Method for measuring the rotational speed of particles

Abstract

To easily measure a rotation speed of a fine particle.SOLUTION: A measurement method for a rotation speed of a particle includes: a similarity calculation step 102 of defining one frame in a plurality of frames with respect to a continuous image, which is obtained by imaging the particles, as a selection frame and calculating similarity between a particle captured in the selection frame and a particle captured in the other frame with the lapse of time; and a rotation speed calculation step 103 of calculating the rotation speed of the particle by defining a time interval, in which a frame of maximal similarity appears, as a time for the particle to rotate once.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to a method for measuring the rotational speed of particles and a cell evaluation apparatus.

Background Art

[0002] Various rotational speed measurement devices for non-contact measurement of the rotational speed of a rotating body are known. For example, there is a device that attaches a reflective tape or the like to a rotating body and measures the rotational speed based on the light reception interval of the reflected light from the reflective tape. Such a device is simple, but the measurement target is limited to those to which a reflective tape or the like can be attached.

[0003] In recent years, devices that capture an image of a rotating body and measure the rotational speed by recognizing the reflection of light in a structural feature having a different luminance, color, etc. from other parts of the rotating body surface have also been studied (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, conventional measurement devices can only measure objects that conveniently include structural features with different reflectivities. In the case of minute particles such as cells, since there are almost no structural features, it is not practical to measure the rotational speed by such a method.

[0006] An object of the present disclosure is to enable easy measurement of the rotational speed of minute particles.

Means for Solving the Problems

[0007] One embodiment of the particle rotation speed measurement method of the present disclosure comprises a similarity calculation step of selecting one frame from a plurality of frames of a series of images of particles and calculating the similarity between the particles captured in the selected frame and the particles captured in the other frames over time, and a rotation speed calculation step of calculating the rotation speed of the particles by defining the time interval in which the frame with the maximum similarity appears as the time it takes for the particle to complete one rotation.

[0008] One method for measuring particle rotation speed calculates the rotation speed of a particle by defining the time interval at which frames with maximum similarity of particle images appear as the time it takes for the particle to complete one rotation. This method allows for easy measurement of the rotation speed of particles with few structural features without the need to mark the particles being measured.

[0009] In one embodiment of a method for measuring particle rotation speed, the change in particle rotation speed over time can be evaluated by selecting a frame over time and performing a similarity calculation step and a rotation speed calculation step.

[0010] One aspect of the drug evaluation method of the present disclosure involves a method for measuring rotational speed of the present disclosure, in which particles are considered cells, and the effect of the drug on cells is evaluated by evaluating the change in the rotational speed of the cells before and after supplying the drug to the cells.

[0011] One embodiment of the cell evaluation apparatus of this disclosure comprises an observation cell to which an electric field can be applied to cells, an imaging unit that captures a series of images of cells in the observation cell, and a rotation speed measurement unit that analyzes the series of images and calculates the rotation speed of the cells. The rotation speed measurement unit calculates the similarity between a cell captured in one selected frame from a plurality of frames and a cell captured in another frame over time, and calculates the rotation speed of the cells by taking the time interval at which the frame with the maximum similarity appears as the time it takes for the cell to complete one rotation.

[0012] In one embodiment of the cell evaluation device, the rotation speed measurement unit can calculate the rotation speed of cells over time.

[0013] One aspect of the cell evaluation device further includes a drug supply unit that supplies a drug to cells in an observation cell, and the rotation speed measurement unit can calculate the rotation speed of the cells over time before and after the supply of the drug.

Advantages of the Invention

[0014] According to the method for measuring the rotation speed of particles of the present disclosure, the rotation speed of minute particles including cells can be easily measured.

Brief Description of the Drawings

[0015] [Figure 1] It is a flowchart showing a method for measuring the rotation speed according to an embodiment. [Figure 2] It is a diagram showing one frame of continuous images. [Figure 3] It is a flowchart showing a similarity calculation step. [Figure 4] It is a diagram showing the formation process of an analysis image group. [Figure 5] It is a diagram showing an example of similarity evaluation. [Figure 6] It is a block diagram showing an evaluation device according to an embodiment. [Figure 7A] It is a schematic diagram showing an example of an electrode chip. [Figure 7B] It is a schematic diagram showing an example of an electrode chip. [Figure 7C] It is a schematic diagram showing an example of an electrode chip. [Figure 7D] It is a schematic diagram showing an example of an electrode chip. [Figure 8] It is a plan view showing an electrode chip used in an example regarding measurement accuracy. [Figure 9] It is a map showing an example of calculation of similarity in an example regarding measurement accuracy. [Figure 10] It is a heat map showing the similarity obtained in an example regarding measurement accuracy. [Figure 11] It is a plan view showing an electrode chip used in an example regarding drug stimulation response. [Figure 12]This is a heatmap showing the similarity obtained in the examples related to drug-induced response. [Figure 13] This graph shows the change in rotational speed over time obtained in the example relating to the drug stimulus response. [Figure 14] This graph shows the time course of relative rotational speed obtained in the examples relating to drug-induced response. [Modes for carrying out the invention]

[0016] As shown in Figure 1, a rotation speed measurement method according to one embodiment includes a similarity calculation step 102 in which a frame is selected from a series of images of the particles to be measured that were prepared in a series of image preparation step, and the similarity between the particles to be measured that appear in the selected frame and the particles to be measured that appear in other frames is calculated over time, and a rotation speed calculation step 103 in which the rotation speed of the particles is calculated by taking the time interval in which the frame in which the similarity is maximized as the time it takes for the particles to be measured to complete one rotation.

[0017] The continuous images prepared in the continuous image preparation step can be, for example, a video of the particles to be measured under a microscope. A continuous image with a constant frame rate is preferred because it is easy to correlate with the passage of time. The frame rate is not particularly limited, but from the viewpoint of accurately calculating the rotation speed, it is preferable that there are several dozen images for one rotation of the particle. For this reason, when the rotation speed of the particle is about 1 to 5 rotations per second, it is preferable to set the frame rate per second to about 30 to 60 fps. Furthermore, if the passage of time in each frame can be accurately determined, a video with a variable frame rate can also be used. The images are preferably grayscale images in order to facilitate analysis.

[0018] Figure 2 illustrates one frame of a sequence of images. In Figure 2, the particles are shown as cells 201 within an observation cell having comb-shaped electrodes 301-304. Multiple cells 201 are visible in the image.

[0019] As shown in Figure 3, the similarity calculation step 102 includes an analysis target extraction step 102A which generates an image from which the analysis target has been extracted by image processing of a series of images, and a calculation step 102B which calculates the similarity of the extracted analysis target.

[0020] In the analysis target extraction step 102A, first, one of several particles in the image is selected as the particle to be measured, and the region in which the selected particle is visible is selected as the analysis target region. The selected region should contain only one particle in all frames of the continuous image. Furthermore, from the viewpoint of preventing non-particle objects from being mistaken for particles, it is preferable that the selected region does not contain anything other than particles. For example, in Figure 2, it is preferable to select region 351 so that electrodes 301 to 304 are not included. The size of the selected region can be determined by considering the size of the continuous image to be analyzed, the size of the microgrid where the particles are located, the magnification of the objective lens of the microscope used for observation, etc. The size of the region can be fixed in advance, or the user can arbitrarily determine the size. The shape of the region is not particularly limited and can be square, rectangular, polygonal, circular, etc.

[0021] Next, the particles to be measured are extracted from the selected region. The extraction operation is not particularly limited, but for example, the image can be binarized and then the particle contours can be extracted. The image binarization is not particularly limited, but for example, "Otsu's Law" is preferred because it automatically calculates a threshold from the input image. Alternatively, binarization can be performed using simple thresholding or adaptive thresholding. These thresholds can be entered by the user to select the optimal value. Contour extraction is not particularly limited, but for example, it can be done using "OpenCV findContours," a type of Open Source Computer Vision Library (OpenCV). Regions other than the extracted particles can be filled with black, for example.

[0022] The extraction of target particles as described above is performed for all frames of the continuous video to be analyzed. As a result, a series of analysis images 406 corresponding to each frame of the continuous video is generated, as shown in Figure 4. Note that while it is possible to create the analysis image set by processing all frames of the prepared continuous video, it is also possible to create the analysis image set by selecting only some frames. Furthermore, while the extraction of target particles can be performed sequentially by selecting regions and extracting particles for each frame, it is also possible to select regions for all frames first and then extract particles all at once.

[0023] The calculation step 102B calculates the similarity between images in the analysis image group 406. It calculates the similarity between one selected analysis target frame fs in the analysis image group 406 and each analysis image up to an analysis image that is a predetermined D frames away from fs. fs can be the first frame in the analysis image group 406, but any position can also be selected. D can be set considering the rotation speed of the particles, the frame rate of the continuous images, etc. For example, if particles that rotate about once per second are photographed at a frame rate of 30 fps, it is preferable to set D to about 30 to 100.

[0024] The calculation of similarity is not particularly limited, but can be performed using template matching. For example, the analysis image of the selected frame fs can be used as a template, and the similarity can be evaluated by matching it with the analysis image of fs+1. The similarity is calculated sequentially from fs+1 to fs+D. The template matching is not particularly limited, but can be performed using OpenCV, such as "OpenCV matchTemplate". Similarity can be evaluated using a normalized correlation function such as "OpenCV TM_CCOEFF_NORMED". In this case, -1 can be used to indicate no similarity between the two images, and 1 can indicate a perfect match. Note that the evaluation of similarity is not limited to the method using a normalized correlation function. Furthermore, similarity can be evaluated using methods other than template matching.

[0025] Figure 5 shows an example of evaluating the similarity between each frame from fs+1 to fs+D and fs. In Figure 5, the similarity is shown by the intensity of the color, with lighter colors indicating a higher degree of similarity.

[0026] Next, in the rotation speed calculation process, the rotation speed of the particle is calculated. As the particle rotates, a surface different from the original state is gradually captured, and the similarity decreases. As the rotation continues and one full rotation is completed, a surface close to the original state is captured, and the similarity reaches its maximum. In Figure 5, the similarity is low in the 2nd, 7th, 12th frames, etc., and high in the 5th, 10th, 15th frames, etc. This means that the particle rotates once at intervals of 5 frames where the similarity is maximized. Therefore, the time required for one rotation of the particle can be determined by the frame interval where the similarity is maximized and the frame rate, and the rotation speed of the particle can be calculated. The method for calculating the rotation speed is not particularly limited, but for example, if the frame interval where the similarity is maximized is Δf and the frame rate is F, the rotation speed Ω of the particle can be calculated by the following equation (1).

[0027]

number

[0028] If the particle's rotation speed does not change, the same result can be obtained regardless of which frame position up to fs+D is used to detect the frame interval at which the similarity is maximized. However, if the rotation speed changes over time, in order to accurately determine the rotation speed Ωs at the time of the selected frame fs, it is preferable to detect the position of the frame in which the particle makes its first rotation from the frame fs and use this to calculate the rotation speed.

[0029] In this case, a minimum frame position fmin and a maximum frame position fmax can be set to detect the frame showing the maximum similarity, and a set of rotation speed calculation images can be created by extracting frames within the range of the minimum frame position fmin and the maximum frame position fmax. By using this set of rotation speed calculation images to detect the position of the frame showing the maximum similarity, the position of the frame in the first rotation can be easily detected. For example, in the case of Figure 5, a set of rotation speed calculation images can be created by setting fmin=2 and fmax=7.

[0030] Furthermore, even for particles whose shape changes over time, detecting the frame position during the first rotation makes it less susceptible to the effects of shape changes, thus simplifying the calculation of rotational speed.

[0031] Alternatively, instead of detecting the frame position of the first rotation, you can set the minimum frame position fmin and maximum frame position fmax to detect the frame position of the (n-1)th rotation and the frame position of the nth rotation, which are set in advance.

[0032] While we have shown an example of pinpointing the rotation velocity Ωs at a specific time point in the frame fs being analyzed, it is possible to evaluate the change in particle rotation velocity over time by shifting the position of the frame fs being analyzed, detecting the frame position where the similarity is maximized, and sequentially calculating the rotation velocity at each time position.

[0033] By evaluating the change in rotational speed over time, it is possible to clarify, for example, how the rotation of a particle changes in response to a specific stimulus.

[0034] The above method for measuring particle rotation speed makes it possible to calculate the rotation speed of particles without assigning any special marks to them. Furthermore, this method can measure the rotation speed and its change over time for particles that have few cellular-like shape features and whose inter-particle shapes are heterogeneous.

[0035] In this embodiment, we have shown an example of evaluating similarity frame by frame from the frame to be analyzed fs to the analysis completion frame fs+D. However, in cases where there are many frames, it is also possible to evaluate similarity at predetermined intervals of any number of frames. Furthermore, although we have shown an example of evaluating similarity in the positive direction, it is also possible to evaluate similarity in the negative direction. When evaluating changes over time, instead of selecting the frame to be analyzed fs by shifting it one frame at a time, it is also possible to select it by shifting it at predetermined intervals of any number of frames.

[0036] The sequential images of the particles can be prepared in any way, but if the particles are cells, a sequential video can be captured using the following evaluation device and the rotation speed can be measured.

[0037] As shown in Figure 6, the evaluation device 500 includes a rotation generation unit 501 that rotates the cells, a liquid delivery unit 502 for stimulating the cells with drugs, an observation unit 503 for observing the rotational movement of the cells, an imaging device 504 for photographing the observed cell rotation, a rotation speed measurement unit 505 that calculates the rotation speed from a series of cell images acquired from the imaging device 504, an input device 506 for inputting instructions to these elements, and a control device 507 for transmitting the instructions input from the input device 506 to each of the aforementioned units.

[0038] The rotation generation unit includes an arbitrary waveform generator 511, a wiring unit 512, an electrode tip 513, and an observation cell 514. The arbitrary waveform generator 511 is a voltage supply unit, and in this example, it is controlled by a control device 507. For example, when supplying an AC voltage to the electrode tip 513, the timing of supply to the electrode tip 513, the frequency and phase of the AC voltage, etc., are controlled by the control device 507.

[0039] The arbitrary waveform generator 511 has the function of outputting AC voltages of multiple types of sine waves with phases shifted by 90° in order to induce electrical rotation of cells. The frequency and voltage ranges of the multiple types of sine waves are the same, preferably in the range of 1kHz to 100MHz and 0Vpp to 20Vpp, and more preferably in the range of 10kHz to 10MHz and 1Vpp to 5Vpp. The four types of sine waves do not necessarily have to be output from a single device, and multiple generators may be used. The output control and waveform control of the arbitrary waveform generator 511 do not necessarily have to be performed by the control device 507, and may be performed by the arbitrary waveform generator 511.

[0040] The wiring unit 512 can apply the AC voltage signal generated by the arbitrary generator 511 to any electrode on the electrode tip 513.

[0041] The electrode tip 513 is not limited in its configuration as long as it can induce electrical rotation in the cells contained in the observation cell 514. For example, it can be an electrical rotation tip as shown in Figures 7A to 7D. The electrical rotation tip 513A shown in Figure 7A consists of four electrodes and can induce electrical rotation in the cell placed in the center of the electrodes. The electrical rotation tip 513A is the most basic electrode tip, and a method for measuring cells using the electrical rotation tip 513A is disclosed, for example, in Analytical Chemistry 1998 70 (13), 2607-2612.

[0042] The electric rotation chip 513B shown in Figure 7B is constructed by placing two comb-shaped electrode substrates, each having two microband electrodes, opposite each other orthogonally. This allows for the creation of a high-density microgrid formed by four microband electrodes. By applying AC voltages with a 90° phase difference to the two comb-shaped electrode substrates, a rotating electric field is induced within the microgrid. By introducing cells into the electric rotation chip 513B and inducing electric rotation, the electric rotation of hundreds of cells can be measured in a single experiment. For example, a cell measurement method using the electric rotation chip 513B is disclosed in Analyst, 2020, 145, 4188-4195.

[0043] The electric rotation chip 513C shown in Figure 7C has four electrodes wired to microwells for capturing cells, and allows for electric rotation measurement with cells captured in the microwells, as disclosed in, for example, Japanese Patent Application Publication No. 2021-185813.

[0044] The electric rotating tip 513D shown in Figure 7D bundles three microelectrodes, and by applying AC voltages with a 120° phase difference to each microelectrode, a rotating electric field is induced at the tip of the microelectrode. By bringing the microelectrode close to a cell, the cell can be rotated (Adv. Sci. 5, 1700711 (2018)).

[0045] In the case of the illustrated electro-rotation chips 513B and 513C, electro-rotation measurements can be performed on multiple cells simultaneously. Furthermore, by using a configuration like that of the electro-rotation chip 513C, electro-rotation measurements can be performed while the cells are held in microwells. This indicates that the solution can be changed to one containing a drug during the electro-rotation measurement process, and the effect of the drug on the cells can be measured as a change in electro-rotation speed.

[0046] The fluid delivery unit 502 can stimulate cells contained in the observation cell 514 by administering a drug solution. The fluid delivery unit 502 has a fluid delivery pump whose flow rate and operating timing are controlled by the control device 507. The fluid delivery unit 502 controls the amount, type, and timing of the solution introduced into the observation cell 514. There are no special restrictions on the fluid delivery unit 502 as long as it can introduce a predetermined solution onto the electrode tip 513 at a predetermined timing and flow rate. The fluid delivery pump also only needs to be able to control the flow rate of the solution, and there are no restrictions on the method of control, but the use of a peristaltic pump, for example, is preferred.

[0047] The observation unit 503 only needs to be able to observe the movement of cells rotating on the electrode tip 513, and in this case, it consists of an illumination device, an objective lens, and a display device.

[0048] The imaging device 504 can capture and record the image formed by the objective lens. There are no restrictions on the imaging method; for example, a CCD camera or a CMOS camera can be used. The imaging interval of the camera is an important factor that affects the accuracy of the rotation speed analysis. Electrically rotating cells rotate about 1 to 10 times per second. Accurate calculation of the rotation speed requires multiple images per rotation. Therefore, an imaging interval of 30 fps (frames per second) or higher is preferable for the imaging device 504. The shorter the imaging interval, the more images are captured per rotation, which leads to a more accurate calculation of the rotation speed, so it is even more preferable to be able to image at high speeds such as 60 fps or 100 fps.

[0049] The timing and duration of the imaging device 504 are controlled by the control device 507. These controls do not necessarily have to be performed by the control device 507; they can also be performed directly by the imaging device 504.

[0050] The rotation speed measurement unit 505 is a unit that calculates the rotation speed of cells on the electrode chip 513 acquired from the imaging device 504, and includes a processing unit 550, a parameter setting unit 551, a recording unit 552, and a display unit 553.

[0051] The processing unit 550 calculates the rotation speed from the continuous images. The processing unit 550 includes an image processing unit 556 that processes the continuous images captured by the imaging device 504 to generate an image for rotation speed measurement, a similarity calculation unit 557 that performs similarity calculation between frames, and a rotation speed calculation unit 558 that performs rotation speed calculation. The frame rate F, fmin, and fmax required for the rotation speed calculation process, when acquiring the continuous images, can be input to the parameter setting unit 551, for example, by the user using the input device 506.

[0052] The image set of the analysis region output from the image processing unit 556, the list of similarity scores for each frame output from the similarity calculation unit 557, and the rotation speed Ω output from the rotation speed calculation unit 558 are stored in the recording unit 552 and referenced according to requests from the processing unit 550.

[0053] The display unit 553 displays information to the user from the data stored in the recording unit 552, according to the information input by the user to the rotation speed measurement unit 505 via the input device 506.

[0054] By using such an evaluation device 500, the rotation speed of cells can be easily and accurately measured. In particular, by evaluating the change in rotation speed over time, the state of cells can be evaluated using the change in electrical rotation speed as an indicator, or the effect of drugs on cells can be evaluated using the state of cells as an indicator. [Examples]

[0055] (Verification of evaluation accuracy) <Evaluation device> The cell rotation speed was measured using the evaluation apparatus shown in Figure 6. As the electrode tip 513, a three-dimensional grid electrode was used, consisting of two comb-shaped electrode substrates 311 superimposed orthogonally to each other via a 30 μm thick spacer, as shown in Figure 8. Each of the two comb-shaped electrode substrates 311 has two comb-shaped microband electrodes 313 and 314 arranged alternately. The width of the combs was 20 μm, and the spacing between the combs was 30 μm. Each actual microband electrode consists of 20 comb teeth, and there are 40 comb teeth on one electrode substrate. Indium tin oxide (ITO) was used as the electrode material for visibility during optical observation. By superimposing the two comb-shaped electrode substrates 311 orthogonally, 1521 microgrids, each composed of four microband electrodes, are formed. Cells can be contained in each microgrid. By applying AC voltages with a phase difference of 90° to each of the four comb-shaped microband electrodes, a rotating electric field can be generated within each individual microgrid.

[0056] To suppress nonspecific adsorption of cells to the electrodes, the 3D grid electrodes were immersed for 2 hours in a 10 mg / mL serum albumin (BSA) solution dissolved in an electro-tachometer solution. The electro-tachometer solution was prepared by diluting cell culture medium (RPMI1640) 10-fold with a 300 mM mannitol aqueous solution. The conductivity of the electro-tachometer solution was 100 mS m. -1 That was the case.

[0057] Jurkat cells were used. Jurkat cells cultured in RPMI1640 medium were harvested, centrifuged (800 rpm, 5 minutes), and then suspended in electrorotating solution. The cell concentration was 1 × 10⁶. 6 The concentration was set to cells / mL.

[0058] After thoroughly washing the BSA solution in the device with an electro-rotational measurement solution, a cell suspension was introduced. Once a sufficient amount of cells had been introduced into the microgrid and the solution flow had stopped, an AC voltage was applied to four comb-shaped microband electrodes to induce a rotating electric field within each microgrid. The AC voltage used was a 2.4 Vpp, 400 kHz sine wave.

[0059] <Acquisition of sequential images for evaluation> To observe cells rotating within the device, the electrode device was placed under an inverted microscope (ECLIPSE Ts2R, Nikon). The objective lens was set to 40x magnification. Images of the cell rotation were recorded on a personal computer via a USB-CMOS camera (DMK33UX174, THE IMAGINGSOURCE). The image resolution was set to 1920 x 1200 pixels, and images were acquired at a frame rate of 60 frames per second (60fps). The cell rotation state was recorded for 5 seconds.

[0060] <Similarity calculation> From the acquired sequence of images, the analysis region was defined as a square shape to include cells rotating within a microgrid, and the region containing cells was extracted. For extraction, a threshold was determined using "Otsu's Law" from the image within the defined region, a binarized image was obtained, and the cell contours were extracted from the obtained binarized image. "OpenCV findContours" was used for contour extraction. The area inside the extracted contour was defined as the cell region, and an image was created by filling the interior. Using the image with the interior filled, only the cell region was extracted, and an image of the cell was created by filling the background with black. This operation was performed for all frames of the acquired sequence of images, and a set of extracted images displaying single cell regions was obtained.

[0061] Next, the similarity of the extracted images was calculated. The image in the selected frame fs was used as a template, and matching was performed with the extracted images in the positive direction from fs. "OpenCV matchTemplate" was used for this template matching. The similarity between the template image and the extracted images in each frame was evaluated using the normalized correlation function matching method, and the similarity was expressed in the range of -1 to 1 using "OpenCV TM_CCOEFF_NORMED".

[0062] <Calculate rotational speed> Figure 9 shows the results of template matching, calculating similarity, and displaying the similarity as a heatmap, with the selected frame fs as the first frame. The image with an interval D of 17 (18th frame) from the selected frame fs had the highest similarity to the image of the selected frame fs. Therefore, the time required for one cell rotation was 17 / 60 seconds. Rotation speed (radians s) -1 When converted to ), it becomes 22.2 radians -1 The results were as follows. Figure 10 shows the heatmap obtained by calculating the similarity for all frames of the acquired images. As the interval D with the selected frame increased, regions with a maximum similarity appeared at regular intervals. Furthermore, regardless of the position of the selected frame fs, the similarity reached a maximum at almost the same frame interval. This indicates that the cells being evaluated continued rotational motion at a nearly constant interval.

[0063] <Comparison with visual inspection> When the rotation speed was measured for cells 1-10 by changing the cells selected for evaluation, the results were 6.01, 8.59, 12.05, 15.11, 17.27, 19.03, 21.99, 24.06, and 29.34 radians, respectively. -1 That's what happened.

[0064] For images of cells 1-10, feature points on the cells were visually identified, and the rotation speed was calculated from the amount of movement of those feature points when the images were played frame by frame. The results were 5.78, 8.77, 12.29, 14.59, 17.13, 18.85, 19.84, 23.94, and 29.00 radians. -1 The correlation coefficient between the two was 0.996. Therefore, it was shown that the method for calculating rotational speed according to the present disclosure has a high correlation with the method for calculating rotational speed by visual inspection.

[0065] [Table 1]

[0066] (Evaluation of changes in rotational speed due to drug administration) <Evaluation device> Evaluation was performed using a well-type rotating electrode as shown in Figure 11 as the electrode tip. Sixteen lower microband electrodes made of ITO (electrode width 40 μm, electrode spacing 20 μm) were placed on a glass substrate. An insulating layer with a thickness of 20 μm was formed on the lower microband electrodes (i, ii) using SU-8, a negative-type thick-film photoresist, and 225 microwells with a width of 30 μm, a length of 20 μm, and a thickness of 20 μm were formed. The microwells were arranged so that two ITO microband electrodes were exposed at both ends of each microwell. Upper microband electrodes made of gold (A, B) were placed on the insulating layer. The upper microband electrodes were coated with an insulating film.

[0067] Lower microband electrodes were exposed up to a position 5 μm from each side of the microwell. Upper microband electrodes were positioned on the top and bottom edges of the microwell, with a width of 20 μm and a length of 5 μm. Four microband electrodes of the same electrode area were placed in each microwell, and a sine wave with a phase shift of 90° was applied to the four microband electrodes (i, ii, A, B) to induce a rotating electric field within the microwell. The AC voltage was supplied by an arbitrary waveform generator (Arb Studio 1104, TELEDYNE LECROY) that can independently output four types of sine signals.

[0068] A well-type rotating electrode was placed inside a solution chamber (outer diameter 10 mm, inner diameter 8 mm, height 6 mm, volume 300 μL) formed from polydimethylsiloxane (PDMS). After oxygen plasma treatment, the solution chamber and rotating electrode were filled with a 10 mg / mL BSA solution and allowed to stand for 2 hours. After rinsing the solution chamber with electro-rotating solution, a suspension of Jurkat cells was introduced into the solution chamber. After standing for about 10 minutes, the cells were allowed to settle by their own gravity and captured in the microwells.

[0069] Jurkat cells were cultured in RPMI1640 medium, with a cell concentration of 1 × 10⁶. 6 The cells were suspended in an electrorotating solution to a concentration of cells / mL. The electrorotating solution used was RPMI1640 medium diluted to 5% (v / v) with a 250 mM mannitol aqueous solution (conductivity 74 mS / m).

[0070] To observe cells rotating within the device, the microwell electric rotation device was placed under an inverted microscope (ECLIPSE Ts2R, Nikon). The objective lens was set to 20x magnification. Images of the cell rotation were recorded on a personal computer via a USB-CMOS camera (DMK33UX174, THE IMAGINGSOURCE). The image resolution was set to 1920 x 1200 pixels, and images were acquired at a rate of 30 frames per second (30fps). The state of cell rotation was imaged for 120 seconds.

[0071] After applying an AC signal to a macroband electrode to rotate the cells, 3 μL of ionomycin solution prepared to 100 μM was gently added to the solution chamber 10 seconds later, until the final ionomycin concentration reached 1 μM. Images were acquired from the start of cell rotation until 120 seconds later, and the time-dependent change in rotation speed of 15 cells that continued to rotate as single cells in a microwell for 120 seconds was obtained.

[0072] As an example, Figure 12 shows a heatmap of the similarity of a selected cell. The region of high similarity observed between D 12 and 26 (the white line region in the figure) represents the number of frames required for one rotation in each frame. By limiting the region of D to fmin=12 and fmax=26, we found the D at which the similarity is maximized in each selected frame. Based on this, we calculated the rotation speed Ωs for each frame.

[0073] Figure 13 shows the change in rotation speed over time, with the position of each selected frame fs converted to elapsed time from the start of rotation. The rotation speed remained almost constant for the first 10 seconds until ionomycin was administered, but gradually decreased, and at 40 seconds, the rotation speed dropped significantly. After that, the rotation speed increased slightly, but did not recover to the initial speed.

[0074] Similar analysis was performed on the remaining 14 cells, and the change in rotational speed for a total of 15 cells was determined. Since the absolute value of rotational speed is affected by the size of the cell, the average rotational speed from immediately after the start of image acquisition to 9.5 seconds was taken as the initial rotational speed Ω0, and the relative rotational speed Ω at each time point was divided by Ω0 to obtain Ω / Ω0. As shown in Figure 14, it was observed that the rotational speed decreased overall from 40 seconds onward, and then gradually recovered. On the other hand, there was a large variation in rotational speed, indicating that each cell had a different response to ionomycin.

[0075] Instead of adding a solution containing ionomycin, a similar evaluation was performed by adding an electrorotating solution without ionomycin. The number of cells analyzed was 19. When only the electrorotating solution was added, the rotation speed remained almost constant for 120 seconds. This indicates that, without adding a drug to the cells, the cells continue to rotate at a constant speed. [Industrial applicability]

[0076] The particle rotation speed evaluation method described herein allows for easy measurement of the rotation speed of minute particles, and for example, the effects of drugs on particles can be easily evaluated, making it useful in various industrial fields, including the pharmaceutical field. [Explanation of Symbols]

[0077] 102 Similarity calculation process 102A Analysis target extraction process 102B Calculation process 103 Rotational speed calculation process 201 cells 300 volume 301, 302, 303, 304 electrode 311 type electrode substrate 313 Microband electrodes 351 areas 406 Analyzed Image Group 500 Evaluation device 501 Rotation generation unit 502 Fluid delivery unit 503 Observation Department 504 Imaging device 505 Rotational speed measurement unit 506 Input device 507 Control Unit 511 Arbitrary Waveform Generator 512 Wiring Unit 513 Electrode Tip 513A, 513B, 513C, 513D Electric Rotary Tip 514 Observation Cells 550 Processing Unit 551 Parameter setting section 552 Records Department 553 Display section 556 Image Processing Unit 557 Similarity calculation unit 558 Rotational speed calculation unit

Claims

1. A similarity calculation step is performed in which, for a series of images capturing particles, one selected frame from among multiple frames is used as the frame to be analyzed, and the similarity between the particles captured in the frame to the particles captured in the other frames is calculated over time. A method for measuring the rotational speed of a particle, comprising: a rotational speed calculation step, which calculates the rotational speed of the particle by defining the time interval at which the frame in which the similarity is maximized appears as the time it takes for the particle to complete one rotation.

2. The method for measuring the rotational speed of a particle according to Claim 1, wherein the position of the first frame is shifted over time and selected as the frame to be analyzed, and the similarity calculation step and the rotational speed calculation step are performed to evaluate the change in the rotational speed of the particle over time.

3. A method for measuring rotational speed according to claim 2, wherein the particles are cells, and the effect of the drug on the cells is evaluated by evaluating the change in rotational speed of the cells before and after supplying the drug to the cells.

4. An observation cell capable of applying an electric field to cells, The imaging unit captures a series of images of cells within the observation cell, The system includes a rotation speed measurement unit that analyzes the aforementioned continuous images and calculates the rotation speed of the cells, The cell evaluation device includes a rotation speed measuring unit which calculates the similarity between the cell shown in one selected frame from a plurality of frames and the cell shown in other frames over time, and calculates the rotation speed of the cell by taking the time interval in which the frame with the maximum similarity appears as the time it takes for the cell to complete one rotation.

5. The cell evaluation apparatus according to claim 4, wherein the rotation speed measuring unit calculates the rotation speed of the cell over time.

6. The unit further comprises a drug supply unit that supplies drugs to cells in the observation cell, The cell evaluation apparatus according to claim 5, wherein the rotation speed measuring unit calculates the rotation speed of the cells over time before and after the supply of the drug.

Images