Intelligent event imaging flow cytometer

Abstract

The present disclosure relates to an intelligent event imaging flow cytometer, comprising a microfluidic system, an optical system and a hardware processing system. The microfluidic system is used for continuously importing a cell sample comprising at least two kinds of cells, and exporting each kind of cell according to the classification result fed back by the hardware processing system. The optical system is used for jointly irradiating the cell sample by a bright field light source and a fluorescent light source, and collecting multiple double-channel coupling information simultaneously comprising bright field cells and fluorescent cells. The hardware processing system is used for obtaining the classification result of the target cell according to multiple target coupling information corresponding to the target cell determined in the double-channel coupling information, and feeding back the classification result to the microfluidic system. In the present disclosure, the cell classification in the cell sample can be realized through the information interaction transmission of the three systems, which simplifies the overall structure of the cytometer. At the same time, the cell classification result is improved in accuracy by jointly judging the cell classification through the cell characteristics collected under the bright field and fluorescent light sources.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of medical equipment, and in particular to an intelligent event imaging flow cytometer. BACKGROUND

[0002] Flow cytometer is an important life science instrument, which can quickly and quantitatively analyze the physical and chemical properties of cell population, and can accurately sort specific cells according to the differences in these properties. The most advanced imaging flow cytometer can achieve multi-fluorescence imaging, high throughput, real-time classification and sorting, but its small field imaging can only detect the transient characteristics of cells, and the system complexity is too high to be popularized. SUMMARY

[0003] Therefore, the present disclosure proposes an intelligent event imaging flow cytometer, which aims to reduce the complexity of the whole system in the cytometer and ensure the accuracy of the specific cells sorted by the cytometer.

[0004] According to a first aspect of the present disclosure, an intelligent event imaging flow cytometer is provided, which comprises:

[0005] A microfluidic system is configured to continuously introduce a cell sample comprising at least two types of cells, and sort at least one type of cells in the cell sample according to a classification result fed back by a hardware processing system;

[0006] An optical system is configured to irradiate the cell sample by a bright field light source and a fluorescence light source, and continuously collect a plurality of double-channel coupling information comprising bright field cells and fluorescence cells at the same time;

[0007] A hardware processing system is configured to determine a plurality of target coupling information corresponding to each target cell in the cell sample in the double-channel coupling information, determine a classification result of the target cell according to the target coupling information, and feed back the classification result to the microfluidic system.

[0008] In a possible implementation, the microfluidic system comprises:

[0009] A microfluidic pump is configured to continuously introduce the cell sample into a microfluidic chip;

[0010] A microfluidic chip is configured to manipulate and transport the introduced cell sample, and the spatial positions of the cell sample in the microfluidic chip do not overlap;

[0011] A sorting mechanism is configured to sort and export the cell sample in the microfluidic chip according to the classification result fed back by the hardware processing system.

[0012] In a possible implementation, the manner in which the microfluidic chip manipulates the cell sample includes at least one of bunching, stretching, rolling, rotating, and dynamic acceleration.

[0013] In a possible implementation, the optical system includes:

[0014] A dual light source illumination system for simultaneously illuminating the microfluidic chip by a bright field light source and a fluorescent light source;

[0015] An inverted microscope for magnifying the cell sample in the microfluidic chip;

[0016] A spatial filter for filtering different parts of the magnified imaging result of the inverted microscope by the bright field light source and the fluorescent light source respectively to obtain a dual-channel imaging result including bright field cells and fluorescent cells, wherein the bright field light source filtering retains the imaging result in the wavelength range of the bright field light source, and the fluorescent light source filtering retains the imaging result in the wavelength range of the fluorescent light source;

[0017] An event camera for acquiring the dual-channel imaging result to obtain dual-channel coupling information.

[0018] In a possible implementation, the dual light source illumination system includes:

[0019] A halogen lamp light source including a halogen lamp and a bright field filter for filtering the light source generated by the halogen lamp through the bright field filter to obtain the bright field light source therein;

[0020] A mercury lamp light source including a mercury lamp and a fluorescent filter for filtering the light source generated by the mercury lamp through the fluorescent filter to obtain the fluorescent light source therein.

[0021] In a possible implementation, the spatial filter includes a spliced bright field filter and a fluorescent filter for filtering half of the magnified imaging result of the inverted microscope by the bright field light source to obtain a bright field imaging result containing bright field cells, and filtering the other half by the fluorescent light source to obtain a fluorescent imaging result containing fluorescent cells.

[0022] In a possible implementation, the hardware processing system includes:

[0023] A central processing unit for receiving a plurality of dual-channel coupling information acquired by the optical system and scheduling a neuromorphic chip to process the dual-channel coupling information;

[0024] The neuromorphic chip is configured to perform sparse space-time decoding, cell detection and identification, cell tracking and real-time classification on the received multiple double-channel coupled information, determine multiple target coupled information corresponding to each target cell in the cell sample according to the cell tracking result, and determine the classification result of the target cell according to the real-time classification result of the multiple target coupled information.

[0025] In a possible implementation, the neuromorphic chip is further configured to perform multi-modal feature extraction on the received double-channel coupled information, and perform image reconstruction according to the result of multi-modal feature extraction to obtain a two-dimensional reconstruction result.

[0026] In a possible implementation, the neuromorphic chip is further configured to perform three-dimensional reconstruction according to the two-dimensional reconstruction result of the multiple target coupled information corresponding to the target cell, to obtain a three-dimensional image of the target cell.

[0027] In a possible implementation, the hardware processing system further includes a graphics processor.

[0028] The central processing unit is configured to schedule the neuromorphic chip and / or the graphics processor to process the double-channel coupled information.

[0029] In the embodiments of the present disclosure, the cytometer includes a microfluidic system, an optical system and a hardware processing system. The microfluidic system is configured to continuously introduce a cell sample including at least two types of cells, and classify and export each type of cell according to the classification result fed back by the hardware processing system. The optical system is configured to irradiate the cell sample by a bright field light source and a fluorescent light source together, and collect multiple double-channel coupled information including both bright field cells and fluorescent cells. The hardware processing system is configured to obtain the classification result of a target cell according to multiple target coupled information corresponding to the target cell determined from the double-channel coupled information, and feed back the classification result to the microfluidic system. The cytometer in the present disclosure can realize cell classification in the cell sample through information interaction and transmission of the three systems, and simplifies the overall structure of the cytometer. Meanwhile, the cell classification result is improved in accuracy by jointly judging the cell type according to the cell characteristics collected under the bright field and fluorescent light sources.

[0030] Other features and aspects of the present disclosure will become apparent from the following detailed description of the example embodiments, taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0031] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate example embodiments, features, and aspects of the present disclosure and serve to explain the principles of the present disclosure.

[0032] Figure 1 FIG. 1 shows a schematic diagram of an intelligent event imaging flow cytometer according to an embodiment of the present disclosure;

[0033] Figure 2 A schematic diagram of a microfluidic system according to an embodiment of the present disclosure is shown.

[0034] Figure 3 A schematic diagram of an optical system according to an embodiment of the present disclosure is shown.

[0035] Figure 4 A schematic diagram of a hardware processing system according to an embodiment of the present disclosure is shown.

[0036] Figure 5 A schematic diagram of a hardware processing system performing cell classification according to an embodiment of the present disclosure is shown.

[0037] Figure 6 A schematic diagram of an overall workflow of a cytometer according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0038] Various exemplary embodiments, features, and aspects of the present disclosure will be described herein below with reference to the accompanying drawings. The same reference numbers in different drawings indicate the same or similar elements. Although various aspects of embodiments are illustrated in the drawings, the drawings are not necessarily drawn to scale unless specifically noted.

[0039] The word "exemplary" is used herein to mean "serving as an example, instance, or illustration." Any implementation described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other implementations.

[0040] In addition, for the purpose of convenience and brevity, detailed descriptions of well-known functions, procedures, components, and circuits will not be described in detail herein. It should be noted that the present disclosure can be implemented in various ways, and that the present disclosure should not be construed as being limited to specific embodiments set forth herein. Rather, embodiments of the present disclosure can be implemented in various combinations and sub-combinations of the embodiments described in the detailed description above and illustrated in the various drawings. It will be apparent that embodiments, as described herein, can be implemented in various ways, and that the

[0041] Figure 1 A schematic diagram of an intelligent event imaging flow cytometer according to an embodiment of the present disclosure is shown. As shown in FIG. 1, the intelligent event imaging flow cytometer 100 includes a sample preparation module 110, a flow cytometer module 120, a data acquisition module 130, a data processing module 140, and a data analysis module 150. Figure 1As shown, the intelligent event imaging flow cytometer of this disclosure includes a microfluidic system, an optical system, and a hardware processing system. Both the optical system and the microfluidic system are connected to the hardware processing system via data communication. The microfluidic system continuously imports cell samples containing at least two types of cells and classifies and exports at least one type of cell from the cell sample based on the classification results fed back by the hardware processing system. The optical system illuminates the cell sample using both a bright-field light source and a fluorescent light source, and continuously acquires multiple dual-channel coupling information samples simultaneously containing bright-field and fluorescent cells. The hardware processing system determines multiple target coupling information samples corresponding to each target cell in the cell sample from the dual-channel coupling information, determines the classification result of the target cell based on the target coupling information, and feeds back the classification result to the microfluidic system.

[0042] Figure 2 A schematic diagram of a microfluidic system according to an embodiment of the present disclosure is shown. Figure 2 As shown, the microfluidic system in the cytometer includes a microfluidic pump, a microfluidic chip, and a sorting mechanism. The microfluidic pump continuously introduces cell samples into the microfluidic chip. The cell sample can be a suspension containing at least two types of cells. The microfluidic pump injects the cell sample into the microfluidic chip through the output port of the injection device. The microfluidic chip is a biochip used to manipulate and transport the introduced cell sample. The cell samples do not overlap in spatial position within the microfluidic chip; that is, no two cells in the cell sample overlap in the spatial position of the microfluidic chip plane. The sorting mechanism communicates continuously with the hardware processing system, receiving the classification structure corresponding to each cell in the cell sample from the hardware processing system, and classifying and exporting the cell sample from the microfluidic chip based on the classification results from the hardware processing system. During the operation of the cytometer, the hardware processing system feeds back the classification results of each cell in the cell sample to the sorting mechanism in real time, and the sorting mechanism classifies and exports the cells in the cell sample from the cytometer according to the corresponding classification results. Meanwhile, as cell samples are exported from the microfluidic chip, the microfluidic pump continuously introduces unclassified cell samples into the microfluidic chip, ensuring the continuous flow of cell samples in the microfluidic chip.

[0043] Optionally, the microfluidic chip can manipulate internal cell samples, and the manipulation methods can include at least one of bundle queuing, stretching deformation, rolling rotation, and dynamic acceleration. This manipulation process can be used to increase the diversity of each cell's morphology, velocity, and position, so as to further acquire images of each cell in the cell sample with different morphologies, velocities, angles, and positions through an optical system, enriching cell features and improving the accuracy of classification results.

[0044] Furthermore, the sorting mechanism can classify and output each cell in the cell sample, or it can classify and output at least one cell of a specific category in the cell sample.

[0045] Figure 3 A schematic diagram of an optical system according to an embodiment of the present disclosure is shown. Figure 3 As shown, the optical system may include a dual-light source illumination system, an inverted microscope, a spatial filter, and an event camera. The dual-light source illumination system is used to simultaneously illuminate the microfluidic chip using both a bright-field light source and a fluorescence light source. In this embodiment, the microfluidic chip can be a transparent chip, allowing each light source to penetrate and illuminate the cell sample within. The objective lens of the inverted microscope is positioned directly over the microfluidic chip to magnify the cell sample, obtaining a clear image including all cells in the microfluidic chip. The spatial filter is used to filter the imaging results from the microscope using both bright-field and fluorescence light sources. Specifically, different portions of the magnified image from the inverted microscope can be filtered using bright-field and fluorescence light sources respectively, resulting in a dual-channel imaging result including both bright-field and fluorescence cells. Bright-field filtering preserves the imaging results within the bright-field wavelength range, while fluorescence filtering preserves the imaging results within the fluorescence wavelength range. The event camera is used to acquire the dual-channel imaging results including both bright-field and fluorescence cells, obtaining dual-channel coupled information.

[0046] Optionally, the dual-source lighting system of this disclosure embodiment may have a bright-field light source and a fluorescent light source. The bright-field light source can be obtained using a halogen lamp light source including a halogen lamp and a bright-field filter, and the fluorescent light source can be obtained using a mercury lamp light source including a mercury lamp and a fluorescent filter. The halogen lamp light source generates a light source through the halogen lamp, and the light source generated by the halogen lamp is filtered by the bright-field filter to obtain the bright-field light source. The mercury lamp light source generates a light source through the mercury lamp, and the light source generated by the mercury lamp is filtered by the fluorescent filter to obtain the fluorescent light source.

[0047] Optionally, the spatial filter in this embodiment includes a spliced ​​bright-field filter and a fluorescence filter; that is, the spatial filter can be a filter obtained by splicing a bright-field filter and a fluorescence filter. The filtering areas of the bright-field filter and the fluorescence filter can be of the same size. After the imaging result obtained by magnification from the inverted microscope passes through the spatial filter, half is filtered by the bright-field filter to obtain a bright-field imaging result containing bright-field cells, and the other half is filtered by the fluorescence filter to obtain a fluorescence imaging result containing fluorescent cells.

[0048] Further, the event camera is a dynamic vision sensor, which is an imaging sensor capable of responding to local brightness changes and outputting change information in each case of local brightness change. That is, when each movement change of the cell sample in the microfluidic chip occurs, the event camera perceives the change of brightness in the field of view and collects the change information to obtain the corresponding dual-channel coupling information, which is used to record the spatial position change of the cell sample. In addition, the dual-channel coupling information can also record the live or dead state of each cell in the cell sample in the field of view in the fluorescence imaging result. Alternatively, the dual-channel coupling information can be in the form of an image. The event camera can be communicatively connected with the hardware processing system. After collecting the dual-channel coupling information each time, the event camera transmits the collected dual-channel coupling information to the hardware processing system in real time for analysis and processing, so as to obtain the classification result of each cell in the microfluidic chip.

[0049] Figure 4 A schematic diagram of a hardware processing system according to an embodiment of the present disclosure is shown. As shown in the figure, the hardware processing system of the present embodiment can include a central processor and a neuromorphic chip. The central processor receives a plurality of dual-channel coupling information collected by the optical system and schedules the neuromorphic chip to process the dual-channel coupling information. The neuromorphic chip performs sparse space-time decoding, cell detection and recognition, cell tracking and real-time classification on the received plurality of dual-channel coupling information, and determines a plurality of target coupling information corresponding to each target cell in the cell sample according to the cell tracking result, and determines the classification result of the target cell according to the real-time classification result of the plurality of target coupling information. Figure 4

[0050] Alternatively, during the operation of the hardware processing system, the central processor is not deployed with an algorithm for cell classification in order to improve the processing efficiency, and is only used to schedule the neuromorphic chip for cell classification. The central processor can receive the dual-channel coupling information collected by the optical system through a data node, and can also distribute the dual-channel coupling information to the scheduled neuromorphic chip through the data node. The neuromorphic chip can be deployed with an algorithm for cell classification, and the neuromorphic chip executes the corresponding algorithm for cell classification after receiving the dual-channel coupling information. The neuromorphic chip can deploy an artificial neural network and / or a spiking neural network, and can schedule any type of neural network to execute the algorithm related to cell classification during the process of cell classification.

[0051] ​Due to the derivable continuous firing mechanism adopted by the traditional artificial neural network algorithm, the information transmitted by the event camera is extracted as non-sparse features, thereby reducing the acceleration and power consumption performance on the neuromorphic chip. In order to fully utilize the sparse characteristics of the double-channel coupled information, the dynamic event-driven characteristics of the spiking neural network are utilized, and the neurons in the neuromorphic chip communicate in the form of binary pulse of 0 / 1. At the same time, the processing of the event camera data is matched, and the acceleration and power consumption performance of the neuromorphic chip after deployment is further improved. Based on the heterogeneous neural network of artificial neural network and spiking neural network, the neuromorphic chip can realize real-time acceleration calculation of double-channel coupled information while obtaining higher algorithm performance.

[0052] Further, in addition to the neuromorphic chip, the hardware processing system can also include a graphics processor, which is also used for cell classification processing. In the case of simultaneously including a graphics processor and a neuromorphic chip in the hardware processing system, the central processing unit can schedule the neuromorphic chip and / or the graphics processor to process the double-channel coupled information. That is, the neuromorphic chip and the graphics processor can respectively deploy part of the algorithm, and the central processing unit can first schedule the neuromorphic chip to process part of the algorithm after receiving the double-channel coupled information collected by the optical system through the data node, and then send the processing result of the neuromorphic chip to the graphics processor for subsequent processing to obtain the final cell classification result.

[0053] In a possible implementation, the format of the double-channel coupled information received by the neuromorphic chip and / or the graphics processor can be the sparse space-time encoded information generated by the event camera. After receiving the information, the double-channel coupled information in the sparse space-time encoded information needs to be obtained by sparse space-time decoding, and then further cell detection is performed to identify each cell included therein, and the same cell is identified in different double-channel coupled information through cell tracking, and the classification algorithm is performed based on the bright field characteristics or fluorescence characteristics of the cell to obtain the corresponding classification result. Optionally, after the cell recognition and detection, the neuromorphic chip and / or the graphics processor can also perform counting to determine the number of cells existing in the field of view when the corresponding double-channel coupled information is collected by the optical system.

[0054] Optionally, the neuromorphic chip and / or the graphics processor can determine a cell as a target cell when the cell appears for the first time in the dual-channel coupled information during the cell classification process, and continuously track the position of the target cell in the dual-channel coupled information acquired subsequently until the target cell no longer exists in the sequentially acquired dual-channel coupled information, and determine the dual-channel coupled information including the target cell previously as target coupled information. The neuromorphic chip and / or the graphics processor can perform class voting according to the real-time classification results of the multiple target coupled information corresponding to each target cell to obtain the final classification result. Since each cell is in a motion state in the microfluidic chip and can appear in the bright field region or the fluorescent region, the real-time classification result of each target coupled information includes the result obtained based on the bright field cell classification and the classification result obtained based on the fluorescent cell classification. The classification results obtained based on the cell features in two different illumination conditions jointly determine the classification result of the cell, increase the richness of the sample, and can improve the accuracy of the classification result and solve the problem that the wide field of view cannot be sorted.

[0055] Figure 5 A schematic diagram of cell classification performed by a hardware processing system according to an embodiment of the present disclosure is shown. As shown in Figure 5 The format of the dual-channel coupled information received by the neuromorphic chip and / or the graphics processor can be sparse spatiotemporal encoding information (i.e., sparse data) generated by an event camera. After receiving the information, the dual-channel coupled information (event frame) in the information needs to be obtained by sparse spatiotemporal decoding of the sparse data, and then further cell detection is performed to identify each cell included therein, and the same cell is identified in different dual-channel coupled information by cell tracking, and classification is performed based on the bright field features or fluorescent features of the cell by a preset classification algorithm to obtain the corresponding classification result. Before performing class voting to determine the final classification result, the neuromorphic chip / graphics processor can also perform multi-modal feature extraction on the received dual-channel coupled information, and perform image reconstruction according to the multi-modal feature extraction result to obtain a two-dimensional reconstruction result, and further perform three-dimensional reconstruction according to the two-dimensional reconstruction results of the multiple target coupled information corresponding to the target cell to obtain a three-dimensional image of the target cell. The three-dimensional image can be displayed by a display device connected to the hardware processing system to visually display the cell morphology in the cell sample to the user.

[0056] Figure 6 A schematic diagram of the overall workflow of a cytometer according to an embodiment of the present disclosure is shown. As shown in Figure 6As shown, after the cell sample is input, the flow cytometer of the embodiment of the present disclosure guides the cell sample into the microfluidic chip through the microfluidic pump. The optical system performs double lighting of bright field and fluorescence on the microfluidic chip, and the microscope in the optical system amplifies the cell sample in the microfluidic chip. The event camera collects the imaging results including two regions of bright field and fluorescence after filtering by the spatial filter, and obtains the double-channel coupling information. After obtaining the double-channel coupling information, the event camera transmits information in the form of sparse event coding, and sends it to the neuromorphic chip in the hardware processing system for cell classification. Among them, the neuromorphic chip is deployed with multiple algorithms such as decoding, detection, tracking, feature extraction, image reconstruction and classification, which are used for cell classification based on the double-channel coupling information sent by the event camera, and the classification results are fed back to the sorting mechanism in the microfluidic system. The sorting mechanism guides the classified cells in the microfluidic chip out according to the classification results.

[0057] Based on the above technical features, the flow cytometer in the embodiment of the present disclosure can realize cell classification in the cell sample through the information interaction and transmission of the microfluidic system, the optical system and the hardware processing system, which simplifies the overall structure of the flow cytometer. At the same time, the cell classification result accuracy is improved by jointly judging the cell category based on the cell characteristics collected under the two light sources of bright field and fluorescence. Further, the hardware for executing the classification algorithm in the hardware processing system of the embodiment of the present disclosure is the neuromorphic chip, which obtains higher algorithm performance, and realizes real-time acceleration calculation of the algorithm through the cooperation of artificial neural network and pulse neural network. The above has described various embodiments of the present disclosure, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes are obvious to those skilled in the art without departing from the scope and spirit of the described embodiments. The choice of terms used herein is intended to best explain the principles, practical applications or technical improvements in the market of the embodiments, or to enable other ordinary skilled persons in the art to understand the embodiments disclosed herein.

Claims

1. A smart event imaging flow cytometer, characterized in that, include: A microfluidic system for continuously introducing cell samples containing at least two types of cells, and classifying and exporting at least one type of cell from the cell sample based on the classification results fed back by the hardware processing system; An optical system is used to illuminate the cell sample with both a bright-field light source and a fluorescent light source, and to continuously acquire multiple dual-channel coupled information that simultaneously includes bright-field cells and fluorescent cells; A hardware processing system is used to determine multiple target coupling information corresponding to each target cell in the cell sample from the dual-channel coupling information, determine the classification result of the target cell based on the target coupling information, and feed the classification result back to the microfluidic system. The hardware processing system includes: The central processing unit is used to receive multiple dual-channel coupled information acquired by the optical system and to schedule the neuromorphic chip and the graphics processor to process the dual-channel coupled information. A neuromorphic chip is used to perform sparse spatiotemporal decoding, cell detection and identification, cell tracking and real-time classification on multiple received dual-channel coupling information, and to determine multiple target coupling information corresponding to each target cell in the cell sample based on the cell tracking results, and to determine the classification result of the target cell based on the real-time classification result of the multiple target coupling information; The microfluidic system includes a microfluidic chip, and the microfluidic chip manipulates the cell sample in the following ways: rolling and rotating. The optical system includes: an inverted microscope and a spatial filter; The inverted microscope is used to magnify cell samples in the microfluidic chip; The spatial filter includes a spliced ​​bright-field filter and a fluorescence filter, which are used to filter half of the magnified imaging result of the inverted microscope with a bright-field light source to obtain a bright-field imaging result containing bright-field cells, and filter the other half with a fluorescence light source to obtain a fluorescence imaging result containing fluorescent cells. The central processing unit is used to schedule the neuromorphic chip and / or the graphics processor to process the dual-channel coupled information together; The neuromorphic chip is also used to perform three-dimensional reconstruction based on multiple target coupling information corresponding to the target cell, thereby obtaining a three-dimensional image of the target cell.

2. The cell analyzer according to claim 1, characterized in that, The microfluidic system includes: A micropump is used to continuously introduce the cell sample into the microfluidic chip; A microfluidic chip is used to manipulate and deliver imported cell samples, wherein the cell samples do not overlap in spatial position within the microfluidic chip; The sorting mechanism is used to classify and export cell samples from the microfluidic chip based on the classification results fed back by the hardware processing system.

3. The cell analyzer according to claim 2, characterized in that, The microfluidic chip manipulates the cell sample in at least one of the following ways: clustering, stretching and deformation, and dynamic acceleration.

4. The cell analyzer according to claim 1, characterized in that, The optical system includes: A dual-light source illumination system is used to simultaneously illuminate the microfluidic chip using a bright field light source and a fluorescent light source; An event camera is used to acquire the dual-channel imaging results to obtain dual-channel coupling information.

5. The cell analyzer according to claim 4, characterized in that, The dual-light source lighting system includes: A halogen lamp light source includes a halogen lamp and a bright field filter, used to filter the light source generated by the halogen lamp through the bright field filter to obtain the bright field light source therein; A mercury lamp light source includes a mercury lamp and a fluorescent filter, used to filter the light source generated by the mercury lamp through the fluorescent filter to obtain the fluorescent light source therein.

6. The cell analyzer according to claim 1, characterized in that, The neuromorphic chip is also used to extract multimodal features from the received dual-channel coupled information, and to reconstruct the image based on the results of the multimodal feature extraction to obtain a two-dimensional reconstruction result.

7. The cell analyzer according to claim 6, characterized in that, The neuromorphic chip is also used to perform three-dimensional reconstruction based on the two-dimensional reconstruction results of multiple target coupling information corresponding to the target cell, so as to obtain a three-dimensional image of the target cell.

8. The cell analyzer according to claim 1, characterized in that, The hardware processing system also includes a graphics processor; The central processing unit is used to schedule the neuromorphic chip and / or the graphics processor to process the dual-channel coupled information together.

Images