JP2024068067A - Magnetic Microscope - Google Patents

Abstract

To provide a magnetic microscope tha t can observe vitality of individual cells.SOLUTION: In a magnetic microscope 61, conventional GSR sensors are downsized, and precisely arranged on a magnetic sensor element grid substrate to fabricate a magnetic sensor grid, the grid is arranged under Petri dish 62 during cell observation, a magnetic field generated from a cell is measured, and distribution of current elements flowing in the cell is calculated, which can be displayed in a PC image as an image figure. Combining the magnetic microscope and an optical microscope 64 makes it possible to easily observe a mechanical motion and movement of a shape change of cells and the degree of activation of each cell in real-time.SELECTED DRAWING: Figure 8

Description

The present invention relates to a magnetic microscope that observes the vitality of an entire cell aggregate by measuring the magnetic field emanating from within the cell aggregate and the electric current fragments that generate the magnetic field, while at the same time observing the structure and movement of the cells using an optical microscope used in cell research.

When observing the growth of cells such as iPS cells, in addition to observing the morphological changes and movement of the cell body using an optical microscope, it is necessary to observe the vitality of the entire cell aggregate by observing the electric current fragments flowing within the cell aggregate.

However, the size of a cell is about 20 μm, and the size of the aggregate is less than 5 mm. By measuring the magnetic field emitted from the current flowing in such a minute cell, the vitality of the cell can be predicted from the strength and direction of the current element. However, there is currently no known magnetic microscope that can measure the current element flowing in such a minute cell.
Here, a magnetic microscope can be defined as a device that detects tiny magnetic fields emitted from tiny cells, amplifies the size of the magnetic fields by 100 to 1000 times in accordance with the shape of the cells, and observes the distribution of the magnetic field or the distribution of electric current fragments that are the source of the magnetic field (Figure 1).

Incidentally, a cell evaluation device using a magnetic sensor is disclosed in Patent Document 1. Figure 17 of the document describes that amorphous wires with a diameter of 20 μm are arranged in a checkerboard structure at intervals of 80 μm, and high-frequency alternating current is passed through both ends of each wire to measure the impedance of the amorphous wire from the voltage at both ends, thereby making it possible to measure the magnetic field Hx or Hy at the intersection of the wires.
However, the magnetic impedance sensor measures the average value of the magnetic field associated with the flowing wire when a voltage is applied to both ends of the wire and an AC current is passed through it, and it is theoretically clear that it cannot measure the magnetic field at the intersection point alone. In fact, no papers supporting this invention have been published, and no products are on the market.

The inventor appears to be mistaken in thinking that this type of sensor, which is a grid sensor in which an element is placed at the (i,j)th position of a grid and a switch is operated so that an AC current flows from the i-th wire to the j-th wire, driving only the (i,j)-th element, is widely known. Although this type of sensor is widely known, the inventor appears to be mistaken in thinking that this technology is not yet developed. In other words, a magnetic grid sensor in which magnetic sensors are placed in a checkerboard pattern with intervals of 2 mm or less has not yet been developed.

The magnetic field emitted from a cell is thought to be about 10 nT or less directly above the cell. Patent Document 1 explains that it is 1 nT or less, but the measurement position of the sensor is 900 μm away from the cell body, whereas the present invention assumes a distance of 300 μm or less, and since the generated magnetic field is inversely proportional to the square of the distance, the views of both documents can be said to be consistent.

Taking into account the size of the cells, the size of the magnetic sensor element required to detect this magnetic field should be approximately 10 μm to 500 μm. Currently known GSR sensors have a detection power of approximately 15 nT when the sensor is 500 μm long, and no small, highly sensitive magnetic sensor that meets the above requirements is known. Furthermore, the manufacturing technology for densely arranging a large number of these magnetic sensor elements in a grid on the X and Y axes has not yet been established.

There is a need to develop a small, highly sensitive magnetic sensor and a magnetic sensor grid using this sensor, in order to develop a magnetic microscope. However, a search using j-platpat did not reveal any prior art literature on magnetic microscopes.

Patent No. 5526384

The present invention measures the magnetic field emitted from a 20 μm-sized cell, and measures the vitality of the cell by observing the tiny magnetic field generated within the cell and the electric current fragment that is the source of the magnetic field.

The inventor came up with the idea that if he developed a small GSR sensor and arranged multiple elements along each row of grid lines in a checkerboard pattern along the X and Y axes, he could invent a magnetic microscope capable of observing cells about 20 μm in size.

Specifically, the idea was that if a magnetic sensor grid with numerous magnetic sensors arranged in a checkerboard pattern along the X and Y axes were attached to the underside of a petri dish being observed under an optical microscope, the micromagnetic field could be measured, and a current element calculated from the measured value, then it would be possible to measure the micromagnetic field, current element, and cell activity at the cellular level in real time. After examining the feasibility of this, the researchers came to the conclusion that the following six challenges lay ahead.

The technological development issues for this purpose are:
The first challenge is to develop a magnetic sensor with a detection power of about 1 pT to 10 nT and a size of 10 μm to 2 mm or less. Specifically, the goal is to reduce the size and improve the performance of current GSR sensors, but the detection power of a magnetic sensor is proportional to its size, and there is a trade-off between the two, making it difficult to improve both characteristics at the same time.

The second challenge is to develop technology to arrange 100 to 4 million magnetic sensor elements in a grid along the X and Y axes on an ASIC board. The number of elements should be selected based on the magnification of 40 to 1,000 times and the measurement area of about 10 mm to 50 mm in diameter.

The third task is to develop a program that uses a magnetic sensor grid to measure the magnetic field emitted by the current element, calculates contour maps of the current element distribution and magnetic field distribution flowing through the cell aggregate, and displays an image on a screen.

The fourth challenge is to develop a magnetic microscope with a magnification of 1,000, using a magnetic sensor grid arranged at an ultra-high density to achieve an image pixel size of about 20 μm and a magnetic field distribution resolution of about 1 μm.

For the first issue, a GSR sensor was used as the magnetic sensor. By setting the coil pitch to 0.1 μm to 3 μm and installing multiple magnetic wires as necessary, the number of coil turns can be set to 150 to 2,000, and the detection power can be increased to 1 pT to 50 pT, making it highly sensitive. This problem can be solved by making the GSR sensor small and highly sensitive. The GSR sensor element is shown in Figure 2.

The GSR sensor is described in detail in Patent Publication No. 5839527 by the present inventor, and is cited in the present invention. The GSR sensor is an ultra-sensitive micromagnetic sensor that includes a GSR element that is composed of a conductive magnetic field detection magnetic wire on a substrate, a detection coil formed of a circular coil wound around the magnetic wire, two electrodes for passing current through the magnetic wire, and wiring that connects two electrodes for detecting the coil voltage, a means for passing a pulse current with a GHz frequency through the magnetic wire, and an electronic circuit that detects the coil voltage generated when the pulse current is passed through the magnetic wire and converts the coil voltage into an external magnetic field H.

The second problem was solved by forming multiple elements on an ASIC board, attaching magnetic wires in a grid pattern along the X and Y axes, and then baking coil wiring and electrode wiring onto the wires. The ASIC and multiple elements were connected using via holes. Multiple ASICs may be used to correspond to multiple elements, or multiple ASICs may be used to control multiple elements with channel switching functions. The magnetic sensor element grid is shown in Figures 3 to 5.
FIG. 3 is a top view of the magnetic sensor element grid, FIG. 4 (FIGS. 4A, 4B, and 4C) is a plan view of the structure of a unit element of the magnetic sensor element grid, and FIG. 5 (FIGS. 5A, 5B, and 5C) is a cross-sectional view of the magnetic sensor element grid.

The magnetic field measurements were taken at the numerous intersections of the X-axis and Y-axis elements, Hz in Figure 4A (a-1), Hx in Figure 4A (a-2), and Hy in Figure 4A (a-3), and these values were used as the magnetic field of the grid to calculate the magnetic field distribution and current element distribution. In other words, they were calculated as Hz = (Hz1 + Hz2 + Hz3 + Hz4)/4, Hx = (Hx1 + Hx2 + Hx3 + Hx4)/4, and Hy = (Hy1 + Hy2 + Hy3 + Hy4)/.

The top surface of the grid element substrate is flattened, and the flat surface is brought into direct contact with the substrate surface on the cell fluid side of the cell observation dish, making the distance from the cell to the element 300 μm or less, facilitating the measurement of the micromagnetic field emitted from the cell. The shorter the distance from the cell to the element, the better.

For the third problem, before observation, the magnetic field at the sensor grid position (i, j) is measured to obtain the initial magnetic field measurement value mH _ij (→) (b), and this value is displayed as a magnetic field distribution corresponding to the grid size on the X-axis and Y-axis plane.

Next, after placing a cell in the petri dish, the magnetic field at the sensor grid position (i, j) is measured to obtain the magnetic field measurement value mH _ij (→)(a), and mH _ij (→)(b) is subtracted from the magnetic field measurement value to obtain the measurement value. In other words, mH _ij (→)=mH _ij (→)(a)-mH _ij (→)(b).

This value is considered to be the magnetic field emanating from the current element Ids within the cell aggregate, and measured by a magnetic sensor at the (i, j)th position of the magnetic sensor element grid, the measured value is taken as mH _ij (→), a contour map of the absolute value of mH _ij is created, and the position P(x, y, z) of the current element Ids from the peak of mH _ij (→) and the length of the current element Ids (→) are assumed from the spread of the mountain of the peak of mH _ij (→), and it is assumed that the current element Ids exists within the cell aggregate. If there are k peak locations, a calculation model is created assuming that there are k current elements Ids.

Next, the kth current element I_kds_kPosition P of (→)_k(x_k,y_k,z_k), and the magnetic field measurement position Gij(x_1j,y_1j, 0), and the current element position P_k(x_k,y_k,z_k) and the measurement position Gij(x_1j,y_1j, 0) is Rijk(→),
The kth current element I_kds_kThe magnetic field that (→) creates at the measurement position G(i,j) is H_ijk(→) = 1/4πRijk³×I_kds_kSince the magnetic fields can be calculated from the basic equation of (→) × Rijk(→), the theoretical value tH created at the measurement position G(i, j) of the magnetic sensor grid is calculated by adding up the magnetic fields._i.j.(→) is tH_i.j.(→) = ΣtH_ijkLet (→). (Σ adds k from 1 to n.)

Let the error between the two be e _ij (→) = mH _ij (→) - tH _ij (→), and create the error function E _ij = Σ(e _ij ) ² .
Regarding the direction of _Ik (→), the inclination angle with respect to the XY-axis plane is _φk and the angle with respect to the X-axis is _θk . From this error function, partial differentiation of the error function is performed using the Gauss-Newton method to derive 7k simultaneous equations, and 7k simultaneous equations consisting of the absolute value of _Ik , _dsk , _θk , _φk , _Xk , _Yk , and _Zk are obtained. 7k unknowns are found from these equations, and a large number of current elements flowing through the cell aggregate are calculated. The results are used to create a distribution map of the current elements and displayed on the screen.

Here, we have decided to affix (→) to the vector representation of each vector physical quantity.
FIG. 6 shows a magnetic field distribution map obtained from magnetic field measurements taken with a magnetic microscope, and FIG. 7 shows a flow chart of a calculation program for obtaining a current element.
It should be noted that the present invention is directed to the configuration of a magnetic microscope, and the method of calculating the current element vector from the error function is not limited to the above method.

For the fourth issue, in the unit element of the sensor grid, the coil pitch is set to 0.1 μm, the element length is 10 μm, the number of magnetic wires is set to two, and the number of coil windings is set to 180, resulting in a pixel of the sensor grid of 10 μm x 10 μm. The detection power of the magnetic sensor is ensured to be 5 pT. If a magnetic field distribution map is created from these measured values, the pixel of the magnetic field distribution map will be approximately 1 μm, and a magnetic microscope with a magnification of 1000 times can be obtained.

Next, we superimposed the current distribution measured by the magnetic microscope on the shape of the cells observed by the optical microscope and displayed the image on a screen, which makes it easy to observe both the mechanical movement and shape change of the cells, as well as the activation level of each cell, in real time.
FIG. 8 shows an image of the combined optical microscope and magnetic microscope, and FIG. 9 shows the results of each observation (the magnetic microscope is an image).

In order to make the spatial resolution of the magnetic field equal to the resolution of an optical microscope, the magnetic field at the intermediate position between the discretely measured magnetic field values is calculated by functionally approximating the magnetic field distribution using an interpolation method. This interpolation method can obtain a spatial resolution of the magnetic field of about 1/20 of the grid interval.
It was confirmed that the positional accuracy of the current element was theoretically about 1/20 of the grid interval, based on a program for calculating the position of the current element.

By combining these four solutions, we were able to invent the magnetic microscope.
It should be noted that the present invention is directed to a small, highly sensitive magnetic sensor and is not limited to a GSR sensor, as is apparent from the configuration.

This invention makes it possible to observe the degree of cell vitality when observing cells. Moreover, by integrating it with an optical microscope, it becomes possible to simultaneously measure the shape, movement, and vitality of individual cells in real time, and it is expected to become a basic tool in cell research.

1 is an image diagram of a magnetic microscope, in which (a) is a perspective view and (b) is a side view. FIG. 2 is a diagram showing a GSR sensor element. FIG. 2 is a plan view of a magnetic sensor element grid. FIG. 2 is a plan view of a grid unit of a magnetic sensor element, showing a one-axis element consisting of a Z-axis element, an X-axis element, or a Y-axis element. FIG. 2 is a plan view of a grid unit of a magnetic sensor element, showing a two-axis element consisting of an X-axis element and a Y-axis element. FIG. 2 is a plan view of a grid unit of a magnetic sensor element, showing a three-axis element consisting of an X-axis element and a Y-axis element arranged on an element base. 1A and 1B are cross-sectional views of the z-axis type, x-axis type, and y-axis type taken along the A1-A2 line in a one-axis element grid. 1 is a cross-sectional view of a two-element grid taken along line A1-A2. 1 is a cross-sectional view of a three-element grid taken along line A1-A2. FIG. 1 is a magnetic field distribution diagram obtained from magnetic field measurements taken with a magnetic microscope. FIG. 13 is a flowchart showing a calculation program for determining a current element. This is an image diagram of an optical microscope and a magnetic microscope combined. FIG. 1 is an image of the observation results obtained using a magnetic microscope.

A first embodiment of the present invention is as follows.
A magnetic sensor element grid is provided on a substrate surface on the cell fluid side of a petri dish for cell observation;
a signal processing circuit that detects a grid voltage corresponding to a minute magnetic field generated by a current element flowing inside a cell using a magnetic sensor element grid and converts the grid voltage into a grid magnetic signal;
The magnetic sensor element grid is a grid of magnetic sensor elements arranged at the intersections of a checkerboard pattern along the X-axis and Y-axis on a sensor grid substrate, or more precisely, at four locations on either side of each intersection, to measure the magnetic field at the intersections.
and a display device that displays the value of the grid magnetic field as a magnetic field contour map.

In addition, the magnetic sensor element of the magnetic microscope is
It is characterized by comprising a single-axis element for measuring one of the magnetic fields, Hz magnetic field, Hx magnetic field, and Hy magnetic field.

In addition, the magnetic sensor element of the magnetic microscope is
It is characterized by comprising a two-axis element for measuring the Hx magnetic field and the Hy magnetic field.

In addition, the magnetic sensor element of the magnetic microscope is
It is characterized by comprising a three-axis element for measuring the Hx magnetic field, the Hy magnetic field and the Hz magnetic field.

In addition, the magnetic microscope
The present invention is characterized by comprising a program for calculating the current element flowing within a cell from the value of the grid magnetic field, and a device for displaying that value as an image on a screen.

In addition, the magnetic microscope
Before observation, the magnetic field at the sensor grid position (i, j) is measured to obtain an initial magnetic field measurement value mH _ij (→) (b), which is then displayed as a magnetic field distribution corresponding to the grid size on the X-axis and Y-axis plane.

Next, after placing a cell in the petri dish, the magnetic field at the sensor grid position (i, j) is measured to obtain the magnetic field measurement value mH _ij (→)(a), from which mH _ij (→)(b) is subtracted to obtain the measurement value mH _ij (→). In other words, the measurement value mH _ij (→)=mH _ij (→)(a)-mH _ij (→)(b).

This measured value mH _ij (→) is considered to be the magnetic field emitted from the current element in the cell aggregate, and is measured by the magnetic sensor at the (i, j)th position of the magnetic sensor element grid. The measured value is taken as mH _ij (→). A contour map of the absolute value of mH _ij is then created.
A computational model is created in which the position P(x, y, z) of the current fragment Ids is assumed from the peak of mH _ij (→), and the length ds of the current fragment Ids is assumed from the spread of the peak of mH _ij (→), and k current fragments Ids (k is 1 to n) are present within the cell population.

Next, let _Pk ( _xk , _yk , _zk ) be the position of the _k -th current element _Ikdsk , _Gij ( _xij , _yij ,0) be the measurement position of the magnetic field, and Rijk be the distance between the current _element position _Pk ( _xk , _yk , _zk ) and the magnetic field measurement position _Gij ( _xij , _yij ,0), then
The magnetic field generated by the k _{-th current element Ikdsk can} be obtained from the equation _Hijk (→) ₌ 1/ _4πRijk3 × ^Ikdsk ₍ →)× _Rijk (→), so by adding up magnetic fields 1 to n, we obtain a theoretical value _tHij (→) generated at the measurement position _Gij (i,j) of the magnetic sensor grid,
The error between the two is e _ij (→) = mH _ij (→) - tH _ij (→),
Create the error function E _ij =Σ(e _ij ) ² .

Regarding the direction of I _k (→), the inclination angle with respect to the XY-axis plane is φ _k , and the angle with respect to the X-axis is θ _k .
From this error function, 7k simultaneous equations related to _Ik , _dsk , _θk , _φk , _Xk , _Yk , and _Zk are obtained, and unknowns are obtained from the equations.
The system calculates k current elements flowing through a cell aggregate, and uses the results to create a distribution map of the current elements, which is then displayed on the screen.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an embodiment of the present invention will be described in detail with reference to FIGS.
<Magnetic microscope>
The positional relationship between the magnetic sensor element grid that constitutes the magnetic microscope and the cells (cell aggregates) of the specimen is shown in Fig. 1. The magnetic microscope is also composed of a signal processing circuit that converts the magnetic field signals detected by the magnetic sensor elements, a display device, etc.
The magnetic sensor element grid 10 is made up of a large number of magnetic sensor elements 101 .
Cells (cell aggregates) 12 are placed on a petri dish (for cell observation) 11, and a magnetic sensor element grid 10 is placed thereon. Alternatively, the petri dish 11 may be placed on the magnetic sensor element grid 10 as shown in FIG.
The positional relationship between the two can be arbitrarily selected depending on the size and volume of the cell (cell aggregate) to be observed, the detection capability of the magnetic sensor element grid, and further on the combination with an optical microscope.

<Magnetic sensor>
A GSR sensor is used as the magnetic sensor. The basic structure of the GSR element is shown in Fig. 2, and the structure of the magnetic sensor element grid is shown and explained in Figs.
The diameter of the magnetic wire 22 is 1 μm to 10 μm. The coil pitch of the detection coil 23 is 0.1 μm to 3 μm, and the coil width is 3 μm to 30 μm. The number of magnetic wires is 1 to 8, as required.
It should be noted that the magnetic sensor constituting the magnetic microscope is not limited to the GSR sensor, so long as it can exhibit the characteristics required for a magnetic microscope in terms of size and performance.

The size of the unit elements (magnetic sensor elements) 32 of the magnetic sensor element grid 3 is 10 μm to 1 mm.
When the size of the unit element 32 is 10 μm, it is preferable that the coil pitch is 0.1 μm, the coil width is 3 μm, the number of magnetic wires is two, and the number of coil turns is 150.
On the other hand, when the size of the unit element 32 is 1 mm, it is preferable that the coil pitch is 3 μm, the coil width is 30 μm, the number of magnetic wires is 1 to 8, and the number of coil turns is 2400.

As shown in FIG. 2, the GSR sensor is an ultra-sensitive micro-magnetic sensor that is made up of a GSR element 2 consisting of a conductive magnetic wire 22 for detecting a magnetic field on a substrate 21, a detection coil 23 formed by a circular coil wound around the magnetic wire, and wiring 26, 29 connecting two electrodes 24 for passing current through the wire and two electrodes 28 for detecting the coil voltage, a means for passing a pulse current with a GHz frequency through the magnetic wire, and an electronic circuit that detects the coil voltage generated when the pulse current is passed and converts the coil voltage into an external magnetic field H.

<Magnetic sensor element grid (grid)>
As shown in FIG. 3, the magnetic sensor element grid 3 includes a number of unit elements (magnetic sensor elements) 32 arranged on a grid substrate 31 .
The size of the magnetic sensor element grid 3 is 5 mm square to 20 mm square, or 5 mm diameter to 20 mm diameter. The size of the unit element 32 is 10 μm to 1 mm. The number of unit elements 32 is a minimum of 25 pixels (5×5) to a maximum of 4 million pixels (2000×2000).

As shown in FIG. 4, the unit elements 32 are of three types: a one-axis element type (FIG. 4A), a two-axis element type (FIG. 4B), and a three-axis element type (FIG. 4C).
5 (5A, 5B, and 5C) show cross-sectional views of the three types taken along line A1-A2 in Fig. 3. The cross-sectional views show a cross-section of a magnetic sensor element grid placed on a petri dish.

First, the type of one-axis element (FIG. 4A) is a type in which the GSR sensor element 321 is arranged symmetrically on the grid substrate 31 with respect to the intersection 321o of the X-axis and Y-axis.
(a-1) A one-axis element type in which the Z-axis elements 321z are arranged one by one in the Z-axis direction;
(a-2) A one-axis element type in which the X-axis elements 321x are arranged one by one in the X-axis direction;
(a-3) There is a one-axis element type in which the Y-axis elements 321y are arranged one on each side in the Y-axis direction.

These cross-sectional views (FIG. 5A) are
(a-1) One-axis element grid - z-axis type (4(41)) is made up of a GSR element 411, an ASIC 412, an electrode 413 and a coating material 415 for protecting the grid wiring 414, and is placed on a petri dish 40.
(a-2) One-axis element grid - x-axis type (4 (42)) is made of a GSR element 421, an ASIC 422, an electrode 423 and a coating material 425 that protects the grid wiring 424, and is placed on a petri dish 40.
(a-3) One-axis element grid - y-axis type (4 (43)) is made up of a GSR element 431, an ASIC 432, an electrode 433 and a coating material 435 for protecting a grid wiring 434, and is placed on a petri dish 40.

Next, the two-axis element type (FIG. 4B) is a type in which the GSR sensor element 321 is attached to the grid substrate 31 along the intersection 321o of the X-axis and the Y-axis.
(b) There is a two-axis element type in which two X-axis elements 321x in the horizontal direction and two Y-axis elements 321y in the vertical direction are arranged symmetrically with respect to the origin 321o.

The cross-sectional view of this biaxial element type (4 (44)) is shown in FIG. 5B. It consists of a GSR element 441, an ASIC 442, and a coating 445 that protects electrodes 443 and grid wiring 444, and is placed on a petri dish 40.

Finally, the type of three-axis element (FIG. 4C) is a three-axis type in which a magnetic field vector sensor is arranged on the grid substrate 31 along the intersection 321o of the X-axis and Y-axis.
This three-axis element type has four GSR elements or on-ASIC type GSR sensors arranged in four-fold symmetry and mirror symmetry on the inclined surface 331 of a base 330 consisting of a square pyramid, an octagonal pyramid, or an irregular octahedral pyramid, so that the inclination direction coincides with the magnetic wire 321w of the GSR element.
The magnetic fields in the X-axis and Y-axis directions are measured from two X-axis elements 321x and two Y-axis elements 321y, and the magnetic field in the Z-axis direction is obtained by calculation.
The on-ASIC type GSR sensor was invented by the inventors and is disclosed in a patent publication (Japanese Patent No. 7062216). Please refer to the description in this patent publication for details. The magnetic field vector sensor was invented by the inventors and is disclosed in a patent publication (Japanese Patent No. 7215702). Please refer to the description in this patent publication for details.

The cross-sectional view of this triaxial element type (4(45)) is shown in FIG. 5C, and it is made up of a GSR element 451, an ASIC 452, electrodes 453, and a coating material 455 that protects grid wiring 454, and is placed on a petri dish 40.
The magnetic field in the Z-axis direction is obtained by calculation, and element 451z is a calculated element.

<Positional relationship between magnetic sensor element grid and cells>
It is preferable that the position of the magnetic sensor element grid 10 is such that the thickness of the petri dish 11 is 200 μm or less, the top surface of the magnetic sensor element grid 10 is flat, the flat surface is directly pressed against the back side of the petri dish, and the distance from the cells to the measurement parts of the magnetic sensor elements 101 is 300 μm or less. Since the magnetic field emitted by the cells is inversely proportional to the square of the distance, it is necessary to make this distance as small as possible.

<Measurement of magnetic field components at intersections of unit elements>
(A) Measurement of magnetic field components in a one-axis element (a-1) From the four pieces of data Hz1, Hz2, Hz3, and Hz4 measured using four unit elements 321z centered on the intersection 321o, the magnetic field component Hz at the intersection can be calculated from Hz = (Hz1 + Hz2 + Hz3 + Hz4)/4.
(a-2) From the four pieces of data Hx1, Hx2, Hx3, and Hx4 measured by the four unit elements 321x centered on the intersection 321o, the magnetic field component Hx of the intersection can be calculated from Hx = (Hx1 + Hx2 + Hx3 + Hx4)/4.
(a-3) From the four data Hy1, Hy2, Hy3, and Hy4 measured by the four unit elements 321y centered on the intersection 321o, Hy, which is the magnetic field component of the intersection, can be calculated from Hy=(Hy1+Hy2+Hy3+Hy4)/4.

(B) Measurement of magnetic field components in a two-axis element (b) From the four pieces of data Hx1, Hx2, Hy1, and Hy2 measured with two unit elements 321x oriented in the X-axis direction and two unit elements 321y oriented in the Y-axis direction centered on the intersection 321o, the magnetic field components Hx and Hy at the intersection can be calculated from Hx = (Hx1 + Hx2)/2, Hy = (Hy1 + Hy2)/2.

(C) Measurement of magnetic field components in a three-axis element (c) In a magnetic field vector sensor, the tilt angle of the base is θ, and from the four data Hx1, Hx2, Hy1, and Hy2 measured with two unit elements 321x oriented in the X-axis direction and two unit elements 321y oriented in the Y-axis direction centered on the intersection 321o, the magnetic field components Hx and Hy at the intersection 321o can be calculated from Hx = (1/2 cos θ) (Hx1-Hx2) and Hy = (1/2 sin θ) (Hy1-Hy2). The Hz in the Z-axis direction can be calculated from Hz = (1/4 sin θ) (Hx1+Hx2+Hy1+Hy2).

In addition, if you first measure the environmental magnetic field before taking measurements, and then subtract the environmental magnetic field from the measured value when placing the cells in the petri dish for observation, there will be no effect from the external magnetic field.

The magnetic sensor element grid also has an electronic circuit that expresses the position of the intersection of the unit element 32 as the i-th position in the X-axis direction and the j-th position in the Y-axis direction as P _ij , expresses the measurement value at P _ij as H _ij , and transfers H _ij to the current element distribution calculation program.
In addition, the position on the grid substrate of each element (elements in the X-axis, Y-axis, and Z-axis directions) that constitutes the unit element 32 may be designated as (i, j), and the measurement values of each element may be directly used as grid magnetic field measurement values to calculate the magnetic field distribution and current element distribution.

<Program to calculate current element distribution>
The program for calculating the current element distribution measures the magnetic field emanating from the current element in the cell aggregate with a magnetic sensor at the (i, j)th position of the magnetic sensor element grid, and defines the measured value as mH _ij .

In practice, since the measurement is affected by the environmental magnetic field, the magnetic field at the magnetic sensor element grid position (i, j) is measured before observation to obtain the initial magnetic field measurement value mH _ij (→) (b), and this value is displayed as the magnetic field distribution corresponding to the grid size on the X-axis and Y-axis plane.

Next, after placing a cell in a petri dish, the magnetic field at the magnetic sensor element grid position (i, j) is measured to obtain the magnetic field measurement value mH _ij (→)(a), from which mH _ij (→)(b) is subtracted to obtain the measurement value.
That is, mH _ij (→)=mH _ij (→)(a)-mH _ij (→)(b).
This value is considered to be the magnetic field emanating from the current element within the cell aggregate, and is measured by a magnetic sensor at the (i, j)th position of the magnetic sensor element grid, and the measured value is taken as mH _ij (→).

A contour map of the absolute value of mH _ij is created, k peak positions are identified, and the position P _k (x _k , y _k , z _k ) of the current fragment from the peak of each absolute value of mH _ij and the length of the current fragment ds _k (→) are assumed from the spread of the peak, and it is assumed that the current fragment I _k ds _k (→) exists at the position P _k (x _k , y _k , z _k ) within the cell aggregate. When there are k peak locations (one or more), a calculation model is created assuming that there are k current fragments.

Next, let _Pk ( _xk , _yk , _zk ) be the position of the _k - _th current element Ikdsk, _Gij ( _xij , _yij ,o) be the measurement position of the magnetic field, and Rijk be the distance between the current _element position _Pk ( _xk , _yk , _zk ) and the measurement position _Gij ( _xij , _yij ,o), then
The _{magnetic field generated by the kth current element Ikdsk can} be obtained from the equation _Hijk (→) = ₁ / _4πRijk3 × ^Ikdsk ₍ →) × _Rijk (→), so by adding these magnetic fields together, the theoretical value _tHij (→) = _ΣtHijk (→) generated at the (i, j)th position of the magnetic sensor element grid is obtained.

The error between the two is e _ij =mH _ij (→)-tH _ij (→), and the error function E=Σ(e _ij ) ² is created.
Regarding the direction of _Ik , the inclination angle with respect to the XY-axis plane is _φk and the angle with respect to the X-axis is _θk . From this error function, the error function is partially differentiated using the Gauss-Newton method to derive 7n simultaneous equations, and 7n simultaneous equations consisting of _Ik , _dsk , _θk , _φk , _Xk , _Yk , and _Zk are obtained. 7k unknowns are obtained from these equations, and a large number of current elements flowing through the cell aggregate are calculated. The results are used to create a distribution map of the current elements, which is then displayed on the screen as a magnetic field distribution map.

The program procedure is as follows (Figure 7). Note that vectors are displayed as (→).
First step (101);
Before observation, the magnetic field at the sensor grid position (i, j) is measured to obtain an initial magnetic field measurement mH _ij (→)(b).
However, the position of each element is specified in advance using a sensor grid coordinate system O-XYZ.

Second step (102);
After placing the cells in the dish, measure the magnetic field at the sensor grid position (i, j) to obtain the magnetic field measurement value mH_i.j.(→) Find (a), and from there find mH_i.j.(→) Subtract (b) to get the measured value mH_i.j.(→), calculate the magnetic field distribution, mH_i.j.(→) If there are k peaks in the absolute value distribution, the peak positions P_k(X_k, Y_k, Z_k) to current element I_kds_kAssume that exists. The length of ds is determined from the peak position and the spread of the mountain.

FIG. 6 shows an example of a magnetic field distribution diagram.
The cell assembly 4 is composed of five cells (=k), namely, cell 41 to cell 45, and the contour lines 411 to ₄₅₁ of the respective cells are illustrated to show a contour map. At the peaks of the respective contour maps, there are current elements _{I1ds1 to I5ds5} , and the positions thereof are positions _P1 to _P5 .
This is transferred to the current element distribution calculation program.

Third step (103):
Assuming that a current element _Ikdsk exists at peak position _{Pk ,} the theoretical magnetic field strength at position _Gij ( _Xij , _Yij ) of each magnetic sensor element grid created by k current elements Ids at position _Pk can be calculated by the formula _tHijk (→)=1 _{/4πRijk3×Ikdsk(→)×Rijk (} ^→ _{) for} the magnetic field emitted by the kth current element. Since there are 1 to n current elements, the theoretical magnetic field strength at position _Gij ( _Xij , _Yij ) is calculated by the sum of _Hij =Σ1/ _4πRijk3 × ^Ikdsk × _Rijk , _{which is} n.

A fourth step (104):
If the distance vector R _ijk (→) between the k-th current element position I _k ds _k and the magnetic field measurement position G _ij (X _ij , Y _ij ) is φ k, the inclination of the current element with respect to the Z axis is φ _k , and the inclination with respect to the X axis is θ _k , then
tH _ijk is a function of the current intensity I _k , the length ds _k of the current element, the positions X _ijk , Y _ijk , and Z _ijk , and the azimuth angles θ _k and φ _k .

Fifth step (105);
Calculate the measurement error.
_eijk (→) = _mHijk (→) - _tHijk (→)

A sixth step (106);
Calculate the sum of squares of the errors.
_Eij ⁼ _Σeijk2

Seventh step (107);
The current intensity _Ik , length _dsk to the current element, Xk _{, Yk} , _Zk, and azimuth angles _θk , _φk that minimize the sum of squares of the error are calculated using the Gauss-Newton method.

Eighth step (108);
A current distribution is calculated from the _{magnitude Ikdsk of} each current element and the values of the positions _Xk , _Yk , and _Zk , and the calculated current distribution is displayed as an image on a PC screen. Alternatively, a current distribution is calculated from the current intensity _Ik of each current element and the values of the positions _Xk , _Yk , and _Zk , and the calculated current distribution is displayed as an image on a PC screen.

The center position of the current segment has a positional accuracy of less than 1/10 of the pixel size, and the current intensity can be calculated with an accuracy of less than ±10%.

In addition, it is constrained by the distribution of each current element and the law of conservation of current, so we check whether there is any contradiction with the law of conservation of current.

As shown in Figures 8 and 9, in an embodiment of the present invention, a magnetic sensor element grid 50 is placed below the substrate surface of a petri dish 51, and an optical microscope is placed above, allowing the movement of cells 52 to be observed optically and simultaneously observed magnetically.

The current element distribution measured by the magnetic microscope is superimposed on a shape diagram of the cell 52 observed and photographed by an optical microscope, and the image is displayed on a screen.
This makes it possible to easily observe both the mechanical movement and shape change of the cells (FIG. 9(a)) and the activation level 71 of each cell (FIG. 9(b)) in real time.

[Example 1]
In the first embodiment of the present invention, the design elements are as follows:
A GSR sensor is used as the magnetic sensor. The structure of the GSR element is shown in Figure 2.
The length of the magnetic wire 22 was 0.95 mm, the wire diameter was 10 μm, the coil pitch was 3 μm, the coil width was 30 μm, and the number of magnetic wires was 4. The number of coil turns was 1,200. The size of the unit element of the sensor grid was 1 mm. As a result, the magnetic detection force of the unit element was 1 pT.

The GSR sensor used here is an ultra-sensitive micro-magnetic sensor consisting of a GSR element, as shown in Figure 2, which is composed of a conductive magnetic wire for detecting a magnetic field on a substrate, a detection coil formed by a circular coil wound around the magnetic wire, two electrodes for passing current through the wire, and wiring connecting two electrodes for detecting the coil voltage, and a means for passing a pulse current having a frequency of 1.2 GHz through the magnetic wire, and an electronic circuit that detects the coil voltage generated when the pulse current is passed and converts the coil voltage into an external magnetic field H.

The magnetic sensor element grid is an arrangement of many unit elements as shown in Figure 3. The unit elements are made up of two-axis elements, and as shown in Figure 4B (b), two GSR sensor elements are arranged on the grid substrate along the X-axis and two along the Y-axis, making it possible to measure the magnetic field (Hx, Hy) at the intersection of the X-axis element and the Y-axis element. The size of the grid sensor is 20 mm square. The number of elements in the grid is 361 pixels.

The grid substrate is made of a flat insulating material or an ASIC substrate coated with an insulating film. The grid substrate consists of a top surface for arranging elements, an electronic circuit surface for signal processing, and a strong material to ensure mechanical strength. The GSR elements are placed on the ASIC substrate surface, which consists of electronic circuits and strong materials, and the two are connected with via holes.

The position of the grid substrate was set so that the thickness of the dish was 160 μm, the top surface of the magnetic sensor element grid was flattened, and the flat surface was pressed directly against the back side of the dish, so that the distance from the cell to the measurement part of the magnetic sensor element at the cell liquid surface was 240 μm.

From the four data points Hx1, Hx2, Hy1, and Hy2 measured by the unit element, the magnetic field components Hx and Hy at the intersection were calculated from Hx = (Hx1 + Hx2)/2 and Hy = (Hy1 + Hy2)/2. Before measurement, the environmental magnetic field was first measured, and when the cells were placed in the dish and observed, the environmental magnetic field was subtracted from the measured value to avoid the influence of external magnetic fields.

The sensor grid represents the intersection of the sensor elements at the i-th position in the X-axis direction and the j-th position in the Y-axis direction as P _ij , and the measured value at P _ij is H _ij (→) to obtain the magnetic field measurement distribution. This measured value H _ij (→) was transferred to the current element distribution calculation program via an electronic circuit.
Furthermore, for the magnetic field between the grids, an approximation curve was created by the interpolation method, and a magnetic field spatial distribution ability of about 1/20 of the grid interval was obtained.

The program that calculates the current element distribution measures the magnetic field emitted from the current element in the cell aggregate using a magnetic sensor at position (i, J) of the magnetic sensor grid, and converts the measured value into mH_i.j.(→), and the theoretical value tH generated by the kth current element is_i.j.(→) The magnetic field created by each current element is H_ijk(→) = 1/4πR_ijk ³×I_kds_k(→) × R_ijk(→), n additions were made. The error between the two is e_i.j.(→) = mH_i.j.(→) -tH_i.j.(→), and the error function E = Σ(e_i.j.)²The direction of I is defined as the inclination angle φ with respect to the XY-axis plane, and the angle with respect to the X-axis is defined as θ.

When there are six current segments, by partially differentiating the error function using the Gauss-Newton method, 42 simultaneous equations can be derived and then the solution can be found.

The program procedure is as follows (Figure 7). Note that vectors are displayed as (→).
The first step (101) is to measure the magnetic field at a sensor grid position (i,j) before observation to obtain an initial magnetic field measurement mH _ij (→)(b).
However, the position of each element is specified in advance using a sensor grid coordinate system O-XYZ.

The second step (102) is to place the cells in the petri dish, measure the magnetic field at the magnetic sensor element grid position (i, j), and obtain the magnetic field measurement value mH_i.j.(→) Find (a), and from there find mH_i.j.(→) Subtract (b) to get the magnetic field measurement value mH_i.j.(→), calculate the magnetic field distribution, mH_i.j.(→) If there are k peaks in the absolute value distribution, the peak positions P_k(x_k, y_k, z_k) to current element I_kds_kAssume that exists. The length of ds is determined from the spread of the peak, centered on the peak position. Transfer this to the current element distribution calculation program.

In the third step (103), assuming that a current element Ids _k exists at the peak position P _k , the theoretical magnetic field strength at position P(i,j) of each magnetic sensor element grid generated by the k current elements Ids at position _{P k} can be calculated by the formula tH _ij = 1/4πR _ij ³ × Ids × R _ij . Since there are k current elements, the theoretical magnetic field strength at position P(i,j) is calculated by the sum of H _ij = Σ 1/4πR _ijk ³ × Ids _k × R _ijk , k pieces.

In the fourth step (104), when the distance vector R _ijk (→) between the kth current element position I _k ds _k and the magnetic field measurement position G _k is φ k, the inclination of the current element with respect to the Z axis is φ _k , and the inclination with respect to the X axis is θ _k , the theoretical magnetic field strength tH _ij is a function of the current strength I _k , the length ds _k of the current element, the positions X _ijk , Y _ijk , Z _ijk , and the azimuth angles θ _k and φ _k .

The fifth step (105) is to calculate the measurement error.
_eijk (→) = _mHijk (→) - _tHijk (→)

The sixth step (106) is to find the sum of the squares of the errors.
_Eij ⁼ _Σeijk2

The seventh step (107) is to calculate _Ik , _dsk , Xk _{, Yk} , _Zk , and azimuth angles _θk and _φk that minimize the sum of squares of the error by the Gauss-Newton method.

The eighth step (108) is to calculate the current element vector distribution or the current intensity of the current element from the magnitude _Ik and _dsk of each current element, the positions _Xk , _Yk , _Zk , and the azimuth angles _θk and _φk , and display it as an image on the PC screen.
The center position, intensity, and direction of the current segment have a positional accuracy of 1/10 of the pixel size or less, and the current intensity and direction can be calculated with an accuracy of ±10% or less.
Furthermore, the distribution of each current element was consistent with the law of conservation of current.

The size of the sensor grid was 20 mm square, and the number of elements in the grid was 400 pixels.
The magnetic field spatial resolution of the magnetic microscope and the positional accuracy of the center position of the current element were 50 μm, which corresponds to the accuracy of a 20x optical microscope. The intensity and direction of the current were estimated to be within ±10% accuracy as a result of simulation calculations using a current element calculation program.

[Example 2]
In Example 1, the GSR sensor had a magnetic wire length of 10 μm, a wire diameter of 8 μm, a coil pitch of 0.1 μm, a coil width of 20 μm, and a number of wires of 2. The size of the unit element of the sensor grid was 0.10 mm. The number of coil turns was 200. As a result, the magnetic detection power of the unit element was 100 pT.

The size of the sensor grid was 10 mm square. The number of elements in the grid was 2,500 pixels.

The spatial resolution of the magnetic field of the magnetic microscope and the positional accuracy of the center position of the current element were 10 μm, which corresponds to the accuracy of a 100x optical microscope. The intensity and direction of the current were theoretically predicted to be accurate to within ±10%.

[Example 3]
In Example 1, the GSR sensor had a magnetic wire length of 10 μm, a wire diameter of 2 μm, a coil pitch of 0.2 μm, a coil width of 4.5 μm, and a number of wires of 2. The size of the unit element of the sensor grid was 0.011 mm. The number of coil turns was 100. As a result, the magnetic detection power of the unit element was 500 pT.

The size of the sensor grid was 5 mm square. The number of elements in the grid was 250,000 pixels.

The magnetic field spatial resolution of the magnetic microscope and the positional accuracy of the center position of the current element were 1 μm, which corresponds to the accuracy of a 1000x optical microscope. The intensity and direction of the current were estimated to be within ±5% as a result of simulation calculations using a current element calculation program.

[Example 4]
In Example 1, the GSR sensor had a magnetic wire length of 2 mm, a wire diameter of 10 μm, a coil pitch of 3 μm, a coil width of 40 μm, and a number of wires of 4. The size of the unit element of the sensor grid was 0.011 mm. The number of coil turns was 2000. As a result, the magnetic detection power of the unit element was 5 pT.

The size of the sensor grid was 50 mm square. The number of elements in the grid was 625 pixels.

The magnetic field spatial resolution of the magnetic microscope and the positional accuracy of the center position of the current element were 50 μm, which corresponds to the accuracy of a 10x optical microscope. The intensity and direction of the current were estimated to be within ±10% as a result of simulation calculations using a current element calculation program.

[Example 5]
This is a combination of the magnetic microscope of Examples 1 to 4 and an optical microscope.
As shown in Fig. 8, a magnetic sensor element grid 50 was placed below the substrate surface of a petri dish 51, and an optical microscope 53 was placed above it, allowing the movement of cells 52 to be optically observed and photographed with a CCD camera 64, while simultaneously being magnetically observed. The current element distribution measured with a magnetic microscope was superimposed on a diagram of the cell shape observed with the optical microscope, and the image was displayed on a screen. This made it easy to observe both the mechanical movement and shape change of the cells, as well as the activation level of each cell, in real time (Fig. 9).

The present invention makes it easy to observe both the mechanical movement and shape change of cells, as well as the degree of activation of each cell, in real time, allowing for more accurate observation of the growth of cells, particularly IPS cells.

10: magnetic sensor element grid, 101: magnetic sensor element, 11: petri dish, 12: cell (cell aggregate)
2: Magnetic sensor element (GSR sensor element)
21: Substrate, 22: Magnetic wire, 23: Detection coil, 24: Wire terminal, 25: Wire electrode, 26: Wiring, 27: Coil terminal, 28: Coil electrode, 29: Wiring 3: Magnetic sensor element grid 31: Sensor grid substrate (substrate)
32: Grid unit element (unit element)
321x: magnetic sensor element in the X-axis direction (GSR sensor element), 321y: magnetic sensor element in the Y-axis direction (GSR sensor element), 321z: magnetic sensor element in the Z-axis direction (GSR sensor element), 321o: origin (origin of unit element, origin of magnetic field vector sensor), 321w: magnetic wire 33: element base 330: base, 331: trapezoidal slope, 332: ridge line, 333: upper surface 4: unit element 4 (41): 1-axis element grid-Z-axis type 40: petri dish, 411: GSR element, 412: ASIC, 413: electrode, 414: grid wiring (wiring), 415: coating agent 4 (42): 1-axis element grid-X-axis type 40: petri dish, 421: GSR element, 4 22: ASIC, 423: electrode, 424: grid wiring (wiring), 425: coating 4 (43): 1-axis element grid - Y-axis type 40: petri dish, 431: GSR element, 432: ASIC, 433: electrode, 434: grid wiring (wiring), 435: coating 4 (44): 2-axis element grid 40: petri dish, 441: GSR element, 442: ASIC, 443: electrode, 444: grid wiring (wiring), 445: coating 4 (45): 3-axis element grid 40: petri dish, 451: GSR element, 452: ASIC, 453: electrode, 454: grid wiring (wiring), 455: coating 5: cell aggregate 51: cell, 511: contour line, 512: current piece
52: cell, 521: contour line, 522: current element 53: cell, 531: contour line, 532: current element 54: cell, 541: contour line, 542: current element 55: cell 551: contour line, 552: current element 6: integrated system of magnetic microscope and optical microscope 61: magnetic microscope, 611: magnetic sensor element (GSR sensor element), 62: petri dish, 63: cell aggregate, 64: optical microscope, 65: CCD camera 7: cell aggregate 71: area of intense movement

The present invention relates to a magnetic microscope that observes the vitality of an entire cell aggregate by measuring the magnetic field emanating from within the cell aggregate and the electric current fragments that generate the magnetic field, while at the same time observing the structure and movement of the cells using an optical microscope used in cell research.

When observing the growth of cells such as iPS cells, in addition to observing the morphological changes and movement of the cell body using an optical microscope, it is necessary to observe the vitality of the entire cell aggregate by observing the electric current fragments flowing within the cell aggregate.

However, the size of a cell is about 20 μm, and the size of the aggregate is less than 5 mm. By measuring the magnetic field emitted from the current flowing in such a minute cell, the vitality of the cell can be predicted from the strength and direction of the current element. However, there is currently no known magnetic microscope that can measure the current element flowing in such a minute cell.
Here, a magnetic microscope can be defined as a device that detects tiny magnetic fields emitted from tiny cells, amplifies the size of the magnetic fields by 100 to 1000 times in accordance with the shape of the cells, and observes the distribution of the magnetic field or the distribution of electric current fragments that are the source of the magnetic field (Figure 1).

Incidentally, a cell evaluation device using a magnetic sensor is disclosed in Patent Document 1. Figure 17 of the document describes that amorphous wires with a diameter of 20 μm are arranged in a checkerboard structure at intervals of 80 μm, and high-frequency alternating current is passed through both ends of each wire to measure the impedance of the amorphous wire from the voltage at both ends, thereby making it possible to measure the magnetic field Hx or Hy at the intersection of the wires.
However, the magnetic impedance sensor measures the average value of the magnetic field associated with the flowing wire when a voltage is applied to both ends of the wire and an AC current is passed through it, and it is theoretically clear that it cannot measure the magnetic field at the intersection point alone. In fact, no papers supporting this invention have been published, and no products are on the market.

The inventor appears to be mistaken in thinking that this type of sensor, which is a grid sensor in which an element is placed at the (i,j)th position of a grid and a switch is operated so that an AC current flows from the i-th wire to the j-th wire, driving only the (i,j)-th element, is widely known. Although this type of sensor is widely known, the inventor appears to be mistaken in thinking that this technology is not yet developed. In other words, a magnetic grid sensor in which magnetic sensors are placed in a checkerboard pattern with intervals of 2 mm or less has not yet been developed.

The magnetic field emitted from a cell is thought to be about 10 nT or less directly above the cell. Patent Document 1 explains that it is 1 nT or less, but the measurement position of the sensor is 900 μm away from the cell body, whereas the present invention assumes a distance of 300 μm or less, and since the generated magnetic field is inversely proportional to the square of the distance, the views of both documents can be said to be consistent.

Taking into account the size of the cells, the size of the magnetic sensor element required to detect this magnetic field should be approximately 10 μm to 500 μm. Currently known GSR sensors have a detection power of approximately 15 nT when the sensor is 500 μm long, and no small, highly sensitive magnetic sensor that meets the above requirements is known. Furthermore, the manufacturing technology for densely arranging a large number of these magnetic sensor elements in a grid on the X and Y axes has not yet been established.

There is a need to develop a small, highly sensitive magnetic sensor and a magnetic sensor grid using this sensor, in order to develop a magnetic microscope. However, a search using j-platpat did not reveal any prior art literature on magnetic microscopes.

Patent No. 5526384

The present invention measures the magnetic field emitted from a 20 μm-sized cell, and measures the vitality of the cell by observing the tiny magnetic field generated within the cell and the electric current fragment that is the source of the magnetic field.

The inventor came up with the idea that if he developed a small GSR sensor and arranged multiple elements along each row of grid lines in a checkerboard pattern along the X and Y axes, he could invent a magnetic microscope capable of observing cells about 20 μm in size.

Specifically, the idea was that if a magnetic sensor grid with numerous magnetic sensors arranged in a checkerboard pattern along the X and Y axes were attached to the underside of a petri dish being observed under an optical microscope, the micromagnetic field could be measured, and a current element calculated from the measured value, then it would be possible to measure the micromagnetic field, current element, and cell activity at the cellular level in real time. After examining the feasibility of this, the researchers came to the conclusion that the following six challenges lay ahead.

The technological development issues for this purpose are:
The first challenge is to develop a magnetic sensor with a detection power of about 1 pT to 10 nT and a size of 10 μm to 2 mm or less. Specifically, the goal is to reduce the size and improve the performance of current GSR sensors, but the detection power of a magnetic sensor is proportional to its size, and there is a trade-off between the two, making it difficult to improve both characteristics at the same time.

The second challenge is to develop technology to arrange 100 to 4 million magnetic sensor elements in a grid along the X and Y axes on an ASIC board. The number of elements should be selected based on the magnification of 40 to 1,000 times and the measurement area of about 10 mm to 50 mm in diameter.

The third task is to develop a program that uses a magnetic sensor grid to measure the magnetic field emitted by the current element, calculates contour maps of the current element distribution and magnetic field distribution flowing through the cell aggregate, and displays an image on a screen.

The fourth challenge is to develop a magnetic microscope with a magnification of 1,000, using a magnetic sensor grid arranged at an ultra-high density to achieve an image pixel size of about 20 μm and a magnetic field distribution resolution of about 1 μm.

For the first issue, a GSR sensor was used as the magnetic sensor. By setting the coil pitch to 0.1 μm to 3 μm and installing multiple magnetic wires as necessary, the number of coil turns can be set to 150 to 2,000, and the detection power can be increased to 1 pT to 50 pT, making it highly sensitive. This problem can be solved by making the GSR sensor small and highly sensitive. The GSR sensor element is shown in Figure 2.

The GSR sensor is described in detail in Patent Publication No. 5839527 by the present inventor, and is cited in the present invention. The GSR sensor is an ultra-sensitive micromagnetic sensor that includes a GSR element that is composed of a conductive magnetic field detection magnetic wire on a substrate, a detection coil formed of a circular coil wound around the magnetic wire, two electrodes for passing current through the magnetic wire, and wiring that connects two electrodes for detecting the coil voltage, a means for passing a pulse current with a GHz frequency through the magnetic wire, and an electronic circuit that detects the coil voltage generated when the pulse current is passed through the magnetic wire and converts the coil voltage into an external magnetic field H.

The second problem was solved by forming multiple elements on an ASIC board, attaching magnetic wires in a grid pattern along the X and Y axes, and then baking coil wiring and electrode wiring onto the wires. The ASIC and multiple elements were connected using via holes. Multiple ASICs may be used to correspond to multiple elements, or multiple ASICs may be used to control multiple elements with channel switching functions. The magnetic sensor element grid is shown in Figures 3 to 5.
FIG. 3 is a top view of the magnetic sensor element grid, FIG. 4 (FIGS. 4A, 4B, and 4C) is a plan view of the structure of a unit element of the magnetic sensor element grid, and FIG. 5 (FIGS. 5A, 5B, and 5C) is a cross-sectional view of the magnetic sensor element grid.

The magnetic field measurements were taken at the numerous intersections of the X-axis and Y-axis elements, Hz in Figure 4A (a-1), Hx in Figure 4A (a-2), and Hy in Figure 4A (a-3), and these values were used as the magnetic field of the grid to calculate the magnetic field distribution and current element distribution. In other words, they were calculated as Hz = (Hz1 + Hz2 + Hz3 + Hz4)/4, Hx = (Hx1 + Hx2 + Hx3 + Hx4)/4, and Hy = (Hy1 + Hy2 + Hy3 + Hy4)/.

The top surface of the grid element substrate is flattened, and the flat surface is brought into direct contact with the substrate surface on the cell fluid side of the cell observation dish, making the distance from the cell to the element 300 μm or less, facilitating measurement of the micromagnetic field emitted from the cell. The shorter the distance from the cell to the element, the better.

For the third problem, before observation, the magnetic field at the sensor grid position (i, j) is measured to obtain the initial magnetic field measurement value mH _ij (→) (b), and this value is displayed as a magnetic field distribution corresponding to the grid size on the X-axis and Y-axis plane.

Next, after placing a cell in the petri dish, the magnetic field at the sensor grid position (i, j) is measured to obtain the magnetic field measurement value mH _ij (→)(a), and mH _ij (→)(b) is subtracted from the magnetic field measurement value to obtain the measurement value. In other words, mH _ij (→)=mH _ij (→)(a)-mH _ij (→)(b).

This value is considered to be the magnetic field emanating from the current element Ids within the cell aggregate, and measured by a magnetic sensor at the (i, j)th position of the magnetic sensor element grid, the measured value is taken as mH _ij (→), a contour map of the absolute value of mH _ij is created, and the position P(x, y, z) of the current element Ids from the peak of mH _ij (→) and the length of the current element Ids (→) are assumed from the spread of the mountain of the peak of mH _ij (→), and it is assumed that the current element Ids exists within the cell aggregate. If there are n peak locations, a calculation model is created assuming that there are n current elements Ids.

Next, the kth current element I_kds_kPosition P of (→)_k(x_k,y_k,z_k) and the magnetic field measurement position Gij(x _i.j. ,y _i.j. ,0)As the current element position P_k(x_k,y_k,z_k) and the measurement position Gij(x _i.j. ,y _i.j. ,0)If the distance between is Rijk(→), then
The kth current element I_kds_kThe magnetic field that (→) creates at the measurement position G(i,j) is H_ijk(→) = 1/4πRijk³×I_kds_kSince it can be calculated from the basic equation of (→) × Rijk(→), adding up these magnetic fields, we getthe magnetic sensor element gridThe theoretical value tH_i.j.(→) is tH_i.j.(→) = ΣtH_ijkLet (→). (Σ adds k from 1 to n.)

Let the error between the two be e _ij (→) = mH _ij (→) - tH _ij (→), and create the error function E = Σ(e _ij ) ² .
Regarding the direction of _Ik (→), the inclination angle with respect to the XY-axis plane is _φk and the angle with respect to the X- _axis is _θk . From this error function, the error function is partially differentiated using the Gauss-Newton method to derive 7n simultaneous equations, and 7n simultaneous equations consisting of the absolute value of _Ik , _dsk , _θk , _φk , _Xk , Yk, and _Zk are obtained. From these equations, 7n unknowns are obtained, and a large number of current elements flowing through the cell aggregate are calculated. The results are used to create a distribution map of the current elements and displayed on the screen.

Here, we have decided to affix (→) to the vector representation of each vector physical quantity.
FIG. 6 shows a magnetic field distribution map obtained from magnetic field measurements taken with a magnetic microscope, and FIG. 7 shows a flow chart of a calculation program for obtaining a current element.
It should be noted that the present invention is directed to the configuration of a magnetic microscope, and the method of calculating the current element vector from the error function is not limited to the above method.

For the fourth issue, in the unit element of the sensor grid, the coil pitch is set to 0.1 μm, the element length is 10 μm, the number of magnetic wires is set to two, and the number of coil windings is set to 180, resulting in a pixel of the sensor grid of 10 μm x 10 μm. The detection power of the magnetic sensor is ensured to be 5 pT. If a magnetic field distribution map is created from these measured values, the pixel of the magnetic field distribution map will be approximately 1 μm, and a magnetic microscope with a magnification of 1000 times can be obtained.

Next, we superimposed the current distribution measured by the magnetic microscope on the shape of the cells observed by the optical microscope and displayed the image on a screen, which makes it easy to observe both the mechanical movement and shape change of the cells, as well as the activation level of each cell, in real time.
FIG. 8 shows an image of the combined optical microscope and magnetic microscope, and FIG. 9 shows the results of each observation (the magnetic microscope is an image).

In order to make the spatial resolution of the magnetic field equal to the resolution of an optical microscope, the magnetic field at the intermediate position between the discretely measured magnetic field values is calculated by functionally approximating the magnetic field distribution using an interpolation method. This interpolation method can obtain a spatial resolution of the magnetic field of about 1/20 of the grid interval.
It was confirmed that the positional accuracy of the current element was theoretically about 1/20 of the grid interval, based on a program for calculating the position of the current element.

By combining these four solutions, we were able to invent the magnetic microscope.
It should be noted that the present invention is directed to a small, highly sensitive magnetic sensor and is not limited to a GSR sensor, as is apparent from the configuration.

This invention makes it possible to observe the degree of cell vitality when observing cells. Moreover, by integrating it with an optical microscope, it becomes possible to simultaneously measure the shape, movement, and vitality of individual cells in real time, and it is expected to become a basic tool in cell research.

1 is an image diagram of a magnetic microscope, in which (a) is a perspective view and (b) is a side view. FIG. 2 is a diagram showing a GSR sensor element. FIG. 2 is a plan view of a magnetic sensor element grid. FIG. 2 is a plan view of a grid unit of a magnetic sensor element, showing a one-axis element consisting of a Z-axis element, an X-axis element, or a Y-axis element. FIG. 2 is a plan view of a grid unit of a magnetic sensor element, showing a two-axis element consisting of an X-axis element and a Y-axis element. FIG. 2 is a plan view of a grid unit of a magnetic sensor element, showing a three-axis element consisting of an X-axis element and a Y-axis element arranged on an element base. 1A and 1B are cross-sectional views of the z-axis type, x-axis type, and y-axis type taken along the A1-A2 line in a one-axis element grid. 1 is a cross-sectional view of a two-element grid taken along line A1-A2. 1 is a cross-sectional view of a three-element grid taken along line A1-A2. FIG. 1 is a magnetic field distribution diagram obtained from magnetic field measurements taken with a magnetic microscope. FIG. 13 is a flowchart showing a calculation program for determining a current element. This is an image diagram of an optical microscope and a magnetic microscope combined. FIG. 1 is an image of the observation results obtained using a magnetic microscope.

A first embodiment of the present invention is as follows.
A magnetic sensor element grid is provided on a substrate surface on the cell fluid side of a petri dish for cell observation;
a signal processing circuit that detects a grid voltage corresponding to a minute magnetic field generated by a current element flowing inside a cell using a magnetic sensor element grid and converts the grid voltage into a grid magnetic signal;
The magnetic sensor element grid is a grid of magnetic sensor elements arranged at the intersections of a checkerboard pattern along the X-axis and Y-axis on a sensor grid substrate, or more precisely, at four locations on either side of each intersection, to measure the magnetic field at the intersections.
and a display device that displays the absolute value of the grid magnetic field as a contour map .

In addition, the magnetic sensor element of the magnetic microscope is
It is characterized by comprising a single-axis element for measuring one of the magnetic fields, Hz magnetic field, Hx magnetic field, and Hy magnetic field.

In addition, the magnetic sensor element of the magnetic microscope is
It is characterized by comprising a two-axis element for measuring the Hx magnetic field and the Hy magnetic field.

In addition, the magnetic sensor element of the magnetic microscope is
It is characterized by comprising a three-axis element for measuring the Hx magnetic field, the Hy magnetic field and the Hz magnetic field.

In addition, the magnetic microscope
The present invention is characterized by comprising a program for calculating the current element flowing within a cell from the value of the grid magnetic field, and a device for displaying that value as an image on a screen.

In addition, the magnetic microscope
Before observation, the magnetic field at the sensor grid position (i, j) is measured to obtain an initial magnetic field measurement value mH _ij (→) (b), which is then displayed as a magnetic field distribution corresponding to the grid size on the X-axis and Y-axis plane.

Next, after placing a cell in the petri dish, the magnetic field at the sensor grid position (i, j) is measured to obtain the magnetic field measurement value mH _ij (→)(a), from which mH _ij (→)(b) is subtracted to obtain the measurement value mH _ij (→). In other words, the measurement value mH _ij (→)=mH _ij (→)(a)-mH _ij (→)(b).

This measured value mH _ij (→) is considered to be the magnetic field emitted from the current element in the cell aggregate , and is measured by the magnetic sensor at the (i, j)th position of the magnetic sensor element grid. The measured value is taken as mH _ij (→). A contour map of the absolute value of mH _ij is then created.
A computational model is created in which n current fragments Ids exist within the cell aggregate by assuming the position P(x, y, z) of the current fragment Ids from the peak of the absolute value of mH _ij (→) and assuming the length ds of the current fragment Ids from the spread of the peak mountain of the absolute value of mH _ij (→).

Next, let _Pk ( _xk , _yk , _zk ) be the position of the _k -th current element _Ikdsk , _Gij ( _xij , _yij ,0) be the measurement position of the magnetic field, and Rijk be the distance between the current _element position _Pk ( _xk , _yk , _zk ) and the magnetic field measurement position _Gij ( _xij , _yij ,0), then
The magnetic field generated by the k _{-th current element Ikdsk can} be obtained from the equation _Hijk (→) ₌ 1/ _4πRijk3 × ^Ikdsk ₍ →)× _Rijk (→), so by adding up magnetic fields 1 to n, the theoretical value of the magnetic field generated at the measurement position _Gij (i,j) of the magnetic sensor element grid is given as _tHij (→)= _ΣHijk (→),
The error between the two is e _ij (→) = mH _ij (→) - tH _ij (→),
Create the error function E=Σ(e _ij ) ² .

Regarding the direction of I _k (→), the inclination angle with respect to the XY-axis plane is φ _k , and the angle with respect to the X-axis is θ _k .
From this error function, 7n simultaneous equations related to I _k , ds _k , θ _k , φ _k , X _k , Y _k , and Z _k are obtained, and unknowns are obtained from the equations.
The device calculates n current elements flowing through a cell aggregate, and uses the results to create a distribution map of the current elements, which is then displayed on the screen.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an embodiment of the present invention will be described in detail with reference to FIGS.
<Magnetic microscope>
The positional relationship between the magnetic sensor element grid that constitutes the magnetic microscope and the cells (cell aggregates) of the specimen is shown in Fig. 1. The magnetic microscope is also composed of a signal processing circuit that converts the magnetic field signals detected by the magnetic sensor elements, a display device, etc.
The magnetic sensor element grid 10 is made up of a large number of magnetic sensor elements 101 .
Cells (cell aggregates) 12 are placed on a petri dish (for cell observation) 11, and a magnetic sensor element grid 10 is placed thereon. Alternatively, the petri dish 11 may be placed on the magnetic sensor element grid 10 as shown in FIG.
The positional relationship between the two can be arbitrarily selected depending on the size and volume of the cell (cell aggregate) to be observed, the detection capability of the magnetic sensor element grid, and further on the combination with an optical microscope.

<Magnetic sensor>
A GSR sensor is used as the magnetic sensor. The basic structure of the GSR element is shown in Fig. 2, and the structure of the magnetic sensor element grid is shown and explained in Figs.
The diameter of the magnetic wire 22 is 1 μm to 10 μm. The coil pitch of the detection coil 23 is 0.1 μm to 3 μm, and the coil width is 3 μm to 30 μm. The number of magnetic wires is 1 to 8, as required.
It should be noted that the magnetic sensor constituting the magnetic microscope is not limited to the GSR sensor, so long as it can exhibit the characteristics required for a magnetic microscope in terms of size and performance.

The size of the unit elements (magnetic sensor elements) 32 of the magnetic sensor element grid 3 is 10 μm to 1 mm.
When the size of the unit element 32 is 10 μm, it is preferable that the coil pitch is 0.1 μm, the coil width is 3 μm, the number of magnetic wires is two, and the number of coil turns is 150.
On the other hand, when the size of the unit element 32 is 1 mm, it is preferable that the coil pitch is 3 μm, the coil width is 30 μm, the number of magnetic wires is 1 to 8, and the number of coil turns is 2400.

As shown in FIG. 2, the GSR sensor is an ultra-sensitive micro-magnetic sensor that is made up of a GSR element 2 consisting of a conductive magnetic wire 22 for detecting a magnetic field on a substrate 21, a detection coil 23 formed by a circular coil wound around the magnetic wire, and wiring 26, 29 connecting two electrodes 24 for passing current through the wire and two electrodes 28 for detecting the coil voltage, a means for passing a pulse current with a GHz frequency through the magnetic wire, and an electronic circuit that detects the coil voltage generated when the pulse current is passed and converts the coil voltage into an external magnetic field H.

<Magnetic sensor element grid (grid)>
As shown in FIG. 3, the magnetic sensor element grid 3 includes a number of unit elements (magnetic sensor elements) 32 arranged on a grid substrate 31 .
The size of the magnetic sensor element grid 3 is 5 mm square to 20 mm square, or 5 mm diameter to 20 mm diameter. The size of the unit element 32 is 10 μm to 1 mm. The number of unit elements 32 is a minimum of 25 pixels (5×5) to a maximum of 4 million pixels (2000×2000).

As shown in FIG. 4, the unit elements 32 are of three types: a one-axis element type (FIG. 4A), a two-axis element type (FIG. 4B), and a three-axis element type (FIG. 4C).
5 (5A, 5B, and 5C) show cross-sectional views of the three types taken along line A1-A2 in Fig. 3. The cross-sectional views show a cross-section of a magnetic sensor element grid placed on a petri dish.

First, the type of one-axis element (FIG. 4A) is a type in which the GSR sensor element 321 is arranged symmetrically on the grid substrate 31 with respect to the intersection 321o of the X-axis and Y-axis.
(a-1) A one-axis element type in which the Z-axis element 321z is arranged one by one in the Z-axis direction,
(a-2) A one-axis element type in which the X-axis elements 321x are arranged one by one in the X-axis direction;
(a-3) There is a one-axis element type in which the Y-axis elements 321y are arranged one on each side in the Y-axis direction.

These cross-sectional views (FIG. 5A) are
(a-1) One-axis element grid - z-axis type (4(41)) is made up of a GSR element 411, an ASIC 412, an electrode 413 and a coating material 415 for protecting the grid wiring 414, and is placed on a petri dish 40.
(a-2) One-axis element grid - x-axis type (4 (42)) is made of a GSR element 421, an ASIC 422, an electrode 423 and a coating material 425 that protects the grid wiring 424, and is placed on a petri dish 40.
(a-3) One-axis element grid - y-axis type (4 (43)) is made up of a GSR element 431, an ASIC 432, an electrode 433 and a coating material 435 for protecting a grid wiring 434, and is placed on a petri dish 40.

Next, the two-axis element type (FIG. 4B) is a type in which the GSR sensor element 321 is attached to the grid substrate 31 along the intersection 321o of the X-axis and the Y-axis.
(b) There is a two-axis element type in which two X-axis elements 321x in the horizontal direction and two Y-axis elements 321y in the vertical direction are arranged symmetrically with respect to the origin 321o.

The cross-sectional view of this biaxial element type (4 (44)) is shown in FIG. 5B. It consists of a GSR element 441, an ASIC 442, and a coating 445 that protects electrodes 443 and grid wiring 444, and is placed on a petri dish 40.

Finally, the type of three-axis element (FIG. 4C) is a three-axis type in which a magnetic field vector sensor is arranged on the grid substrate 31 along the intersection 321o of the X-axis and Y-axis.
This three-axis element type has four GSR elements or on-ASIC type GSR sensors arranged in four-fold symmetry and mirror symmetry on the inclined surface 331 of a base 330 consisting of a square pyramid, an octagonal pyramid, or an irregular octahedral pyramid, so that the inclination direction coincides with the magnetic wire 321w of the GSR element.
The magnetic fields in the X-axis and Y-axis directions are measured from two X-axis elements 321x and two Y-axis elements 321y, and the magnetic field in the Z-axis direction is obtained by calculation.
The on-ASIC type GSR sensor was invented by the inventors and is disclosed in a patent publication (Japanese Patent No. 7062216). Please refer to the description in this patent publication for details. The magnetic field vector sensor was invented by the inventors and is disclosed in a patent publication (Japanese Patent No. 7215702). Please refer to the description in this patent publication for details.

The cross-sectional view of this triaxial element type (4(45)) is shown in FIG. 5C, and it is made up of a GSR element 451, an ASIC 452, electrodes 453, and a coating material 455 that protects grid wiring 454, and is placed on a petri dish 40.
The magnetic field in the Z-axis direction is obtained by calculation, and element 451z is a calculated element.

<Positional relationship between magnetic sensor element grid and cells>
It is preferable that the position of the magnetic sensor element grid 10 is such that the thickness of the petri dish 11 is 200 μm or less, the top surface of the magnetic sensor element grid 10 is flat, the flat surface is directly pressed against the back side of the petri dish, and the distance from the cells to the measurement parts of the magnetic sensor elements 101 is 300 μm or less. Since the magnetic field emitted by the cells is inversely proportional to the square of the distance, it is necessary to make this distance as small as possible.

<Measurement of magnetic field components at intersections of unit elements>
(A) Measurement of magnetic field components in a one-axis element (a-1) From the four pieces of data Hz1, Hz2, Hz3, and Hz4 measured using four unit elements 321z centered on the intersection 321o, the magnetic field component Hz at the intersection can be calculated from Hz = (Hz1 + Hz2 + Hz3 + Hz4)/4.
(a-2) From the four pieces of data Hx1, Hx2, Hx3, and Hx4 measured by the four unit elements 321x centered on the intersection 321o, the magnetic field component Hx of the intersection can be calculated from Hx = (Hx1 + Hx2 + Hx3 + Hx4)/4.
(a-3) From the four data Hy1, Hy2, Hy3, and Hy4 measured by the four unit elements 321y centered on the intersection 321o, Hy, which is the magnetic field component of the intersection, can be calculated from Hy=(Hy1+Hy2+Hy3+Hy4)/4.

(B) Measurement of magnetic field components in a two-axis element (b) From the four pieces of data Hx1, Hx2, Hy1, and Hy2 measured with two unit elements 321x oriented in the X-axis direction and two unit elements 321y oriented in the Y-axis direction centered on the intersection 321o, the magnetic field components Hx and Hy at the intersection can be calculated from Hx = (Hx1 + Hx2)/2, Hy = (Hy1 + Hy2)/2.

(C) Measurement of magnetic field components in a three-axis element (c) In a magnetic field vector sensor, the tilt angle of the base is θ, and from the four data Hx1, Hx2, Hy1, and Hy2 measured with two unit elements 321x oriented in the X-axis direction and two unit elements 321y oriented in the Y-axis direction centered on the intersection 321o, the magnetic field components Hx and Hy at the intersection 321o can be calculated from Hx = (1/2 cos θ) (Hx1-Hx2) and Hy = (1/2 sin θ) (Hy1-Hy2). The Hz in the Z-axis direction can be calculated from Hz = (1/4 sin θ) (Hx1+Hx2+Hy1+Hy2).

In addition, if you first measure the environmental magnetic field before taking measurements, and then subtract the environmental magnetic field from the measured value when placing the cells in the petri dish for observation, there will be no effect from the external magnetic field.

The magnetic sensor element grid also has an electronic circuit that expresses the position of the intersection of the unit element 32 as the i-th position in the X-axis direction and the j-th position in the Y-axis direction as P _ij , expresses the measurement value at P _ij as H _ij , and transfers H _ij to the current element distribution calculation program.
In addition, the position on the grid substrate of each element (elements in the X-axis, Y-axis, and Z-axis directions) that constitutes the unit element 32 may be designated as (i, j), and the measurement values of each element may be directly used as grid magnetic field measurement values to calculate the magnetic field distribution and current element distribution.

<Program to calculate current element distribution>
The program for calculating the current element distribution measures the magnetic field emanating from the current element in the cell aggregate with a magnetic sensor at the (i, j)th position of the magnetic sensor element grid, and defines the measured value as mH _ij .

In practice, since the measurement is affected by the environmental magnetic field, the magnetic field at the magnetic sensor element grid position (i, j) is measured before observation to obtain the initial magnetic field measurement value mH _ij (→) (b), and this value is displayed as the magnetic field distribution corresponding to the grid size on the X-axis and Y-axis plane.

Next, after placing a cell in the petri dish, the magnetic field at the magnetic sensor element grid position (i, j) is measured to obtain the magnetic field measurement value mH _ij (→)(a), from which mH _ij (→)(b) is subtracted to obtain the measurement value.
That is, mH _ij (→)=mH _ij (→)(a)-mH _ij (→)(b).
This value is considered to be the magnetic field emanating from the current element within the cell aggregate, and is measured by the magnetic sensor at the (i, j)th position of the magnetic sensor element grid, and the measured value is taken as mH _ij (→).

A contour map of the absolute value of mH _ij is created, n peak positions are identified, and the position P _k (x _k , y _k , z _k ) of the current element from the peak of each absolute value of mH _ij and the length of the current element ds _k (→) are assumed from the spread of the peak, and it is assumed that the current element I _k ds _k (→) exists at the position P _k (x _k , y _k , z _k ) within the cell aggregate. When there are n peak locations (one or more), a calculation model is created assuming that there are n current elements.

Next, let _Pk ( _xk , _yk , _zk ) be the position of the _k - _th current element Ikdsk, _Gij ( _xij , _yij ,o) be the measurement position of the magnetic field, and Rijk be the distance between the current _element position _Pk ( _xk , _yk , _zk ) and the measurement position _Gij ( _xij , _yij ,o), then
The magnetic field generated by the kth _{current element Ikdsk can} be obtained from the equation _Hijk (→) = ₁ / _4πRijk3 × ^Ikdsk ₍ →) × _Rijk (→), so the magnetic fields generated by n current elements are added together to obtain the theoretical value _tHij (→) = _ΣtHijk (→) generated at position (i, j) of the magnetic sensor element grid.

The error between the two is e _ij =mH _ij (→)-tH _ij (→), and the error function E=Σ(e _ij ) ² is created.
Regarding the direction of _Ik , the inclination angle with respect to the XY-axis plane is _φk and the angle with respect to the X-axis is _θk . From this error function, the error function is partially differentiated using the Gauss-Newton method to derive 7n simultaneous equations. 7n simultaneous equations consisting of _Ik , _dsk , _θk , _φk , _Xk , _Yk , and _Zk are obtained, and 7n unknowns are obtained from these equations. A large number of current elements flowing through the cell aggregate are calculated, and the results are used to create a distribution map of the current elements, which is then displayed on the screen as a magnetic field distribution map.

The program procedure is as follows (Figure 7). Note that vectors are displayed as (→).
First step (101);
Before observation, the magnetic field at the sensor grid position (i, j) is measured to obtain an initial magnetic field measurement mH _ij (→)(b).
However, the position of each element is specified in advance using a sensor grid coordinate system O-XYZ.

Second step (102);
After placing the cells in the dish, measure the magnetic field at the sensor grid position (i, j) to obtain the magnetic field measurement value mH_i.j.(→) Find (a), and from there find mH_i.j.(→) Subtract (b) to get the measured value mH_i.j.(→), calculate the magnetic field distribution, mH_i.j.(→) If there are k peaks in the absolute value distribution, the peak positions P_k(X_k, Y_k, Z_k) to current element I_kds_kAssume that exists. The length of ds is determined from the peak position and the extent of the peak.

FIG. 6 shows an example of a magnetic field distribution diagram.
The cell assembly 4 is composed of five cells (n=5) , namely, cell 41 to cell 45, and the contour lines 411 to 451 of the respective cells are illustrated to show a contour map. Current elements I ₁ ds ₁ to I ₅ ds ₅ are present at the peaks of the respective contour maps, and their positions are positions P ₁ to P ₅ .
This is transferred to the current element distribution calculation program.

Third step (103):
Assuming that _a current element _Ikdsk exists at peak position _Pk , the theoretical magnetic field strength at position _Gij ( _Xij , _Yij ) of each magnetic sensor element grid created by n current elements Ids at position _Pk , that is, the magnetic field emitted by the kth current element, can be calculated by the formula _tHijk (→)=1/ _4πRijk3 × ^Ikdsk ₍ →)× _Rijk ( _→ ). Since there are 1 to n current elements, the theoretical magnetic field strength at position _Gij ( _Xij , _Yij ) is calculated by the sum of n _, tHij ₌ Σ1/ _4πRijk3 × ^Ikdsk × _{Rijk .}

A fourth step (104):
If the distance vector R _ijk (→) between the position I _k ds _k of the kth current element and the measurement position G _ij (X _ij , Y _ij ) of the magnetic field, the inclination of the current element with respect to the Z axis is φ _k , and the inclination with respect to the X axis is θ _k , then
tH _ij is a function of the current intensity I _k , the length ds _k of the current element, the positions X _ijk , Y _ijk , and Z _ijk , and the azimuth angles θ _k and φ _k .

Fifth step (105);
Calculate the measurement error.
eij ₍ →) = mHij ₍ →) - _tHij (→)

A sixth step (106);
Calculate the sum of squares of the errors.
^E = _Σeij2

Seventh step (107);
The current intensity _Ik , length _dsk to the current element, Xk _{, Yk} , _Zk, and azimuth angles _θk , _φk that minimize the sum of squares of the error are calculated using the Gauss-Newton method.

Eighth step (108);
A current distribution is calculated from the _{magnitude Ikdsk of} each current element and the values of the positions _Xk , _Yk , and _Zk , and the calculated current distribution is displayed as an image on a PC screen. Alternatively, a current distribution is calculated from the current intensity _Ik of each current element and the values of the positions _Xk , _Yk , and _Zk , and the calculated current distribution is displayed as an image on a PC screen.

The center position of the current segment has a positional accuracy of less than 1/10 of the pixel size, and the current intensity can be calculated with an accuracy of less than ±10%.

In addition, it is constrained by the distribution of each current element and the law of conservation of current, so we check whether there is any contradiction with the law of conservation of current.

In an embodiment of the present invention, as shown in Figures 8 and 9, a magnetic sensor element grid 50 is placed below the substrate surface of a petri dish 51, and an optical microscope is placed above, so that the movement of cells 52 can be observed optically and simultaneously magnetically .

The current element distribution measured by the magnetic microscope is superimposed on a shape diagram of the cell 52 observed and photographed by an optical microscope, and the image is displayed on a screen.
This makes it possible to easily observe both the mechanical movement and shape change of the cells (FIG. 9(a)) and the activation level 71 of each cell (FIG. 9(b)) in real time.

[Example 1]
In the first embodiment of the present invention, the design elements are as follows:
A GSR sensor is used as the magnetic sensor. The structure of the GSR element is shown in Figure 2.
The length of the magnetic wire 22 was 0.95 mm, the wire diameter was 10 μm, the coil pitch was 3 μm, the coil width was 30 μm, and the number of magnetic wires was 4. The number of coil turns was 1,200. The size of the unit element of the sensor grid was 1 mm. As a result, the magnetic detection force of the unit element was 1 pT.

The GSR sensor used here is an ultra-sensitive micro-magnetic sensor consisting of a GSR element, as shown in Figure 2, which is composed of a conductive magnetic wire for detecting a magnetic field on a substrate, a detection coil formed by a circular coil wound around the magnetic wire, two electrodes for passing current through the wire, and wiring connecting two electrodes for detecting the coil voltage, and a means for passing a pulse current having a frequency of 1.2 GHz through the magnetic wire, and an electronic circuit that detects the coil voltage generated when the pulse current is passed and converts the coil voltage into an external magnetic field H.

The magnetic sensor element grid is an arrangement of many unit elements as shown in Figure 3. The unit elements are made up of two-axis elements, and as shown in Figure 4B (b), two GSR sensor elements are arranged on the grid substrate along the X-axis and two along the Y-axis, making it possible to measure the magnetic field (Hx, Hy) at the intersection of the X-axis element and the Y-axis element. The size of the grid sensor is 20 mm square. The number of elements in the grid is 361 pixels.

The grid substrate is made of a flat insulating material or an ASIC substrate coated with an insulating film. The grid substrate consists of a top surface for arranging elements, an electronic circuit surface for signal processing, and a strong material to ensure mechanical strength. The GSR elements are placed on the ASIC substrate surface, which consists of electronic circuits and strong materials, and the two are connected with via holes.

The position of the grid substrate was set so that the thickness of the dish was 160 μm, the top surface of the magnetic sensor element grid was flattened, and the flat surface was pressed directly against the back side of the dish, so that the distance from the cell to the measurement part of the magnetic sensor element at the cell liquid surface was 240 μm.

From the four data points Hx1, Hx2, Hy1, and Hy2 measured by the unit element, the magnetic field components Hx and Hy at the intersection were calculated from Hx = (Hx1 + Hx2)/2 and Hy = (Hy1 + Hy2)/2. Before measurement, the environmental magnetic field was first measured, and when the cells were placed in the dish and observed, the environmental magnetic field was subtracted from the measured value to avoid the influence of external magnetic fields.

The sensor grid represents the intersection of the sensor elements at the i-th position in the X-axis direction and the j-th position in the Y-axis direction as P _ij , and the measured value at P _ij is H _ij (→) to obtain the magnetic field measurement distribution. This measured value H _ij (→) was transferred to the current element distribution calculation program via an electronic circuit.
Furthermore, for the magnetic field between the grids, an approximation curve was created by the interpolation method, and a magnetic field spatial distribution ability of about 1/20 of the grid interval was obtained.

The program for calculating the current element distribution measures the magnetic field emitted from the current element in the cell aggregate with the magnetic sensor at the (i, J)th position of the magnetic sensor grid, and the measured value is mH _ij (→). The theoretical value tH _ij (→) generated by the kth current element is the magnetic field generated by each current element , and tH _ijk (→). = 1/4πR _ijk ³ × I _k ds _k (→) × R _ijk (→), so the magnetic fields created by n current elements were added together to find the error between the two. The error between the two is e _ij (→) = mH _ij (→) - tH _ij (→), and the error function E is defined as E = Σ(e _ij ) ^2. For the direction of I, the inclination angle with respect to the XY-axis plane is defined as φ, and the angle with respect to the X-axis is defined as θ.

When there are six current elements (that is, n=6) , by partially differentiating the error function using the Gauss-Newton method, 42 simultaneous equations can be derived and then the solution can be obtained.

The program procedure is as follows (Figure 7). Note that vectors are displayed as (→).
The first step (101) is to measure the magnetic field at a sensor grid position (i,j) before observation to obtain an initial magnetic field measurement mH _ij (→)(b).
However, the position of each element is specified in advance using a sensor grid coordinate system O-XYZ.

The second step (102) is to place the cells in the dish, measure the magnetic field at the magnetic sensor element grid position (i, j), and obtain the magnetic field measurement value mH_i.j.(→) Find (a), and from there find mH_i.j.(→) Subtract (b) to get the magnetic field measurement value mH_i.j.(→), calculate the magnetic field distribution, mH_i.j.(→) Absolute value distributionn piecesIf there is a peak inThe kthPeak position P_k(x_k, y_k, z_k) to current element I_kds_kAssume that exists. The length of ds is determined from the spread of the peak, centered on the peak position. Transfer this to the current element distribution calculation program.

In the third step (103), assuming that a current element Ids _k exists at the peak position P _k , the theoretical magnetic field strength at position P(i,j) of each magnetic sensor element grid created by n current elements Ids (k ranges from 1 to n) at position _P k can be calculated using the formula tH _ij = 1/4πR _ij ³ × Ids × R _ij . Since there are n current elements, the theoretical magnetic field strength at position P(i,j) is calculated by the sum of n , tH _ij = Σ 1/4πR _ijk ³ × I _k ds _k × R _ijk .

In the fourth step (104), when the distance vector R _ijk (→) between the kth current element position I _k ds _k and the magnetic field measurement position G _k is φ k, the inclination of the current element with respect to the Z axis is φ _k , and the inclination with respect to the X axis is θ _k , the theoretical magnetic field strength tH _ij is a function of the current strength I _k , the length ds _k of the current element, the positions X _ijk , Y _ijk , Z _ijk , and the azimuth angles θ _k and φ _k .

The fifth step (105) is to calculate the measurement error.
eij ₍ →) = mHij ₍ →) - _tHij (→)

The sixth step (106) is to find the sum of the squares of the errors.
^E = _Σeij2

The seventh step (107) is to calculate _Ik , _dsk , Xk _{, Yk} , _Zk , and azimuth angles _θk and _φk that minimize the sum of squares of the error by the Gauss-Newton method.

The eighth step (108) is to calculate the current element vector distribution or the current intensity of the current element from the magnitude _Ik and _dsk of each current element, the positions _Xk , _Yk , _Zk , and the azimuth angles _θk and _φk , and display it as an image on the PC screen.
The center position, intensity, and direction of the current segment have a positional accuracy of 1/10 of the pixel size or less, and the current intensity and direction can be calculated with an accuracy of ±10% or less.
Furthermore, the distribution of each current element was consistent with the law of conservation of current.

The size of the sensor grid was 20 mm square, and the number of elements in the grid was 400 pixels.
The magnetic field spatial resolution of the magnetic microscope and the positional accuracy of the center position of the current element were 50 μm, which corresponds to the accuracy of a 20x optical microscope. The intensity and direction of the current were estimated to be within ±10% accuracy as a result of simulation calculations using a current element calculation program.

[Example 2]
In Example 1, the GSR sensor had a magnetic wire length of 10 μm, a wire diameter of 8 μm, a coil pitch of 0.1 μm, a coil width of 20 μm, and a number of wires of 2. The size of the unit element of the sensor grid was 0.10 mm. The number of coil turns was 200. As a result, the magnetic detection power of the unit element was 100 pT.

The size of the sensor grid was 10 mm square. The number of elements in the grid was 2,500 pixels.

The spatial resolution of the magnetic field of the magnetic microscope and the positional accuracy of the center position of the current element were 10 μm, which corresponds to the accuracy of a 100x optical microscope. The intensity and direction of the current were theoretically predicted to be accurate to within ±10%.

[Example 3]
In Example 1, the GSR sensor had a magnetic wire length of 10 μm, a wire diameter of 2 μm, a coil pitch of 0.2 μm, a coil width of 4.5 μm, and a number of wires of 2. The size of the unit element of the sensor grid was 0.011 mm. The number of coil turns was 100. As a result, the magnetic detection power of the unit element was 500 pT.

The size of the sensor grid was 5 mm square. The number of elements in the grid was 250,000 pixels.

The magnetic field spatial resolution of the magnetic microscope and the positional accuracy of the center position of the current element were 1 μm, which corresponds to the accuracy of a 1000x optical microscope. The intensity and direction of the current were estimated to be within ±5% as a result of simulation calculations using a current element calculation program.

[Example 4]
In Example 1, the GSR sensor had a magnetic wire length of 2 mm, a wire diameter of 10 μm, a coil pitch of 3 μm, a coil width of 40 μm, and a number of wires of 4. The size of the unit element of the sensor grid was 0.011 mm. The number of coil turns was 2000. As a result, the magnetic detection power of the unit element was 5 pT.

The size of the sensor grid was 50 mm square. The number of elements in the grid was 625 pixels.

The magnetic field spatial resolution of the magnetic microscope and the positional accuracy of the center position of the current element were 50 μm, which corresponds to the accuracy of a 10x optical microscope. The intensity and direction of the current were estimated to be within ±10% as a result of simulation calculations using a current element calculation program.

[Example 5]
This is a combination of the magnetic microscope of Examples 1 to 4 and an optical microscope.
As shown in Fig. 8, a magnetic sensor element grid 50 was placed below the substrate surface of a petri dish 51, and an optical microscope 53 was placed above it, allowing the movement of cells 52 to be optically observed and photographed with a CCD camera 64, while simultaneously being magnetically observed. The current element distribution measured with a magnetic microscope was superimposed on a diagram of the cell shape observed with the optical microscope, and the image was displayed on a screen. This made it easy to observe both the mechanical movement and shape change of the cells, as well as the activation level of each cell, in real time (Fig. 9).

The present invention makes it easy to observe both the mechanical movement and shape change of cells, as well as the degree of activation of each cell, in real time, allowing for more accurate observation of the growth of cells, particularly IPS cells.

10: magnetic sensor element grid, 101: magnetic sensor element, 11: petri dish, 12: cell (cell aggregate)
2: Magnetic sensor element (GSR sensor element)
21: Substrate, 22: Magnetic wire, 23: Detection coil, 24: Wire terminal, 25: Wire electrode, 26: Wiring, 27: Coil terminal, 28: Coil electrode, 29: Wiring 3: Magnetic sensor element grid 31: Sensor grid substrate (substrate)
32: Grid unit element (unit element)
321x: magnetic sensor element in the X-axis direction (GSR sensor element), 321y: magnetic sensor element in the Y-axis direction (GSR sensor element), 321z: magnetic sensor element in the Z-axis direction (GSR sensor element), 321o: origin (origin of unit element, origin of magnetic field vector sensor), 321w: magnetic wire 33: element base 330: base, 331: trapezoidal slope, 332: ridge line, 333: upper surface 4: unit element 4 (41): 1-axis element grid-Z-axis type 40: petri dish, 411: GSR element, 412: ASIC, 413: electrode, 414: grid wiring (wiring), 415: coating agent 4 (42): 1-axis element grid-X-axis type 40: petri dish, 421: GSR element, 4 22: ASIC, 423: electrode, 424: grid wiring (wiring), 425: coating 4 (43): 1-axis element grid - Y-axis type 40: petri dish, 431: GSR element, 432: ASIC, 433: electrode, 434: grid wiring (wiring), 435: coating 4 (44): 2-axis element grid 40: petri dish, 441: GSR element, 442: ASIC, 443: electrode, 444: grid wiring (wiring), 445: coating 4 (45): 3-axis element grid 40: petri dish, 451: GSR element, 452: ASIC, 453: electrode, 454: grid wiring (wiring), 455: coating 5: cell aggregate 51: cell, 511: contour line, 512: current piece
52: cell, 521: contour line, 522: current element 53: cell, 531: contour line, 532: current element 54: cell, 541: contour line, 542: current element 55: cell 551: contour line, 552: current element 6: integrated system of magnetic microscope and optical microscope 61: magnetic microscope, 611: magnetic sensor element (GSR sensor element), 62: petri dish, 63: cell aggregate, 64: optical microscope, 65: CCD camera 7: cell aggregate 71: area of intense movement

Claims (6)

A magnetic sensor element grid is provided on a substrate surface on the cell fluid side of a petri dish for cell observation;
a signal processing circuit for detecting a grid voltage corresponding to a minute magnetic field generated by a current element flowing in a cell in the magnetic sensor element grid and converting the grid voltage into a value of a grid magnetic field;
the magnetic sensor element grid is provided with magnetic sensor elements at intersections of a checkerboard pattern along the X-axis and the Y-axis on a sensor grid substrate, and measures a magnetic field at the intersections;
and a display device for displaying the value of the grid magnetic field as a magnetic field contour map. In claim 1,
The magnetic microscope according to the present invention, wherein the magnetic sensor element is a one-axis element for measuring one of a Hz magnetic field, a Hx magnetic field, and a Hy magnetic field. In claim 1,
The magnetic microscope according to the present invention, wherein the magnetic sensor element is a two-axis element for measuring a Hx magnetic field and a Hy magnetic field. In claim 1,
The magnetic microscope, wherein the magnetic sensor element is a three-axis element for measuring a Hx magnetic field, a Hy magnetic field, and a Hz magnetic field. In any one of claims 1 to 4,
A magnetic microscope comprising a program for calculating the current segments flowing within a cell from the value of the grid magnetic field, and a display device for displaying the values calculated by the program as an image on a screen. In claim 5,
The magnetic field generated from the current element in the cell is measured by a magnetic sensor at the (i, j)th position of the magnetic sensor element grid, the measured value is designated as mH _ij (→), and a contour map of the absolute value of mH _ij is created;
A calculation model is created in which k current pieces Ids are present in the cell assembly by assuming a position P(x, y, z) of the current piece Ids from the peak of the mH _ij (→) and assuming a length of the current piece Ids from the spread of the peak of the mH _ij (→),
Next, let _Pk ( _xk , _yk , _zk ) be the position of the _k -th current element _Ikdsk , _Gij ( _xij , _yij , 0) be the measurement position of the magnetic field, and Rijk be the distance between the current _element position _Pk ( _xk , _yk , _zk ) and the magnetic field measurement position _Gij ( _xij , _yij , 0), then
The magnetic field generated by _{the k-th current element Ikdsk on} the sensor grid substrate _{can be calculated from the equation Hk(→)=1/4πRk3×Ikdsk ( →} ⁾ _{× Rk} (→), and these magnetic fields are added together to obtain a theoretical value _tHij (→) generated at the measurement position G(i,j) of the magnetic sensor grid,
The error between the two is e _ij (→) = mH _ij (→) - tH _ij (→),
Create the error function _Eij = Σ( _eij ) ² ,
Regarding the direction of I _k (→), the inclination angle with respect to the XY-axis plane is φ _k , and the angle with respect to the X-axis is θ _k .
From this error function, 7k simultaneous equations related to _Ik , _dsk , _θk , _φk , _Xk , _Yk , and _Zk are obtained, and unknowns are obtained from the equations.
A magnetic microscope characterized in that k current pieces Ids flowing through the cell aggregate are calculated, and a distribution map of the current pieces Ids is created using the results and displayed on a screen.

Images