Distributed artificial intelligence-based three-dimensional cell automatic production platform

Abstract

The present invention relates to a distributed three-dimensional cell production platform for personalized cell synthesis and treatment, in which, in cell culture and culture monitoring processes, cell culture conditions can be optimized through ultrasonic multimodal imaging and artificial intelligence analysis, and cells can be automatically cultured. Therefore, the platform according to the present invention enables a reduction in the cost required for culture by increasing the survival rate of cells during cell culture, and can improve the efficiency of cell culture.

Description

Distributed 3D Cell AI Automated Production Platform

[0001] The present invention was carried out under the support of the Ministry of Science and ICT under project number 2710005353 and project number 2024050019, with the research management agency for the above project being the National Research Foundation of Korea, the research project and research project title being “Research on the Development of a Snapshot Multispectral / Holographic Ultrasound Imaging Platform for 3D Label-Free Detection of Gastrointestinal Cancer”, the lead institution being Daegu Gyeongbuk Institute of Science and Technology, and the research period being 2024-05-01 - 2025-04-30.

[0002] In addition, the present invention was carried out under the support of the Ministry of Science and ICT under project number 2410001013 and project number 2024040054, the research management agency for the said project is the Korea Institute of Industrial Technology Planning and Evaluation, the research project and research project name is “Development of SoC for Multi-Channel Ultrasound Catheter and Probe for Cardiac Examination Applying the Same”, the lead organization is R&S Lab Co., Ltd., and the research period is 2024-05-01 - 2024-12-31.

[0003]

[0004] The present invention relates to a distributed three-dimensional cell production platform for personalized cell synthesis and treatment, and more specifically, to a platform capable of automatically culturing specific cells in three dimensions through ultrasound holograms, multimodal imaging, and artificial intelligence analysis.

[0005] Cell culture technology is essential in various biological studies, including research on tissue regeneration, drug screening, toxicity testing, and disease modeling. In particular, two-dimensional cell culture systems have been widely used in research for a long time, but they have limitations in that they do not adequately reflect the actual environment of cells. In two-dimensional culture, cells diffuse in a planar manner, and as a result, intercellular interactions, signal transduction, and interactions with the extracellular matrix may appear differently from the actual biological environment, making it difficult to represent the actual characteristics of the cultured cells.

[0006] Accordingly, research and development on 3D cell culture systems are actively underway to provide an environment similar to what cells experience within the actual human body. 3D cell culture is a technology that enables cells to grow and interact within a three-dimensional structure, which can contribute to maintaining or enhancing the functional characteristics of cells by providing a more accurate biomimetic environment.

[0007] Currently commercialized 3D cell culture systems include hydrogel-based, scaffold-based, bioprinting, and microfluidics-based systems. Hydrogel-based systems provide a gel-like matrix in which cells can be cultured, allowing them to naturally form 3D structures. They primarily utilize materials capable of mimicking the characteristics of the extracellular matrix (ECM), enabling cells to grow in an environment similar to the ECM. Scaffold-based systems use 3D structures made of various materials to allow cells to grow along specific shapes; they are widely used in the field of tissue engineering and are utilized in basic research for the regeneration of various tissues and organs. Bioprinting is a technology that creates 3D structures by stacking bioinks containing cells layer by layer, enabling the realization of precise structures. Microfluidics-based systems supply cells, nutrients, and gases through fine fluid channels and have the advantage of reproducing complex microenvironments.

[0008] However, these 3D culture systems have limitations depending on their respective methodologies. For example, hydrogel-based systems provide an environment for cell growth but present difficulties in controlling the gel's composition and strength according to specific cell types; scaffold-based systems offer structural stability but may result in uneven cell distribution, and the biocompatibility of the scaffold itself can be an issue. Bioprinting technology can realize highly precise structures but is costly and time-consuming. Microfluidics-based systems can provide complex microenvironments, but their design and fabrication are difficult, and they are often unsuitable for large-scale culture.

[0009] In addition, existing 3D culture systems faced difficulties in automating the entire cell culture process. In the existing 3D culture system, the process of monitoring cell growth and the process of controlling culture conditions and variables based on the monitoring results were carried out as separate processes. Since processes such as cell culture, cell monitoring, and changes in culture conditions based on monitoring results were all performed individually, it was difficult to culture large quantities of cells and to culture a large number of cells simultaneously.

[0010] Accordingly, the present invention aims to solve these problems of existing 3D cell culture systems and propose a new system capable of providing a more efficient and reproducible 3D cell culture environment. In particular, the present invention relates to an integrated culture platform, which utilizes an artificial intelligence model in the cell culture process to automatically configure the processes of cell culture, cell monitoring, and control of culture conditions, thereby making it possible to produce a variety of cells in small or large quantities, and to automatically produce target cells individually.

[0011] The inventors have constructed a three-dimensional cell culture system in which a cell culture vessel, a cell stimulation-capture unit responsible for stimulating and capturing cells, and an image acquisition unit for acquiring cell images are integrated, and have confirmed that the system can culture cells more efficiently.

[0012] Accordingly, the objective of the present invention is to provide a three-dimensional cell culture system.

[0013] Another objective of the present invention is to provide a three-dimensional cell culture method.

[0014] One aspect of the present invention is a three-dimensional cell culture system based on an ultrasound hologram, comprising: a cell culture vessel including a culture well where a sample containing one or more cells is located; a first light source that irradiates light toward the culture well from the upper side of the cell culture vessel; a first ultrasound transducer that is spaced apart from the culture well and generates ultrasound toward the culture well; a cell stimulation-capture unit including an ultrasound hologram generating lens that spatially phase-modulates the ultrasound generated from the first ultrasound transducer to form a set pattern shape on a target cell; and an image acquisition unit that acquires a multimodal three-dimensional cell image of a cell and a second ultrasound transducer that outputs high-frequency ultrasound toward the culture well by electrical signal control.

[0015] In the present invention, the sample may include one or more types of cells and may include an ultrasonic medium for ultrasonic holographic stimulation together with the cells.

[0016] In one embodiment of the present invention, the three-dimensional cell culture system may further include a control unit that analyzes the characteristics of a cell through a pre-learned analysis model from a cell image acquired through an image acquisition unit, and changes the cell stimulation parameters by changing the shape of a set pattern based on the analyzed results.

[0017] The pre-trained analysis model may be a deep neural network (DNN). A deep neural network may include an input layer and an output layer, or it may include multiple separate hidden layers in addition to the input layer and the output layer. A deep neural network may include a convolutional neural network (CNN), a recurrent neural network (RNN), a Long Short-Term Memory algorithm, a restricted Boltzmann machine (RBM), a deep belief network (DBN), a Q network, a U network, a Siamese network, etc.

[0018] In one embodiment of the present invention, the cell stimulation and capture unit may include an exchange wheel having a plurality of ultrasonic hologram generating lenses having different surface structures.

[0019] In one embodiment of the present invention, a first ultrasonic transducer may be disposed adjacently to the exchange wheel, so that the pattern of the ultrasonic hologram changes according to the rotational drive of the exchange wheel.

[0020] In one embodiment of the present invention, the exchange wheel may be driven to rotate in order to output a corresponding ultrasonic hologram pattern based on the analysis result of the control unit.

[0021] In one embodiment of the present invention, the image acquisition unit may include a second light source that irradiates a wavelength in the ultraviolet region.

[0022] In one embodiment of the present invention, the image acquisition unit may include an objective lens that focuses light output from a second light source to a second ultrasonic transducer.

[0023] In one embodiment of the present invention, the image acquisition unit may include a variable filter for acquiring an image of a desired wavelength region from light reflected from a target cell.

[0024] In one embodiment of the present invention, the image acquisition unit may include a camera that acquires a multimodal image of a target cell.

[0025] In one embodiment of the present invention, the first ultrasonic transducer may generate ultrasound to a target cell, capture the target cell through ultrasound, and culture the cell.

[0026] In one embodiment of the present invention, the multimodal three-dimensional cell image of the cell may be a high-frequency ultrasound B-Mode image, an ultrasound scattering coefficient image (IBC), and an ultrasound radiation force (ARFI) image.

[0027] In one embodiment of the present invention, the set pattern shape may be learned using the loss between the amplitude value of the acoustic field for each pixel AASM(x,y) derived on the target surface through the Angular Spectrum Method (ASM) or any ultrasound simulation based on the phase value for each pixel, and the amplitude value of the acoustic field for each pixel Agoal(x,y) of the target image.

[0028] In one embodiment of the present invention, the second ultrasonic transducer may be formed in a ring type to secure the path of light irradiated from the second light source.

[0029] In one embodiment of the present invention, the second ultrasonic transducer may emit a high-frequency ultrasonic microbeam for a single cell sample and emit a low-frequency ultrasonic beam for a plurality of cell samples to measure mechanical properties of the cell samples.

[0030] The present invention relates to a distributed three-dimensional cell production platform for personalized cell synthesis and treatment, which can optimize cell culture conditions and automatically culture cells through ultrasonic multimodal imaging and artificial intelligence analysis during the cell culture and culture monitoring processes. Accordingly, the platform according to the present invention can increase the viability of cells during cell culture, thereby reducing the cost of culture and improving the efficiency of cell culture.

[0031] FIG. 1 is a diagram illustrating the operation flow of a three-dimensional cell culture system according to one embodiment.

[0032] FIG. 2 is a perspective view of a three-dimensional cell culture system according to one embodiment.

[0033] FIG. 3 is a diagram illustrating the components of a three-dimensional cell culture system according to one embodiment.

[0034] FIG. 4 is a drawing showing an ultrasonic hologram generating lens according to one embodiment.

[0035] FIG. 5 is a diagram illustrating the various types of images that can be captured in a three-dimensional cell culture system according to one embodiment.

[0036] FIG. 6 is a diagram illustrating the operation method of a three-dimensional cell culture system according to one embodiment.

[0037] FIG. 7 is a diagram illustrating a deep neural network according to one embodiment.

[0038] In an ultrasound hologram-based three-dimensional cell culture system,

[0039] A cell culture vessel comprising a culture well in which a sample containing one or more cells is located;

[0040] A first light source that irradiates light toward the culture well from the upper side of the cell culture vessel;

[0041] A first ultrasonic transducer spaced at a certain distance from the culture well and generating ultrasound toward the culture well, and an ultrasonic hologram generating lens cell stimulation-capture unit that spatially phase-modulates the ultrasound generated from the first ultrasonic transducer to form a set pattern shape on a target cell; and

[0042] A three-dimensional cell culture system comprising: a second ultrasound transducer that outputs high-frequency ultrasound toward the culture well by electrical signal control; and an image acquisition unit that acquires a multimodal three-dimensional cell image of the cell.

[0043] The present invention will be described below with reference to the attached drawings. However, the present invention may be implemented in various different forms and is therefore not limited to the embodiments described herein. Furthermore, in order to clearly explain the present invention in the drawings, parts unrelated to the explanation have been omitted, and similar parts throughout the specification have been given similar reference numerals.

[0044] Throughout the specification, when it is stated that a part is "connected (connected, in contact, combined)" with another part, this includes not only cases where they are "directly connected," but also cases where they are "indirectly connected" with other members interposed between them. Furthermore, when it is stated that a part "includes" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but rather allows for the inclusion of additional components.

[0045] The terms used herein are merely for describing specific embodiments and are not intended to limit the invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this specification, terms such as “comprising” or “having” are intended to indicate the presence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0046]

[0047] Embodiments of the present invention will be described in detail below with reference to the attached drawings.

[0048]

[0049] FIG. 1 is a diagram illustrating the operation flow of a three-dimensional cell culture system according to one embodiment.

[0050] Referring to FIG. 1, a three-dimensional cell culture system according to one embodiment can perform cell stimulation through an ultrasonic hologram and enable three-dimensional cell culture in a non-contact environment by capturing cells through ultrasound, and can perform three-dimensional cell culture by acquiring a multimodal image through an equipped imaging system to acquire a three-dimensional cell image and analyzing the three-dimensional cell image quantitatively and qualitatively through an embedded artificial intelligence model.

[0051] A three-dimensional cell culture system according to one embodiment can operate as a closed-loop system. Accordingly, multimodal images acquired during the entire culture process are used as data to provide feedback on culture conditions, allowing the system to operate by continuously observing cells and adjusting culture conditions during the culture process. Specifically, under certain conditions, an ultrasound hologram is transmitted from the cell to acquire an image of the cultured cell. If the cell growth and culture conditions are found to be unsuitable through the analysis of the image, feedback is performed to modify the cell hologram or cell capture conditions, thereby enabling the cell culture to be integrally controlled under feedback within a single system.

[0052] As such, the 3D cell culture system according to one embodiment can be utilized as a fully automated 3D cell culture platform by having a closed-loop system in the stimulation and manipulation environment, and can solve the problem of unbalanced nutrient supply to cultured cells by performing appropriate feedback during the culture process. In addition, it can be utilized to produce a small amount of multi-variety cells or a large amount of a few varieties of cells. In particular, the 3D cell culture system according to one embodiment is a fully automated cell culture platform capable of culturing various varieties of cells, and has strengths in producing a small amount of multi-variety cells, so it can be utilized for the synthesis of personalized spheroids or organoids.

[0053] FIG. 2 is a perspective view of a three-dimensional cell culture system according to one embodiment.

[0054] Referring to FIG. 2, a three-dimensional cell culture system (1000) according to one embodiment may be composed of a single independent device.

[0055] Accordingly, a three-dimensional cell culture system according to one embodiment can provide optimized biophysical stimulation and a microculture environment without the implementation of complex and bulky hardware.

[0056] A three-dimensional cell culture system according to one embodiment may include a cell culture vessel, a cell stimulation-capture unit, and an image acquisition unit inside. The cell culture vessel, the cell stimulation-capture unit, and the image acquisition unit will be described below.

[0057] FIG. 3 is a diagram illustrating the components of a three-dimensional cell culture system according to one embodiment.

[0058] Referring to FIG. 3, a three-dimensional cell culture system (1000) according to one embodiment may include a cell stimulation-capture unit (100).

[0059] The cell stimulation-capture unit (100) may include a power supply unit (110), a control unit (120), a pulse generation unit (130), a first ultrasonic transducer (140), and an ultrasonic hologram generating lens (150). The cell stimulation-capture unit (100) may include a power supply unit (110) that applies power, a control unit (120) that generates an electrical control signal through power supplied from the power supply unit, a first ultrasonic transducer (140) that generates ultrasound using the electrical control signal as input, and an ultrasonic hologram generating lens (150) that changes the phase of the ultrasound generated by the ultrasonic transducer (140) to form phase-changed ultrasound at a plurality of focal points.

[0060] The cell stimulation-capture unit (100) may include an independent memory, and the memory may be electrically connected to the control unit (120) of the cell stimulation-capture unit.

[0061] The cell stimulation-capture unit (100) may be positioned at a certain distance from the lower side of the cell culture vessel (400) in a three-dimensional cell culture system (1000) according to one embodiment. Accordingly, the first ultrasonic transducer (140) within the cell stimulation-capture unit (100) may generate ultrasound from the lower side of the cell culture vessel (400) toward the upper side, and may apply ultrasound to a biological sample (e.g., a cell) contained in the cell culture vessel (400).

[0062] As described above, the first ultrasonic transducer (140) and the cell culture vessel (400) may be spaced apart by a certain distance, and an ultrasonic medium may be present in the spaced-apart space. When an ultrasonic medium is present, the influence of mechanical vibration of the ultrasonic transducer is reduced, so that the ultrasonic signal can be applied more clearly to the cell, and the phenomenon of ultrasonic loss during transmission can be minimized.

[0063] One or more first ultrasonic transducers may be included and may be arranged in an array form in the cell stimulation-capture section. When the first ultrasonic transducers are arranged in an array form, the phase of the ultrasound can be electrically controlled. When the phase of the ultrasound is controlled through multiple first ultrasonic transducers, it is possible to more easily move or control the position of cells within the cell culture vessel using ultrasound.

[0064] The ultrasonic medium can be a liquid medium, hydrogel, gaseous medium, or solid medium. When a liquid medium is used as the ultrasonic medium, the medium may be water or a culture medium commonly used in cell culture processes. When a hydrogel is used as the ultrasonic medium, alginate, gelatin, or polyacrylamide may be used. When a gas is used as the ultrasonic medium, carbon fluoride may be utilized, and when a solid is used, a polymer matrix may be used as the medium. In other words, mechanical vibrations caused by the generation of ultrasound by the ultrasonic transducer can be minimized by the ultrasonic medium, ensuring that only the ultrasonic signal is transmitted without vibrations being transmitted to the cells, thereby eliminating the influence of mechanical vibrations rather than ultrasound and excluding the probability of unintended results.

[0065] The first ultrasonic transducer (140) may be a single-element transducer or a multi-element transducer, and can be manufactured as a PZT (piezoelectric transducer), pMUT (piezoelectric micromachined ultrasonic transducer), cMUT (capacitive micromachined ultrasonic transducer), etc. The frequency of the ultrasonic transducer (140) may be set differently depending on the type of cell, the purpose of culture, and the culture environment.

[0066] Ultrasonic hologram refers to a technology that handles ultrasonic energy in three dimensions by utilizing cancellation between multiple ultrasonic waves, and a three-dimensional cell culture system (1000) according to one embodiment can be used to apply specific stimulation to cells in a biological sample or to capture cells through the ultrasonic hologram.

[0067] Ultrasound propagates in the form of plane or spherical waves, and by intentionally varying the transmission start times—that is, the phase of the transmitted signal—at multiple locations, it can be focused or moved to a desired point. Utilizing this principle, if a lens with an intentionally varied height distribution is placed in front of a single-element ultrasonic transducer, it is possible not only to focus the ultrasound to a desired point but also to generate ultrasound of a specific shape.

[0068] An ultrasonic hologram generating lens (150) may be positioned above the first ultrasonic transducer (140). Accordingly, an ultrasonic signal generated from the first ultrasonic transducer (140) may pass through the ultrasonic hologram generating lens (150) and reach a biological sample located in a cell culture vessel (400) with a phase change.

[0069] The ultrasonic hologram generating lens (150) can be designed to be detachable from the cell stimulation-capture unit (100), and the pattern of the ultrasonic hologram generating lens (150) can be used in different patterns depending on the type of cell and culture environment to be cultured. In addition, the pattern of the ultrasonic hologram generating lens (150) can be appropriately designed according to the target ultrasonic hologram pattern, and can be designed by utilizing an artificial intelligence model for the purpose of designing the ultrasonic hologram generating lens (150) with variables regarding the target ultrasonic hologram pattern and the structure of the ultrasonic hologram generating lens (150) as inputs.

[0070] For example, as illustrated in FIG. 4, an ultrasonic hologram generating lens (150) according to one embodiment can be designed with different height distributions of the lens in specific sections according to the target hologram pattern, thereby allowing the formation pattern of the ultrasonic hologram to be set differently.

[0071] The ultrasonic hologram generating lens (150) can be formed from a conventional general lens material and can be conveniently manufactured by a 3D printing technique. By using a 3D printing technique, a lens surface structure designed with varying height distributions can be quickly and easily realized, and a desired surface structure can be realized with high resolution.

[0072] The ultrasonic hologram generating lens (150) can be designed with different height distributions for each unit area of ​​the lens. When ultrasound is transmitted through the lens (150) designed with different height distributions, the spatial phase of the ultrasound can be modulated, and an ultrasonic hologram in the shape of a preset pattern can be formed on the surface of a target (biological sample). At this time, the pattern of the ultrasonic hologram generated by the ultrasonic hologram generating lens (150) can have various forms depending on the purpose, and can be various forms such as geometric structures like polygons and circles, specific patterns, letter shapes, specific designs, etc.

[0073] When an ultrasonic hologram is irradiated onto a biological sample, such as a cell to be cultured, within a cell culture vessel (400), the cell can be selectively stimulated. Specifically, since the ultrasonic hologram is delivered to the cell in a specific pattern, ultrasound can be applied to specific parts of the cell, and unnecessary ultrasonic energy can be prevented from being delivered to unwanted parts. Additionally, when ultrasound is selectively applied within a single culture well, an energy difference occurs between the area receiving the ultrasonic energy and the area not receiving it, thereby generating a force. This force can be used to capture cells, and the energy difference induced by the ultrasonic hologram can be used to apply force to the cells, thereby inducing them to grow into a desired shape.

[0074] The first ultrasonic transducer (140) and the ultrasonic hologram generating lens (150) may be configured as multiple channels for each culture well within a three-dimensional cell culture system (1000) according to one embodiment. For example, if four culture wells for cell culture are provided within a cell culture vessel (400), four first ultrasonic transducers (140) and four corresponding ultrasonic hologram generating lenses (150) may be provided for each culture well to apply different ultrasonic holograms. Therefore, when multiple culture wells are provided, cells can be stimulated and cultured in each culture well in a different pattern shape through the differently designed ultrasonic hologram generating lens and the first ultrasonic transducer (140).

[0075] The cell stimulation-capture unit (100) may include a pulse generation unit (130). The control unit (120) may generate an electrical control signal that changes according to parameters regarding the ultrasonic pulse signal and transmit it to the pulse generation unit (130), and the pulse generation unit (130) may generate a pulse signal for ultrasonic generation through the electrical signal received from the control unit (120).

[0076] The cell stimulation-capture unit (100) can induce cell growth by stimulating a specific part of a cell through an ultrasonic hologram, or capture and move a specific cell or cell cluster contained in a biological sample for cell culture. The ultrasonic transducer (140) can generate high-frequency ultrasound, thereby applying ultrasound around a specific cell so that the sample can move toward the focus of the ultrasound. And, when the cell is captured and fixed at the focus of the ultrasound, it can be subjected to pressure by the ultrasound.

[0077] As illustrated in FIG. 6, the cell stimulation-capture unit (100) may further include a lens exchange wheel (160). A plurality of ultrasonic hologram generating lenses (150) may be disposed on the lens exchange wheel (160), and the ultrasonic hologram generating lenses (150) disposed therein may each be for generating different ultrasonic holograms. A first ultrasonic transducer (140) may be disposed adjacent to the lens exchange wheel, and the ultrasonic hologram generating lens (150) through which the ultrasonic waves generated from the first ultrasonic transducer (160) pass may be changed according to the rotational drive of the lens exchange wheel (160). For example, when the cell stimulation-capture unit (100) operates to first generate one ultrasonic hologram and then generate another ultrasonic hologram, the lens exchange wheel (160) can be rotated so that the ultrasonic hologram generating lens (150) for generating the first ultrasonic hologram is adjacent to the first ultrasonic transducer (140), thereby aligning the ultrasonic hologram generating lens (150) and the first ultrasonic transducer (140). Subsequently, to generate another ultrasonic hologram, the lens exchange wheel (160) is rotated again to align the other ultrasonic hologram generating lens (150) with the first ultrasonic transducer, thereby allowing the other ultrasonic hologram to be generated into the desired biological sample.

[0078] The lens exchange wheel (160) may be driven by direct operation by the user, but it may also be driven by receiving an electrical signal from the control unit (120). Accordingly, a control signal according to a set protocol or user interface operation is transmitted from the control unit, and the lens exchange wheel (160) is driven by rotation to generate an appropriate ultrasonic hologram as a biological sample.

[0079] When cells are captured and moved using ultrasound, it is easier to acquire multimodal images through the image acquisition unit described later, and the shape of the cell cluster cultured in three dimensions can be maintained through cell capture even during the cell growth process, so it is more efficient to form a three-dimensional cell cluster, such as a spheroid or organoid, while minimizing unnecessary stimulation to the cells.

[0080] The control unit (120) may be part of a processor, implemented as a single semiconductor, or configured on a single printed circuit board (PCB). Alternatively, the control unit (120) may be implemented as software that can be executed on a computer based on a general-purpose processor and may be stored in a recording medium such as memory or storage. Alternatively, the control unit may be implemented as a programmable logic controller (PLC), a field-programmable gate array (FPGA), etc.

[0081] As described above, the control unit (120) may exist separately in the cell stimulation-capture unit (100) to control the operation of the first ultrasonic transducer (140) or the lens drive wheel (160), but may also exist integrated with the control unit (300) of the three-dimensional cell culture platform (1000) according to one embodiment. When the control unit (120) of the cell stimulation-capture unit (100) is integrated into the control unit of the three-dimensional cell culture platform (1000), the control unit (120) in the cell stimulation-capture unit (100) may be omitted, and the operation of the lens exchange wheel (160) and the ultrasonic transducer (140) may be controlled by the control unit (300) of the three-dimensional cell culture platform.

[0082] A three-dimensional cell culture system according to one embodiment may include an image acquisition unit (200).

[0083] The image acquisition unit (200) may include a first light source (210), a second light source (220), an objective lens (230), a variable filter (240), a second ultrasonic transducer (250), and a camera (260).

[0084] The first light source (210) can irradiate light of a specific wavelength from the upper side of the cell culture vessel (400) to the lower side, that is, toward a biological sample inside the cell culture vessel (400). At this time, the light of the specific wavelength may be visible light, and for example, the first light source (210) may be a white light source.

[0085] The light irradiated from the first light source (210) can increase the brightness of the cell culture vessel (400), making it easier to acquire images by the image acquisition unit (200).

[0086] The image acquisition unit (200) may include a second light source (220), and the second light source (220) may be a light source in a wavelength range capable of analyzing the state of a cell, including ultraviolet light. The light irradiated from the second light source (220) may be set differently depending on the type of image to be acquired, and the light irradiated through the second light source (220) may be reflected from the sample and incident on the variable filter (240) and camera (260) of the image acquisition unit (200) to acquire an image.

[0087] Although not illustrated in FIG. 3, the image acquisition unit (200) may further include a computation unit for calculating strain according to changes in cell thickness within a biological sample.

[0088] The second ultrasonic transducer (250) can apply ultrasound to a biological sample, and the image acquisition unit (200) can acquire multiple images of cells within the biological sample that are deformed by ultrasound. At this time, the computation unit can calculate the strain rate according to the change in thickness of the cells from the multiple images, and through this, the mechanical properties of the cells can be quantitatively measured.

[0089] The second ultrasonic transducer (250) can convert an input electrical signal into ultrasound and can generate ultrasound through cells in a biological sample. The ultrasound generated through the second ultrasonic transducer (250) can oscillate toward the biological sample in the same axial direction as the light irradiated from the second light source (220), and can generate ultrasound toward the biological sample in a direction perpendicular to the biological sample.

[0090] The second ultrasonic transducer (250) may have a ring-type shape, and if it has a ring-type shape, it may be easier to secure the path of light irradiated from the second light source (220).

[0091] The variable filter (240) can selectively transmit an image of a specific wavelength range from the image magnified through the objective lens (230) toward the camera (260). At this time, the specific wavelength can be set by considering various variables such as the type of image to be acquired and the characteristics of the cell.

[0092] The camera (260) can capture an image of a specific wavelength region transmitted through a variable filter (240) and can capture the shape of cells in a biological sample being deformed and restored by ultrasound. The camera (260) may be an ultra-high-speed camera and, for example, can acquire images of cells at a speed of 1000 fps or more.

[0093] The computational unit can measure the change in cell thickness pixel by pixel from multiple acquired phase difference images. Accordingly, the computational unit can calculate the strain before, during, and after ultrasound is applied to the cell, and consequently, the computational unit can calculate the elastic modulus of the cell using the intensity of the ultrasound stimulation and the strain of the sample within the biological sample.

[0094] Specifically, when a short impulse-shaped ultrasonic stimulus is applied to a biological sample from the aforementioned second ultrasonic transducer, multiple XY plane images of the cell before, during, and after the ultrasound is applied are obtained, and in addition, by obtaining information on the change in thickness of the cell in the Z-axis direction, i.e., displacement information, the computing unit can calculate the time it takes for the cell to recover to its original thickness after stimulation and map the local elasticity.

[0095] In addition, as shown in FIG. 5, the image acquisition unit (200) can acquire various types of images, such as multispectral images, ultrasound B-mode / reflection coefficient images, and elastic ultrasound images, in addition to calculating the elasticity modulus of the cell.

[0096] Multispectral imaging is an imaging technique that acquires images across various spectral bands using light of multiple wavelengths, and can be used to non-invasively analyze the tissue composition, oxygenation status, and blood flow of cells within biological samples. Because multispectral imaging combines information obtained from different wavelengths, it is possible to analyze the tissue characteristics of cells within biological samples more accurately.

[0097] Ultrasound B-mode / Reflectivity Imaging is the most common ultrasound imaging method, which generates two-dimensional images and allows for real-time visualization of cell cross-sections, thereby enabling clear identification of the internal structure of cells in real time without the use of invasive methods.

[0098] Elastography is a method of generating images by measuring the elasticity of tissues. It is used to evaluate tissue hardness and can be used to visually assess the hardness of cellular tissues within biological samples.

[0099] In this way, a three-dimensional cell culture system (1000) according to one embodiment can acquire multimodal images through a light source (210, 220) provided in an image acquisition unit (200) and an ultrasonic transducer (250), and can simultaneously determine various tissue characteristics of cells within a biological sample. The acquired images of the cells can be used to analyze the characteristics of the cells using an artificial intelligence model that has been pre-trained through a control unit (300) of the three-dimensional cell culture system.

[0100] A three-dimensional cell culture system according to one embodiment may include a control unit (300).

[0101] The control unit (300) can analyze the characteristics of cells through a pre-trained analysis model. The control unit (300) can receive various images of cells in a biological sample from the image acquisition unit.

[0102] Additionally, the received image can be analyzed through an analysis model learned in the control unit (300), and the characteristics of the cell analyzed in the control unit (300) can be used to subsequently change cell stimulation parameters or change the cell culture environment. To this end, the analysis results can be transmitted to the cell stimulation-capture unit (100) or the image acquisition unit (200).

[0103] For example, if, as a result of analyzing a cell image received from an image acquisition unit (200) through an analysis model, growth of a specific part of the cell is insufficient, the control unit (300) can select and transmit an appropriate ultrasonic hologram pattern to the cell stimulation-capture unit (100), and the cell stimulation-capture unit (100) can generate an ultrasonic hologram capable of stimulating the insufficient growth part toward the cell culture vessel based on this information.

[0104] The analysis model may be a deep neural network (DNN). A deep neural network may include an input layer and an output layer, or it may include multiple separate hidden layers in addition to the input layer and the output layer. A deep neural network may include a convolutional neural network (CNN), a recurrent neural network (RNN), a Long Short-Term Memory algorithm, a restricted Boltzmann machine (RBM), a deep belief network (DBN), a Q network, a U network, a Siamese network, etc.

[0105] The control unit (300) may include one or more cores and may include a processor for data analysis and deep learning, such as a central processing unit (CPU), a graphics processing unit (GPU), or a tensor processing unit (TPU) of a computing device. The processor may read a computer program stored in memory and perform data processing for machine learning according to one embodiment. According to one embodiment, the processor may perform operations for learning a neural network. The processor may perform calculations for learning a neural network, such as processing input data for learning in deep learning (DL), extracting features from input data, calculating errors, and updating the weights of the neural network using backpropagation. At least one of the CPU, GPU, and TPU of the processor may process the learning of a network function.

[0106] A schematic diagram of a deep neural network is shown in FIG. 7, and a deep neural network such as an analysis model according to one embodiment will be described based on FIG. 7.

[0107] A neural network can represent a model of a machine learning structure designed to extract feature data from input data and provide inference operations using the feature data. In this case, the feature data may represent data regarding features abstracted from the input data. Although FIG. 3 is illustrated as having three hidden layers for convenience of explanation, the hidden layers may include various numbers of layers. A neural network may include one or more layers, and each layer may include one or more nodes.

[0108] A node (or unit) is an element that constitutes each layer, and each layer can be composed of a node or a set of nodes. Within a neural network, nodes in layers other than the output layer can be connected to nodes in the next layer through links to transmit output signals. In this case, nodes in each layer can be interconnected via links, and nodes in the connected layers may be in an input node or output node relationship depending on whether signals are being transmitted or received. Among the nodes connected via links, the value of the output node's data can be determined based on the data input into the input node. Each node included in the hidden layer may receive the output of an activation function regarding the weighted inputs of the nodes included in the previous layer. Weighted inputs are inputs of the nodes included in the previous layer to which weights have been applied. In this case, weights can be variable and may vary depending on the function and algorithm of the neural network. Weights can be referred to as parameters of a neural network, and activation functions may include sigmoid, hyperbolic tangent (tanh), and rectified linear unit (ReLU).

[0109] Initial input nodes may refer to one or more nodes within a neural network to which data is directly input without passing through links in their relationships with other nodes. Alternatively, in terms of link-based relationships between nodes within the neural network, they may refer to nodes that do not have other input nodes connected by links. Similarly, final output nodes may refer to one or more nodes within a neural network that do not have output nodes in their relationships with other nodes. Furthermore, hidden nodes may refer to nodes constituting the neural network that are neither initial input nodes nor final output nodes.

[0110] A neural network can be trained using at least one of the following methods: supervised learning, which uses training data labeled with correct answers; unsupervised learning, which uses training data without labeled answers; semi-supervised learning; or reinforcement learning. In this process, an error can be calculated by comparing the output of the neural network with the labels or training data. The calculated error is backpropagated within the neural network (i.e., from the output layer to the input layer), and the connection weights of each node in each layer of the neural network can be updated based on this backpropagation. The amount of change in the connection weights of each updated node can be determined by the learning rate.

[0111] Overfitting is a phenomenon in neural networks where errors increase despite an increase in the number of training iterations due to excessive learning on the training data. Overfitting can act as a cause of increased errors in machine learning algorithms, and various optimization methods can be used to prevent it. To prevent overfitting, methods such as increasing the training data, regularization, dropout (which disables some nodes in the network during training), and the use of batch normalization layers can be applied.

[0112]

[0113] A three-dimensional cell culture system according to one embodiment may include a cell culture vessel (400).

[0114] The cell culture vessel (400) may include at least one culture well. The culture well may have a structure with an open bottom, and a transparent sheet capable of ultrasonic transmission may be placed in the open bottom portion.

[0115] A permeable sheet can be positioned to cover the cell culture vessel (400) from the bottom of the cell culture vessel (400), and a biological sample (e.g., a cell sample) can be placed on the surface area corresponding to the culture well. That is, when a permeable sheet is stacked on the bottom of the culture vessel, a cell can be positioned on the surface area exposed by the culture well.

[0116] The permeable sheet is formed to cover the entire lower surface of the cell culture vessel (400) and can be formed as an ultrasonic permeable sheet including a Mylar sheet or a PDMS (polydimethylsiloxane) sheet.

[0117] While general cell culture vessel materials such as plastic and glass are subject to ultrasonic attenuation and reflection effects, the embodiment of the present invention is implemented with a structure in which the bottom of the culture well is open, and at the same time, the bottom is covered with a Mylar sheet or PDMS sheet to eliminate concerns regarding related effects.

[0118] The cell culture vessel (400) can be installed on the upper part of the platform body with its bottom blocked by a permeable sheet.

[0119] FIG. 6 is a diagram illustrating the operation method of a three-dimensional cell culture system according to one embodiment.

[0120] A three-dimensional culture system (1000) according to one embodiment may be equipped with a cell culture vessel (400), an image acquisition unit (200) positioned below the cell culture vessel, a cell stimulation-capture unit (100) and a control unit (300) positioned below the cell culture vessel.

[0121] The first light source (220) of the image acquisition unit (200) may be a light source such as a white light source, and white light may be irradiated toward the biological sample of the cell culture vessel (400) to increase the brightness of the biological sample so that the image acquisition unit can acquire an image.

[0122] In the cell stimulation-capture unit (100), ultrasound generated from the first ultrasound transducer (140) at the lower side of the cell culture vessel (400) passes through an ultrasound hologram generating lens, allowing an ultrasound hologram with a changed phase of ultrasound to reach a biological sample within the cell culture vessel (400). When the ultrasound hologram is applied to the biological sample, a specific part of the cells within the sample may be stimulated by the ultrasound hologram. At this time, the stimulated location may be a part with insufficient nutrient supply, and through ultrasound hologram stimulation, more nutrients can be supplied, allowing the overall cell culture to be balanced.

[0123] In addition, the cell stimulation capture unit (100) can fix and capture cells using ultrasound, which makes it easy to stimulate the cells using ultrasound holograms. Also, the position of the cells can be moved arbitrarily in a non-contact environment, so three-dimensional cell culture can be performed more efficiently.

[0124] The image acquisition unit (200) can transmit ultraviolet light irradiated from the second light source (220) toward the cell culture vessel. The ultraviolet light generated by the second light source can be transmitted to the cell to be image acquired by passing through the objective lens (230) and the second ultrasonic transducer (250), and the ultraviolet light reflected from the cell in the biological sample can be incident on the variable filter by passing through the second ultrasonic transducer (250) and the objective lens (230) of the image acquisition unit (200). The variable filter (240) can selectively transmit light in a specific wavelength range among the reflected ultraviolet light to the camera, and by setting the wavelength band differently depending on the type of image to be obtained, the camera (260) can capture a multimodal image. The second ultrasonic transducer (250) can be formed in a ring type and can pass light generated from the second light source (220) through the hole in the center of the ring.

[0125] The image acquisition unit (200) is equipped with a second light source (220) for generating ultraviolet rays, a second ultrasonic transducer (250) for generating ultrasonic rays, and a first light source (210) capable of generating visible light such as white light, so that various types of images such as multispectral images, ultrasonic B-mode / reflectance coefficient images, and elastic ultrasonic images can be acquired and transmitted to the control unit.

[0126] The control unit (300) can receive images acquired from the image acquisition unit and can analyze the received images through an artificial intelligence model. The control unit (300) can generate analysis information by analyzing images generated from the image acquisition unit, such as multispectral images, ultrasonic B-mode / reflectance coefficient images, and elastic ultrasonic images, to analyze culture conditions, whether there is a lack of nutrients in a specific part of a cell, and whether there is an abnormality in a specific part of a cell. The analysis information can then be transmitted to the cell stimulation-capture unit (100), and the cell stimulation-capture unit (100) can perform feedback operations, such as applying ultrasonic holographic stimulation to cells that are not supplied with nutrients, through operations such as adjusting the ultrasonic hologram based on the analysis information. That is, in a 3D cell culture system (1000) according to one embodiment, the cell stimulation-capture unit (100), the image acquisition unit (200), and the control unit (300) are organically combined and undergo an internal feedback process in a closed loop form, making it possible to culture 3D cells, such as spheroids or organoids, more efficiently.

[0127]

[0128] The foregoing description of the present invention is for illustrative purposes only, and those skilled in the art will understand that other specific forms can be easily modified without altering the technical spirit or essential features of the present invention. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. For example, each component described as a single unit may be implemented in a distributed manner, and components described as distributed may likewise be implemented in a combined form.

[0129] The scope of the present invention is defined by the claims set forth below, and all modifications or variations derived from the meaning and scope of the claims and equivalent concepts thereof should be interpreted as being included within the scope of the present invention.

[0130] The present invention relates to a distributed three-dimensional cell production platform for personalized cell synthesis and treatment, and more specifically, to a platform capable of automatically culturing specific cells in three dimensions through ultrasound holograms, multimodal imaging, and artificial intelligence analysis.

Claims

1. In an ultrasound hologram-based three-dimensional cell culture system, A cell culture vessel comprising a culture well in which a sample containing one or more cells is located; A first light source that irradiates light toward the culture well from the upper side of the cell culture vessel; A first ultrasonic transducer spaced at a certain distance from the culture well and generating ultrasound toward the culture well, and an ultrasonic hologram generating lens cell stimulation-capture unit that spatially phase-modulates the ultrasound generated from the first ultrasonic transducer to form a set pattern shape on a target cell; and A three-dimensional cell culture system comprising: a second ultrasound transducer that outputs high-frequency ultrasound toward the culture well by electrical signal control; and an image acquisition unit that acquires a multimodal three-dimensional cell image of the cell.

2. In Paragraph 1, The above three-dimensional cell culture system is, A control unit that analyzes the characteristics of a cell from a cell image acquired through the above image acquisition unit using a pre-trained analysis model, and changes the cell stimulation parameters by changing the shape of the above-mentioned set pattern based on the analyzed results; A three-dimensional cell culture system that further includes 3. In Paragraph 1, The cell stimulation and capture unit described above includes an exchange wheel having a plurality of ultrasonic hologram generating lenses with different surface structures, and A first ultrasonic transducer is disposed adjacently to the above-mentioned exchange wheel, so that the pattern of the ultrasonic hologram changes according to the rotational drive of the exchange wheel. 3D cell culture system.

4. In Paragraph 3, A three-dimensional cell culture system in which the above-described exchange wheel is driven to rotate in order to output a corresponding ultrasonic hologram pattern based on the analysis results of the above-described control unit.

5. In Paragraph 1, The above image acquisition unit is, A second light source that irradiates wavelengths in a specific region, An objective lens that focuses light output from a second light source to a second ultrasonic transducer; A variable filter for acquiring an image of a desired wavelength region from light reflected from a target cell; and A camera for acquiring multimodal images of target cells; A three-dimensional cell culture system comprising 6. In Paragraph 1, A three-dimensional cell culture system in which the first ultrasound transducer generates ultrasound toward a target cell, captures the target cell through ultrasound, and cultures the cell.

7. In Paragraph 1, Multimodal cell imaging of the above cells is radiofrequency ultrasound B-Mode imaging, radiofrequency ultrasound scattering coefficient imaging (IBC) and radiofrequency ultrasound radiation force (ARFI) imaging, and a three-dimensional cell culture system.

8. In Paragraph 2, The above-mentioned setting pattern shape is a 3D cell culture system composed of a phase control map generated through a network trained using an arbitrary optimization algorithm or by utilizing the loss between AASM(x,y), the amplitude value of the acoustic field per pixel derived on the target surface through the Angular Spectrum Method (ASM) or arbitrary ultrasound simulation based on the phase value per pixel, and Agoal(x,y), the amplitude value of the acoustic field per pixel of the target image.

9. In Paragraph 5, A three-dimensional cell culture system in which the second ultrasonic transducer is formed in a ring type and can secure the path of light irradiated from the second light source.

10. In Paragraph 1, A three-dimensional cell culture system in which the second ultrasonic transducer oscillates a high-frequency ultrasonic micro beam for a single cell sample and oscillates a low-frequency ultrasonic beam for a plurality of cell samples to measure the mechanical properties of the cell samples.

11. A three-dimensional mass cell culture platform comprising one or more of the three-dimensional cell culture systems of claim 1.

Images