Method, device and equipment for evaluating interaction between nano preparation and cells

Abstract

The invention discloses a nano preparation and cell interaction evaluation method, device and equipment, and relates to the technical field of nano preparation and cell interaction research. A digital holographic imaging system is used for obtaining holograms of a nano preparation and cells in different time periods; numerical diffraction reproduction is carried out on the holograms in different time periods to obtain a real nano-scale phase distribution diagram of the cells, characteristic parameters of the cells are extracted from the real nano-scale phase distribution diagram, and a neural network diffusion model is constructed by using the characteristic parameters; a neural network diffusion model is trained by using a real nano-scale phase distribution diagram, so that the model can perform nano-scale accurate display on real dynamic evolution under the interaction of a nano preparation and cells, the design and development of novel nano drugs are promoted, and important technical support is provided for the field of nano medicine.

Description

Technical Field

[0001] The present invention relates to the technical field of research on the interaction between nano-preparations and cells, and in particular to a method, device and equipment for evaluating the interaction between nano-preparations and cells. Background Art

[0002] Nanotechnology is an emerging research field in medicine and pharmaceutical science. Nanoformulation is a drug preparation prepared by nanotechnology, whose core components have a size or structural unit within the nanoscale range (1-100 nanometers). This kind of preparation can directly nanosize the drug components, or it can be a system formed by combining drugs with nanocarrier materials. The main types of nanomaterials are liposomes, polymer micelles, nanoemulsions, nanocrystals, etc. As an emerging drug delivery system, it has great application potential in cancer treatment, gene delivery, vaccine development and other fields due to its unique physicochemical properties, adjustable size, surface functionalization ability and good biocompatibility. However, although nanoformulations have shown great potential in laboratory research, the cases of successful transformation into marketed nanomedicines are still extremely limited. The key lies in the lack of a comprehensive understanding of the behavior, distribution and action mechanism of nanoformulations in cells.

[0003] At present, the traditional evaluation methods for studying the interaction between nanoparticles and cells mainly include fluorescence microscopy and cryo-transmission electron microscopy. Fluorescence microscopy relies on the labeling of nanoparticles or cells, which can obtain the location of nanoparticles in cells, thereby revealing their action properties and cytotoxicity. Cryo-transmission electron microscopy is a method of choice for studying liposome structure and a powerful tool for analyzing core-shell nanoparticles by rapidly freezing samples to maintain their near-native state. However, it requires more careful sample preparation and is usually not suitable for long-term observation of living cells. However, these two types of research methods usually rely on the labeling of nanoparticles or cells, which may change the physical and chemical properties of nanoparticles, interfere with their natural behavior in cells, and may cause changes in cell function, leading to distortion and uncertainty of experimental results.

[0004] The emergence of digital holographic imaging technology provides a possible solution to the above problems. Through the two steps of interference recording and diffraction reproduction, digital holographic imaging technology can simultaneously restore the amplitude and phase distribution of the object light field, quantitatively acquire and analyze the three-dimensional information of the object, and has a high imaging resolution. However, when analyzing objects, it can often only analyze the three-dimensional information of static objects, and it is difficult to analyze the real dynamic evolution of the interaction between nano-agents and cells. Summary of the invention

[0005] The embodiments of the present invention provide a method, device and equipment for evaluating the interaction between nano-agents and cells, which can solve the problem in the prior art that it is difficult to analyze the real dynamic evolution of the interaction between nano-agents and cells.

[0006] The present invention provides a method for evaluating the interaction between a nanoparticle and a cell, comprising the following steps:

[0007] Obtaining holograms of the nanoparticles and cells at different time periods when the nanoparticles and cells interact;

[0008] Acquire the amplitude map and phase map of the hologram in different time periods; extract the real phase distribution area of ​​the cell from the phase map according to the amplitude map to form a real phase distribution map of the nanoscale cell; extract the characteristic parameters of the cell from the real phase distribution map of the nanoscale cell;

[0009] Construct a neural network diffusion model based on artificial neural network ANN; use the phase distribution map of cells in different time periods to train the neural network diffusion model in the forward diffusion process and reverse attention mechanism, and constrain the weights in the forward diffusion process and reverse attention mechanism according to the characteristic parameters to obtain the trained neural network diffusion model;

[0010] The phase distribution map of the nanoscale cells to be tested is input into the trained neural network diffusion model to obtain a cell image with highlighted nanopreparations, and the interaction mechanism between the nanopreparations and cells is identified from the cell image.

[0011] Preferably, the acquisition of the holograms of the nanoparticles and cells in different time periods comprises:

[0012] Obtain two groups of cells containing the nanoparticle and one group of cells not containing the nanoparticle, place the two groups of cells in culture dishes respectively, and place the culture dishes containing the cells horizontally on a sample platform;

[0013] Every 30 seconds, an off-axis digital holographic system is used to record a hologram of the nanoparticles and cells in the culture dish to obtain holograms of the nanoparticles and cells in different time periods.

[0014] Preferably, the step of acquiring the amplitude map and phase map of the hologram in different time periods comprises:

[0015] Identify the center point of the positive first-order spectrum in the hologram in different time periods, extract the positive first-order spectrum of the hologram in different time periods with this point as the center, and separate the diffraction terms of the positive first-order spectrum in the hologram in different time periods through inverse Fourier transform IFT;

[0016] Based on the diffraction terms of the positive first-order spectrum in the hologram in different time periods, the object light wavefront of the nanopreparations and cells in the holograms in different time periods is propagated from the holographic surface to the imaging surface using the angular spectrum diffraction reconstruction algorithm; and in the propagation process, the nanopreparations and cells in the holograms in different time periods are identified, and each cell and nanopreparation is digitally focused to obtain the amplitude diagram and wrapped phase diagram of the hologram in different time periods.

[0017] Preferably, the forming of a true phase distribution diagram of nanoscale cells comprises:

[0018] According to the amplitude diagrams in different time periods, the area corresponding to the plane part of the real object is extracted from the phase diagram in the corresponding hologram, and the extracted area of ​​the plane part is subjected to surface fitting using Zernike polynomials to obtain the phase form of the phase distortion corresponding to the distribution of the plane part;

[0019] The fitting surface is screened out from the area of ​​the plane part in different time periods according to the phase form, and aberration compensation is performed to obtain the phase image corresponding to the area of ​​the plane part of the real object in the phase image in different time periods;

[0020] The phase image was unwrapped to obtain the phase distribution map of cells in different time periods.

[0021] Preferably, the extracting characteristic parameters of cells includes:

[0022] According to the phase distribution diagram of cells in different time periods and using the standard resolution plate phase calibration, the characteristic parameters in the cells are obtained;

[0023] The standard resolution plate phase calibration steps are as follows:

[0024] Use a camera to take a picture of the calibration plate, use a projector to project a phase distribution map on the calibration plate, and use the camera to take pictures at the same time; and change the position of the calibration plate, and repeatedly take multiple pictures of the calibration plate with the projected phase distribution map;

[0025] Use the captured calibration plate images to calibrate the camera, and obtain the corner point position information on each calibration plate image; obtain the upper absolute phase map in the x and y directions on the calibration plate image of each projected phase distribution map, and convert the range of the phase map to the image row and column range; obtain the image point coordinates corresponding to each corner point according to the corner point position of each image and the converted phase map, and identify the characteristic parameters in the cell in the phase map;

[0026] The characteristic parameters include: projected surface area, sphericity coefficient, perimeter, roundness, volume, refractive index distribution, cell contour gradient and uncertainty of the cell.

[0027] Preferably, the obtained trained neural network diffusion model includes:

[0028] The phase distribution diagrams of cells in different time periods are used to train the forward diffusion process of the neural network diffusion model. During the training process, the weights of the neural network diffusion model are randomly initialized and constrained according to the characteristic parameters.

[0029] The phase distribution maps of cells in different time periods are used to train the reverse attention mechanism of the neural network diffusion model. During the training process, the weights of the neural network diffusion model are adjusted using characteristic parameters to reduce the loss value during the reverse attention mechanism training. The weights of the neural network diffusion model are updated according to the loss value to obtain the trained neural network diffusion model.

[0030] Preferably, the identification of the interaction mechanism between the nano-agent and the cell is carried out by identifying the amount of the highlighted nano-agent at different locations of the cell, thereby determining the degree of damage to the cell membrane caused by the nano-agent when the nano-agent interacts with the cell, as well as the degree of fusion of the cell.

[0031] The embodiment of the present invention also provides a device for evaluating the interaction between a nano-preparation and a cell, comprising:

[0032] An acquisition module, used to obtain holograms of the nano-agent and the cells at different time periods when the nano-agent and the cells interact with each other;

[0033] The phase unwrapping module is used to obtain the amplitude map and phase map of the hologram in different time periods; and extract the real phase distribution area of ​​the cell from the phase map according to the amplitude map to form a real phase distribution map of the nanoscale cell; extract the characteristic parameters of the cell from the real phase distribution map of the nanoscale cell;

[0034] The phase recognition module is used to construct a neural network diffusion model based on an artificial neural network ANN; the phase distribution diagrams of cells in different time periods are used to train the neural network diffusion model in the forward diffusion process and the reverse attention mechanism, and the weights in the forward diffusion process and the reverse attention mechanism are constrained according to the characteristic parameters to obtain the trained neural network diffusion model;

[0035] The evaluation module is used to input the phase distribution map of the nanoscale cells to be tested into the trained neural network diffusion model, obtain the cell image with the nano-preparation highlighted, and identify the interaction mechanism between the nano-preparation and the cells from the cell image.

[0036] An embodiment of the present invention further provides an electronic device, including a memory and a processor;

[0037] The memory is used to store computer programs;

[0038] The processor is used to implement the steps of the method for evaluating the interaction between nano-preparations and cells as described above when executing the computer program stored in the memory.

[0039] The embodiments of the present invention provide a method, device and apparatus for evaluating the interaction between nano-preparations and cells. Compared with the prior art, the beneficial effects thereof are as follows:

[0040] The present invention utilizes a digital holographic imaging system to obtain holograms of nano-preparations and cells in different time periods, and then performs numerical diffraction reproduction on the holograms in different time periods to obtain a true nanoscale phase distribution map of the cells, and extracts characteristic parameters of the cells from the true nanoscale phase distribution map, uses the characteristic parameters to construct a neural network diffusion model, and uses the true nanoscale phase distribution map to train the neural network diffusion model, so that the model can accurately display the true dynamic evolution of the interaction between the nano-preparation and the cell at the nanoscale, promote the design and development of new nano-drugs, and provide important technical support for the field of nanomedicine. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A schematic diagram of the overall process of a method for evaluating the interaction between a nano-preparation and cells provided in an embodiment of the present invention;

[0042] Figure 2 A 180nm liposome hologram of a method for evaluating the interaction between a nano-preparation and a cell provided in an embodiment of the present invention;

[0043] Figure 3 A red blood cell hologram of a method for evaluating the interaction between a nano-preparation and a cell provided in an embodiment of the present invention;

[0044] Figure 4 A schematic diagram of the positive primary encapsulation phase of a liposome in a method for evaluating the interaction between a nano-preparation and a cell provided in an embodiment of the present invention;

[0045] Figure 5 A schematic diagram of the positive primary encapsulation phase of red blood cells in a method for evaluating the interaction between a nano-preparation and cells provided in an embodiment of the present invention;

[0046] Figure 6 A liposome dedistorted phase image of a method for evaluating the interaction between a nano-preparation and a cell provided in an embodiment of the present invention;

[0047] Figure 7 A red blood cell dedistorted phase image of a method for evaluating the interaction between a nano-preparation and a cell provided in an embodiment of the present invention;

[0048] Figure 8A three-dimensional phase image of a single liposome in a method for evaluating the interaction between a nanoformulation and a cell provided in an embodiment of the present invention;

[0049] Fig. 9 A three-dimensional phase image of a single red blood cell in a method for evaluating the interaction between a nanoformulation and cells provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present invention, so the present invention is not limited by the specific embodiments disclosed below.

[0051] See also Figure 1 The present invention provides a method for evaluating the interaction between a nano-preparation and a cell, comprising the following steps:

[0052] Step 1: Use an off-axis digital holographic system to record holograms of cells and nanoparticles, including:

[0053] ① Take two groups of cells, one containing nano-preparations and the other not containing nano-preparations, and put them into culture dishes respectively. Place the culture dishes containing cells horizontally on the sample platform. The one with nano-preparations added is the experimental group, and the one without nano-preparations is the control group.

[0054] ②Every 30 seconds, record a hologram of the cells and nanoparticles in the culture dish.

[0055] Step 2: Perform numerical diffraction reconstruction on the recorded hologram to obtain the quantitative phase map of the cell at each time, specifically including:

[0056] ① Spatial filtering: Separate the diffraction terms of the hologram spectrum and extract the positive first-order spectrum in the hologram at different time periods; that is, find the center point of the positive first-order spectrum in the hologram spectrum, extract the positive first-order spectrum of the hologram with this point as the center, and separate the positive first-order diffraction term separately through inverse Fourier transform.

[0057] ② The object light wavefront is propagated from the holographic surface to the imaging surface through the angular spectrum diffraction reconstruction algorithm.

[0058] ③ Identify and distinguish cells and nanoparticles from the hologram, digitally focus each cell and nanoparticle, and obtain clear amplitude and package phase images;

[0059] ④ Extract the area corresponding to the plane part of the real object in the phase distribution, use the Zernike polynomial to fit the surface, get the distribution form of phase distortion, subtract the fitting surface from the phase distribution, and perform aberration compensation.

[0060] ⑤ Unwrap the phase image of the cell to restore the true phase distribution of the cell.

[0061] Step 3: Combined with the standard resolution plate phase calibration results, calculate the cell's projected surface area, sphericity coefficient, circumference, roundness, volume, refractive index distribution, cell contour gradient and uncertainty from the cell's phase distribution map. The specific steps for standard resolution plate phase calibration are:

[0062] Use a camera to take a picture of the calibration plate, use a projector to project a phase distribution map on the calibration plate, and use the camera to take pictures at the same time; and change the position of the calibration plate, and repeatedly take multiple pictures of the calibration plate with the projected phase distribution map;

[0063] The camera is calibrated using the captured calibration plate images to obtain the corner point position information on each calibration plate image; the upper absolute phase map in the x and y directions on the calibration plate image of each projected phase distribution map is obtained, and the range of the phase map is converted to the image row and column range; the image point coordinates corresponding to each corner point are obtained according to the corner point position of each image and the converted phase map, and the characteristic parameters in the cell in the phase map are identified.

[0064] Step 4: Using the eight characteristic parameters obtained in step 3 and the artificial neural network ANN, a neural network diffusion model is established to train a model that can identify whether cells contain nanoparticles, including:

[0065] ① Label the control group and the experimental group separately. In the experimental group, it is necessary to manually determine whether the cells are invaded by the nanoparticles. Divide these data into training set and test set in a ratio of 7:3.

[0066] ② Construct a diffusion model: input the phase image of unlabeled cells and training set data respectively. The model includes a forward diffusion process and a reverse attention mechanism training process. The output is the judgment result of whether the cell contains the nanoparticle and the cell image with the nanoparticle highlighted. Under the guidance of the input parameters and the input cell image, a model is obtained, and then tested with the test set until a model with an accuracy rate of more than 95% is obtained.

[0067] During the forward diffusion process and reverse attention mechanism training of the network diffusion model, the characteristic parameters are used to adjust the weights of the neural network diffusion model. In the forward diffusion process training, the weights of the model are randomly initialized. In the reverse attention mechanism training, the weights of the model are adjusted to reduce the loss value during training, and the loss value is used to update the weights of the model again to make the weights of the model optimal.

[0068] Step 5: Monitor the interaction process between the nano-preparation and the cells in real time, input the characteristic parameters obtained from the hologram into the model, obtain the judgment results and the cell image with the nano-preparation highlighted, and thus monitor the action mechanism between the nano-preparation and the cells in real time.

[0069] In the specific experiment, digital holography was used to track the interaction between polyalkylcyanoacrylate nanoparticles loaded with cabazitaxel and NIH-3T3 mouse fibroblasts, specifically:

[0070] 1. Preparation of Nanoparticles:

[0071] Cabazitaxel-loaded polyalkylcyanoacrylate (PACA) nanoparticles can be used to treat cancer; such polyethylene glycol nanoparticles are synthesized by emulsion polymerization of an aqueous phase containing cyanoacrylate monomers and an aqueous phase containing hydrochloric acid and polyethylene glycol surfactants.

[0072] 2. Cell culture:

[0073] NIH-3T3 mouse fibroblasts ( CRL-1658TM) was used for cell line and cell culture experiments; the cells were cultured in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum, 1 mM pyruvate and 2 mM glutamine, passaged 3 times a week, and the 5th to 20th generations were used in the nanoformulation interaction experiments.

[0074] 3. Nanoparticle-cell interaction experiment:

[0075] Mouse fibroblasts were cultured for 3-5 days until the cell density reached 90%, seeded at 15,000 cells / well in a 96-well plate, and incubated with 5 mg / mL of cabazitaxel-loaded PACA at 37°C and 5% CO2 for 24 hours.

[0076] 4. Label-free tracing data collection of nanoparticle-cell interactions:

[0077] At 0h, 1h, 2h, 4h, 6h, and every 2h thereafter, a digital holographic imaging system was used to observe the interaction between the nanoparticles and cells, and the three-dimensional phase images of the cells at different times and their parameters such as gradient and uncertainty were obtained. They were then divided into two categories, the results of cells without nanoparticles and the results of cells containing nanoparticles.

[0078] like Figure 2 and Figure 3 Shown are the holograms of 180 nm liposomes containing nanoparticles and the holograms of red blood cells without nanoparticles; Figure 4 and Figure 5 Shown are the positive first-order encapsulation phases extracted from the hologram of 180 nm liposomes containing nanoparticles and the hologram of red blood cells without nanoparticles; Figure 6 and Figure 7 The following are phase images obtained by dedistorting the positive first-order encapsulation phase from a 180nm liposome hologram containing nanoparticles and a red blood cell hologram without nanoparticles; Figure 8 and Fig. 9 Shown are three-dimensional phase images of 180 nm liposomes containing nanoparticles and red blood cells without nanoparticles. The characteristic parameters of each cell can be obtained based on the three-dimensional phase image.

[0079] 5. Establishment of a real-time monitoring model for cell status:

[0080] A diffusion model is constructed, and through the forward diffusion process and the reverse attention mechanism training process, under the specified conditions of the input parameters and the guidance of the input cell image, the denoising process is completed and the judgment result and the cell image with the nano-preparation highlighted are output, thereby obtaining a model that can judge the status of the nano-preparation and the cell in real time and can clearly locate the position of the nano-preparation in the cell.

[0081] The present invention utilizes the advantages of digital holographic microscopy, such as its nanometer-level sensitivity to phase, and combines it with a neural network diffusion model to construct a new research strategy for label-free, real-time monitoring of the interaction between nanoformulations and cells. This strategy breaks through the limitations of existing technologies and provides a new tool for studying the distribution, metabolism and action mechanism of nanoformulations in cells. It not only helps to deeply understand the biological action mechanism of nanoformulations, but also will promote the design and development of new nanodrugs, providing important technical support for the field of nanomedicine.

[0082] The embodiment of the present invention also provides a device for evaluating the interaction between a nano-preparation and a cell, comprising:

[0083] The acquisition module is used to obtain holograms of the nano-preparation and the cells in different time periods when the nano-preparation and the cells interact with each other.

[0084] The phase unwrapping module is used to obtain the amplitude map and phase map of the hologram in different time periods; and extract the real phase distribution area of ​​the cell from the phase map according to the amplitude map to form a real phase distribution map of nanoscale cells; extract the characteristic parameters of the cell from the real phase distribution map of nanoscale cells.

[0085] The phase recognition module is used to construct a neural network diffusion model based on the artificial neural network ANN; the phase distribution diagram of cells in different time periods is used to train the forward diffusion process and reverse attention mechanism of the neural network diffusion model, and the weights in the forward diffusion process and the reverse attention mechanism are constrained according to the characteristic parameters to obtain the trained neural network diffusion model.

[0086] The evaluation module is used to input the phase distribution map of the nanoscale cells to be tested into the trained neural network diffusion model, obtain the cell image with the nano-preparation highlighted, and identify the interaction mechanism between the nano-preparation and the cells from the cell image.

[0087] An embodiment of the present invention further provides an electronic device, including a memory and a processor.

[0088] The memory is used to store computer programs.

[0089] When the processor is used to execute the computer program stored in the memory, the steps of the above-mentioned method for evaluating the interaction between nano-preparations and cells are implemented.

[0090] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for evaluating the interaction between a nanoparticle preparation and a cell, characterized in that: The following steps are involved: Obtaining holograms of the nanoparticles and cells at different time periods when the nanoparticles and cells interact; Acquire the amplitude map and phase map of the hologram in different time periods; extract the real phase distribution area of ​​the cell from the phase map according to the amplitude map to form a real phase distribution map of the nanoscale cell; extract the characteristic parameters of the cell from the real phase distribution map of the nanoscale cell; Construct a neural network diffusion model based on artificial neural network ANN; use the phase distribution map of cells in different time periods to train the neural network diffusion model in the forward diffusion process and reverse attention mechanism, and constrain the weights in the forward diffusion process and reverse attention mechanism according to the characteristic parameters to obtain the trained neural network diffusion model; The phase distribution map of the nanoscale cells to be tested is input into the trained neural network diffusion model to obtain a cell image with highlighted nanopreparations, and the interaction mechanism between the nanopreparations and cells is identified from the cell image.

2. The method for evaluating the interaction between a nanoformulation and a cell according to claim 1, characterized in that: The acquisition of holograms of the nanoparticles and cells in different time periods includes: Obtain two groups of cells containing the nanoparticle and one group of cells not containing the nanoparticle, place the two groups of cells in culture dishes respectively, and place the culture dishes containing the cells horizontally on a sample platform; Every 30 seconds, an off-axis digital holographic system is used to record a hologram of the nanoparticles and cells in the culture dish to obtain holograms of the nanoparticles and cells in different time periods.

3. The method for evaluating the interaction between a nanoformulation and a cell according to claim 1, characterized in that: The step of obtaining the amplitude graph and phase graph of the hologram in different time periods comprises: Identify the center point of the positive first-order spectrum in the hologram in different time periods, extract the positive first-order spectrum of the hologram in different time periods with this point as the center, and separate the diffraction terms of the positive first-order spectrum in the hologram in different time periods through inverse Fourier transform IFT; Based on the diffraction terms of the positive first-order spectrum in the hologram in different time periods, the object light wavefront of the nanopreparations and cells in the holograms in different time periods is propagated from the holographic surface to the imaging surface using the angular spectrum diffraction reconstruction algorithm; and in the propagation process, the nanopreparations and cells in the holograms in different time periods are identified, and each cell and nanopreparation is digitally focused to obtain the amplitude diagram and wrapped phase diagram of the hologram in different time periods.

4. The method for evaluating the interaction between a nanoformulation and a cell according to claim 1, characterized in that: The phase distribution diagram of forming a real nanoscale cell includes: According to the amplitude diagrams in different time periods, the area corresponding to the plane part of the real object is extracted from the phase diagram in the corresponding hologram, and the extracted area of ​​the plane part is subjected to surface fitting using Zernike polynomials to obtain the phase form of the phase distortion corresponding to the distribution of the plane part; The fitting surface is screened out from the area of ​​the plane part in different time periods according to the phase form, and aberration compensation is performed to obtain the phase image corresponding to the area of ​​the plane part of the real object in the phase image in different time periods; The phase image was unwrapped to obtain the phase distribution map of cells in different time periods.

5. The method for evaluating the interaction between a nano preparation and a cell according to claim 1, characterized in that: The characteristic parameters of the extracted cells include: According to the phase distribution diagram of cells in different time periods and using the standard resolution plate phase calibration, the characteristic parameters in the cells are obtained; The standard resolution plate phase calibration steps are as follows: Use a camera to take a picture of the calibration plate, use a projector to project a phase distribution map on the calibration plate, and use the camera to take pictures at the same time; and change the position of the calibration plate, and repeatedly take multiple pictures of the calibration plate with the projected phase distribution map; Use the captured calibration plate images to calibrate the camera, and obtain the corner point position information on each calibration plate image; obtain the upper absolute phase map in the x and y directions on the calibration plate image of each projected phase distribution map, and convert the range of the phase map to the image row and column range; obtain the image point coordinates corresponding to each corner point according to the corner point position of each image and the converted phase map, and identify the characteristic parameters in the cell in the phase map; The characteristic parameters include: projected surface area, sphericity coefficient, perimeter, roundness, volume, refractive index distribution, cell contour gradient and uncertainty of the cell.

6. The method for evaluating the interaction between a nanoformulation and a cell according to claim 1, characterized in that: The trained neural network diffusion model is obtained, including: Construct a neural network diffusion model based on artificial neural network ANN; The phase distribution diagrams of cells in different time periods are used to train the forward diffusion process of the neural network diffusion model. During the training process, the weights of the neural network diffusion model are randomly initialized and constrained according to the characteristic parameters. The phase distribution maps of cells in different time periods are used to train the reverse attention mechanism of the neural network diffusion model. During the training process, the weights of the neural network diffusion model are adjusted using characteristic parameters to reduce the loss value during the reverse attention mechanism training. The weights of the neural network diffusion model are updated according to the loss value to obtain the trained neural network diffusion model.

7. The method for evaluating the interaction between a nanoformulation and a cell according to claim 1, characterized in that: When identifying the interaction mechanism between the nano-agent and the cell, the presence of the highlighted nano-agent at different locations of the cell is identified to determine the degree of damage to the cell membrane and the degree of fusion of the cell caused by the nano-agent when the nano-agent interacts with the cell.

8. A device for evaluating the interaction between nanoparticles and cells, characterized in that: include: An acquisition module, used to obtain holograms of the nano-agent and the cells at different time periods when the nano-agent and the cells interact with each other; The phase unwrapping module is used to obtain the amplitude map and phase map of the hologram in different time periods; and extract the real phase distribution area of ​​the cell from the phase map according to the amplitude map to form a real phase distribution map of the nanoscale cell; extract the characteristic parameters of the cell from the real phase distribution map of the nanoscale cell; The phase recognition module is used to construct a neural network diffusion model based on an artificial neural network ANN; the phase distribution diagrams of cells in different time periods are used to train the neural network diffusion model in the forward diffusion process and the reverse attention mechanism, and the weights in the forward diffusion process and the reverse attention mechanism are constrained according to the characteristic parameters to obtain the trained neural network diffusion model; The evaluation module is used to input the phase distribution map of the nanoscale cells to be tested into the trained neural network diffusion model, obtain the cell image with the nano-preparation highlighted, and identify the interaction mechanism between the nano-preparation and the cells from the cell image.

9. An electronic device, characterized in that: include: Memory and processor; The memory is used to store computer programs; The processor is used to implement the steps of a method for evaluating the interaction between a nanoformulation and a cell as described in any one of claims 1 to 7 when executing the computer program stored in the memory.

Images