0040] 2. Perform fluorescence source segmentation on the candidate images to generate fluorescence source information. The fluorescence source information refers to the proportion of the fluorescence source area to the candidate image. In practice, the execution entity can use a segmentation model, such as the FastFCN model, to segment the candidate images to generate fluorescence source regions and determine the proportion of the fluorescence source regions to the candidate images, thus obtaining the fluorescence source information.

[0041] 3. Based on the fluorescence source information described above, generate a focal length score for the aforementioned denoised microscope image. In practice, each microscope image corresponds to fluorescence source information. Since fluorescence source information represents the proportion of the fluorescence source area to the candidate image, the aforementioned execution entity can standardize the fluorescence source information to obtain a focal length score. Specifically, due to the fluorescence denoising process, at a suitable focal length, the fluorescence source should be clearly visible, and the proportion of the fluorescence source area to the candidate image should be moderate, avoiding problems such as excessively small fluorescence sources increasing the difficulty of subsequent counting, and excessively large fluorescence sources causing greater observation difficulty. Therefore, the median of the fluorescence source information corresponds to a larger focal length score. The smaller value of the fluorescence source information corresponds to the next higher focal length score. The larger value of the fluorescence source information corresponds to the lowest focal length score.

[0042] The third step is to take the focal length of the denoised image that meets the focal length scoring condition from the above set of denoised microscope images as the target focal length. The focal length scoring condition can be: the denoised image in the set of denoised microscope images has the highest focal length score.

[0043] Step 105: Acquire images of the biological sample at the target focal length.

[0044] In some embodiments, the executing entity can acquire an image of the biological sample at the target focal length. This biological sample image is an image with a global field of view. Specifically, the biological sample image can be a microscope image with a global field of view of the entire target slide at the target focal length. For example, the executing entity can control a fluorescence microscope to acquire an image of the entire target slide at the target focal length as the biological sample image.

[0045] Step 106: In response to determining the detection mode as the first detection mode, perform full bioactivity detection on the above biological sample image to generate bioactivity detection information.

[0046] In some embodiments, the execution entity may, in response to determining the detection mode as a first detection mode, perform full-scale bioactivity detection on the biological sample image to generate bioactivity detection information. The detection modes include a first detection mode and a second detection mode, where the first detection mode is a full-scale bioactivity detection mode and the second detection mode is a non-full-scale bioactivity detection mode. The full-scale bioactivity detection mode refers to detecting the bioactivity of the biological sample within the entire slide. The second detection mode refers to a detection mode that uses a quadrat method to detect the bioactivity of biological samples within a preset area. The quadrat method is a biological survey technique that estimates the population size or density by randomly selecting and fixing the area of ​​quadrats within a survey area and counting the number of biological individuals within each quadrat. The detection accuracy of the second detection mode is lower than that of the first detection mode, but its detection speed is faster. In practice, the detection mode can be selected according to experimental needs.

[0047] In practice, the aforementioned implementing entity can use a target detection model to perform full bioactivity detection on biological sample images to generate bioactivity detection information. Specifically, the target detection model can be the YOLO model. For example, bioactivity detection information can include: the number of surviving cells and the number of dead cells.

[0048] In practice, the aforementioned implementing entity can perform full bioactivity detection on the above-mentioned biological sample images through the following steps to generate bioactivity detection information:

[0049] The first step involves using an image feature extraction model to extract features from the biological sample images at different scales, generating a biological sample image feature set. This method avoids the need to retrain other image feature extraction models, reducing training costs.

[0050] The second step is to perform image feature fusion on the image features of each biological sample in the above biological sample image feature set to generate a fused feature image.

[0051] In practice, the aforementioned execution entity can use a feature pyramid network to fuse the biological sample image features from the aforementioned biological sample image feature set to generate a fused feature image. The feature pyramid network can include a first convolutional layer, a second convolutional layer, and a third convolutional layer. The receptive field of the first convolutional layer is larger than that of the second convolutional layer. The receptive field of the second convolutional layer is larger than that of the third convolutional layer. The first, second, and third convolutional layers are connected sequentially. The second preprocessed image features with the largest feature scale are extracted through the first, second, and third convolutional layers to obtain the image features to be fused. The biological sample image features with the next smaller feature scale are extracted through the second and third convolutional layers to obtain the image features to be fused. The biological sample image features with the smallest feature scale are extracted through the third convolutional layer to obtain the image features to be fused. Finally, the multiple fused image features are superimposed to obtain the fused feature image.

[0052] The third step involves generating bioactivity detection information using the biological target segmentation model and the aforementioned fused feature image. The biological target segmentation model comprises a segmentation model and a labeling model. The segmentation model is used to segment biological targets from the fused feature image, while the labeling model is used to classify the segmented biological targets. In practice, the segmentation model can be the YOLO model, and the labeling model can be a prediction layer connected to the segmentation model, outputting the location of the bounding boxes and a classifier for the cell categories within the bounding boxes. Specifically, the classifier can be a three-class classifier.

[0053] Here, the bioactivity detection information includes: the number of surviving cells, the number of dead cells, a set of surviving cell location information, a set of dead cell location information, and a set of suspicious cell location information. A classifier distinguishes between surviving, dead, and suspicious cells, and corresponding counters count the number of surviving and dead cells. A prediction layer outputting bounding box positions is used to determine the location information of dead, surviving, and suspicious cells. Suspicious cells are cell types whose status cannot be accurately determined as either dead or surviving. Dead cell location information represents the bounding box positions of dead cells. Surviving cell location information represents the bounding box positions of surviving cells. Suspicious cell location information represents the bounding box positions of cells whose status cannot be accurately determined as either dead or surviving.

[0054] Optionally, in response to the above detection mode being a second detection mode, a detection region set is determined with the image center of the above biological sample image as the center.

[0055] In some embodiments, the execution entity may, in response to the detection mode being a second detection mode, determine a set of detection regions centered on the image center of the biological sample image. The detection regions within the set of detection regions have the same size, and the intervals between the detection regions in the set of detection regions are the same.

[0056] Optionally, region images corresponding to the detection regions in the above detection region set are segmented from the above biological sample images to obtain a region image set.

[0057] In some embodiments, the execution entity can segment region images corresponding to the detection regions in the detection region set from the biological sample image to obtain a region image set. A local image within the detection region of the biological sample image can be extracted as a region image.

[0058] Optionally, full bioactivity detection is performed on each region image in the above-mentioned region image set to generate local bioactivity detection information, thus obtaining a local bioactivity detection information set.

[0059] In some embodiments, the aforementioned execution entity can perform full bioactivity detection on each region image in the aforementioned region image set using a target detection model to generate local bioactivity detection information, thereby obtaining a local bioactivity detection information set. The local bioactivity detection information characterizes the bioactivity within the target region.

[0060] Optionally, bioactivity detection information can be generated based on the aforementioned local bioactivity detection information set.

[0061] In some embodiments, the execution entity can sum the aforementioned local bioactivity detection information set to generate bioactivity detection information. For example, the local bioactivity detection information corresponding to region image A could be: {number of surviving cells: 200; number of dead cells: 12}. The local bioactivity detection information corresponding to region image B could be: {number of surviving cells: 200; number of dead cells: 10}. Therefore, the obtained bioactivity detection information could be: {number of surviving cells: 400; number of dead cells: 22}.

[0062] In some optional implementations of some embodiments, the above method further includes:

[0063] The first step is to generate a first labeled image based on the above set of surviving cell location information.

[0064] In practice, the aforementioned executing entity can mark the location of surviving cells in red to obtain the first marked image.

[0065] The second step is to generate a second labeled image based on the above set of dead cell location information.

[0066] In practice, the aforementioned executing entity can mark the location of the dead cells in green to obtain the second marked image.

[0067] The third step is to generate a third labeled image based on the above set of suspicious cell location information.

[0068] The first, second, and third labeled images are located on different layers, and the cells in the first, second, and third labeled images are labeled with different colors.

[0069] In practice, the aforementioned executing entity can mark the location of the suspicious cell as gray to obtain the aforementioned third marked image.

[0070] In some optional implementations of some embodiments, the above method further includes:

[0071] Display the merged marker image based on the layer selected by the user.

[0072] The selected layer includes at least one layer, and the layer in the at least one layer corresponds to the marker image in the first marker image, the second marker image, and the third marker image. For example, if the user selects the first marker image in the first layer and the second marker image in the second layer, the execution entity can overlay the first marker image and the second marker image to obtain the fused marker image.

[0073] Further reference Figure 3 As an implementation of the methods shown in the above figures, this disclosure provides some embodiments of a biological sample activity detection device, which are similar to... Figure 1 Corresponding to the method embodiments shown, this biological sample activity detection device can be specifically applied to various electronic devices.

[0074] like Figure 3As shown, a biological sample activity detection device 300 in some embodiments includes: a reading unit 301, a first acquisition unit 302, a noise reduction unit 303, a determination unit 304, a second acquisition unit 305, and a detection unit 306. The reading unit 301 is configured to read the biological sample tube label code using a label code reading device to obtain biological sample information. The biological sample tube label code is a label code located at the bottom of the biological sample tube, which contains a sample to be tested for biological activity. The first acquisition unit 302 is configured to acquire a set of microscope images in response to successful reading of the biological sample information and a target slide being mounted at the observation position of the fluorescence microscope. The target slide contains the biological sample from the biological sample tube and the corresponding cell dye. The microscope images in the set are images acquired at different focal lengths and under local fields of view. The noise reduction unit 303 is configured to process the microscope images... Each microscope image in the set is subjected to fluorescence denoising to generate a denoised microscope image, resulting in a denoised microscope image set; a determination unit 304 is configured to determine a target focal length based on the denoised microscope image set; a second acquisition unit 305 is configured to acquire biological sample images at the target focal length, wherein the biological sample images are images under a global field of view; a detection unit 306 is configured to perform full-scale bioactivity detection on the biological sample images in response to determining the detection mode as a first detection mode, to generate bioactivity detection information, wherein the detection mode includes: a first detection mode and a second detection mode, the first detection mode being a full-scale bioactivity detection mode and the second detection mode being a non-full-scale bioactivity detection mode.

[0075] It is understandable that the units described in the biological sample activity detection device 300 are similar to those in the reference. Figure 1 The steps in the described method correspond to each other. Therefore, the operations, features, and beneficial effects described above for the method also apply to the biological sample activity detection device 300 and the units contained therein, and will not be repeated here.

[0076] The following is for reference. Figure 4 It illustrates a schematic diagram of the structure of an electronic device (such as a computing device) suitable for implementing some embodiments of the present disclosure. Figure 4 The electronic device shown is merely an example and should not be construed as limiting the functionality or scope of the embodiments of this disclosure. Figure 4As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The memory may include non-volatile storage media and internal memory. The non-volatile storage media may store an operating system and a computer program. The computer program includes program instructions that, when executed, cause the processor to perform any biological sample activity detection method. The processor provides computational and control capabilities to support the operation of the entire computer device. The internal memory provides an environment for the execution of the computer program in the non-volatile storage media; when executed by the processor, this program causes the processor to perform any biological sample activity detection method. The network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present disclosure and does not constitute a limitation on the computer device to which the present disclosure is applied. A specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0077] 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.

[0078] In one embodiment, the processor is configured to run a computer program stored in a memory to perform the following steps: reading the biological sample tube label code via a label code reading device to obtain biological sample information, wherein the biological sample tube label code is a sample tube label code located at the bottom of the biological sample tube, and the biological sample tube contains a sample to be tested for bioactivity; in response to the successful reading of the biological sample information and the presence of a target slide at the observation position of the fluorescence microscope, acquiring a set of microscope images, wherein the target slide contains the biological sample from the biological sample tube and the corresponding cell dye, and the microscope images in the set of microscope images are acquired at different focal lengths. Images from a local field of view; fluorescence denoising is performed on each microscopic image in the aforementioned microscopic image set to generate a denoised microscopic image, resulting in a denoised microscopic image set; the target focal length is determined based on the aforementioned denoised microscopic image set; images of biological samples are acquired at the aforementioned target focal length, wherein the aforementioned biological sample images are images from a global field of view; in response to determining the detection mode as a first detection mode, full bioactivity detection is performed on the aforementioned biological sample images to generate bioactivity detection information, wherein the aforementioned detection mode includes: a first detection mode and a second detection mode, the aforementioned first detection mode being a full bioactivity detection mode and the aforementioned second detection mode being a non-full bioactivity detection mode.

[0079] This disclosure also provides a computer-readable storage medium storing a computer program, the computer program including program instructions, and the method implemented when the program instructions are executed can be referred to various embodiments of the biological sample activity detection method of this disclosure.

[0080] The aforementioned computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiments, such as the 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, SmartMedia Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the computer device.

[0081] 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.

[0082] The above description is merely a selection of preferred embodiments of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. A method for detecting the activity of biological samples, characterized in that, The method comprises the following steps: reading a biological sample tube label code through a label code reading device to obtain biological sample information, wherein the biological sample tube label code is a sample tube label code arranged at the bottom of a biological sample tube, and the biological sample tube contains a sample to be detected for biological activity; in response to the biological sample information indicating that the reading is successful and that the observation position of a fluorescence microscope is loaded with a target slide, collecting a microscope image set, wherein the target slide contains the biological sample in the biological sample tube and corresponding cell dyes, and the microscope images in the microscope image set are images collected at different focal lengths and in a local field of view; performing fluorescence noise reduction on each microscope image in the microscope image set to generate a microscope noise-reduced image, thereby obtaining a microscope noise-reduced image set; determining a target focal length based on the microscope noise-reduced image set; collecting a biological sample image at the target focal length, wherein the biological sample image is an image in a global field of view; in response to determining that the detection mode is a first detection mode, performing full-quantity biological activity detection on the biological sample image to generate biological activity detection information, wherein the detection mode includes a first detection mode and a second detection mode, the first detection mode is a full-quantity activity detection mode, and the second detection mode is a non-full-quantity activity detection mode, different detection modes correspond to different cell counting quantities; wherein the determining of the target focal length based on the microscope noise-reduced image set comprises: randomly generating a candidate region, wherein the candidate region is a square region, and the area size of the candidate region is smaller than the image size of the microscope noise-reduced image; for each microscope noise-reduced image in the microscope noise-reduced image set, the following processing steps are performed: cropping a region image corresponding to the candidate region to obtain a candidate image; performing fluorescence source segmentation on the candidate image to generate fluorescence source information, wherein the fluorescence source information is the proportion of the fluorescence source area in the candidate image; generating a focal length score for the microscope noise-reduced image based on the fluorescence source information; taking the image focal length corresponding to the microscope noise-reduced image in the microscope noise-reduced image set that meets the focal length score condition as the target focal length.

2. The method of claim 1, wherein, The fluorescence noise reduction on each microscope image in the microscope image set to generate a microscope noise-reduced image comprises: extracting features of different scales from the microscope image through an image feature extraction model to generate a first image feature set, wherein the image feature extraction model comprises a first scale image feature preprocessing model, a second scale image feature preprocessing model, and a third scale image feature preprocessing model, and the first scale image feature preprocessing model, the second scale image feature preprocessing model, and the third scale image feature preprocessing model are all based on a Transformer structure; generating a microscope noise-reduced image corresponding to the microscope image through a fluorescence noise reduction model and the first image feature set, wherein the fluorescence noise reduction model adopts a U-shaped network structure.

3. The method of claim 2, wherein, The full-quantity biological activity detection on the biological sample image to generate biological activity detection information comprises: The image feature extraction model is used for feature extraction of the biological sample image at different scales to generate a biological sample image feature set; Each biological sample image feature in the biological sample image feature set is subjected to image feature fusion to generate a fused feature image; The biological target segmentation model and the fused feature image are used to generate biological activity detection information.

4. The method of claim 3, wherein, The method further comprises: in response to the detection mode being the second detection mode, determining a set of detection regions centered on the image center of the biological sample image, wherein the area size of the detection regions in the set of detection regions is the same, and the area interval between the detection regions in the set of detection regions is the same; segmenting the regions in the set of detection regions from the biological sample image to obtain a set of region images; performing full-quantity biological activity detection on each region image in the set of region images to generate local biological activity detection information and obtain a set of local biological activity detection information; generating biological activity detection information based on the set of local biological activity detection information.

5. A biological sample activity detection device, characterized by, comprises: a reading unit configured to read a biological sample tube label code by a label code reading device to obtain biological sample information, wherein the biological sample tube label code is a sample tube label code arranged at the bottom of a biological sample tube, and the biological sample tube contains a sample to be subjected to biological activity detection; a first acquisition unit configured to, in response to the biological sample information being successfully read and the observation position of a fluorescence microscope being loaded with a target slide, acquire a set of microscope images, wherein the target slide contains biological samples in the biological sample tube and corresponding cell dyes, and the microscope images in the set of microscope images are images under a local field of view acquired at different focal lengths; a noise reduction unit configured to perform fluorescence noise reduction on each microscope image in the set of microscope images to generate a set of microscope noise reduction images; a determination unit configured to determine a target focal length based on the set of microscope noise reduction images; the determination unit is further configured to: randomly generate a candidate region, wherein the candidate region is a square region, and the area size of the candidate region is smaller than the image size of the microscope noise reduction image; for each microscope noise reduction image in the set of microscope noise reduction images, the following processing steps are performed: cropping a region image corresponding to the candidate region to obtain a candidate image; performing fluorescence source segmentation on the candidate image to generate fluorescence source information, wherein the fluorescence source information is the proportion of the fluorescence source area in the candidate image; generating a focal length score for the microscope noise reduction image based on the fluorescence source information; taking the image focal length corresponding to the microscope noise reduction image in the set of microscope noise reduction images that meets the focal length score condition as the target focal length; a second acquisition unit configured to acquire a biological sample image at the target focal length, wherein the biological sample image is an image under a global field of view; The detection unit is configured to, in response to determining that the detection mode is the first detection mode, perform full-amount biological activity detection on the biological sample image to generate biological activity detection information, wherein the detection mode includes: a first detection mode and a second detection mode, the first detection mode is a full-amount activity detection mode, and the second detection mode is a non-full-amount activity detection mode. Different detection modes correspond to different cell counting quantities.

6. An electronic device, comprising: The method comprises: one or more processors; a storage device having stored thereon one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 4.

7. A computer readable medium characterized by a computer program stored thereon, wherein the computer program is executed by a processor to implement the method according to any one of claims 1 to 4.

Images