JP7061412B1 - Dielectric particle estimation device and dielectric particle type estimation system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a dielectric particle type estimation device capable of estimating the type of a bacterium to be inspected. SOLUTION: A dielectric particle estimation device 7 acquires an image of captured dielectric particles by causing dielectric migration of the dielectric particles contained in a sample solution, and estimates the type of the dielectric particles. The particle estimation unit is classified into the first category based on the image processing unit 60 that performs image processing for removing noise from the captured image and the shape of the dielectric particles contained in the image that has undergone image processing. The shape classification unit 62, the growth environment analysis unit 64 that classifies the dielectric particles into the second category based on the growth environment, and the image analysis of the dielectric particles having substantially the same shape in the second category are performed. The estimation unit that estimates the type of dielectric particles based on the data classified by the same-shape in-shape classification unit 66 and the same-shape in-shape classification unit 66 that extracts the differences in the above and classifies them into the third category based on the differences. 70 and. [Selection diagram] Fig. 1

Description

The present invention relates to a dielectric particle estimation device and a dielectric particle type estimation system.

Conventionally, an inspection method for inspecting dielectric particles such as bacteria and cells contained in a sample liquid by using dielectrophoresis has been known.

As a technique related to the present invention, for example, an inspection device for inspecting dielectric particles contained in a sample liquid, a pair of at least a pair of electrodes and a pair of flow paths extending in a predetermined direction on the pair of electrodes. A dielectric collecting section having a plurality of slit regions separated from each other while arranging in a predetermined direction between the electrodes of the above, a pump section for feeding the sample liquid so as to proceed in the predetermined direction through the flow path, and a feeding section. An inspection device including an AC voltage supply unit that supplies an AC voltage of a predetermined frequency to a pair of electrodes so that the dielectric particles in the sample liquid to be liquid cause dielectrophoresis is disclosed.

International Publication No. 2017/061496

When inspecting bacteria using an inspection device using dielectrophoresis, it is desirable to be able to estimate the type of bacteria to be inspected.

An object of the present invention is to provide a dielectric particle estimation device and a dielectric particle type estimation system capable of estimating the type of bacteria to be inspected.

The dielectric particle estimation device according to the present invention acquires an image of the dielectric particles captured by causing dielectric migration of the dielectric particles contained in the sample liquid, and estimates the type of the dielectric particles. It is a dielectric particle estimation device having a body particle estimation unit, and the dielectric particle estimation unit performs image processing for removing noise from an image captured by the image pickup unit and recognizing / extracting the dielectric particle corresponding portion. An image processing unit to be performed, a shape classification unit that classifies the dielectric particles into a first category that is roughly classified based on the shape of the dielectric particles contained in the image on which the image processing is performed, and the shape classification. After being classified by the growth environment analysis unit, the dielectric particles are classified into the second category, which is a subordinate concept of the first category, based on the growth environment, and after being classified by the growth environment analysis unit. Image analysis of the dielectric particles having substantially the same shape in the second category is performed to extract differences in feature amounts, and based on the differences, the dielectric particles are classified into the third category of the subordinate concept of the second category. It is characterized by including a classification unit and an estimation unit for estimating the type of the dielectric particles based on the data classified by the classification unit in the same shape.

Further, in the dielectric particle estimation device according to the present invention, when the estimation unit cannot be estimated based on the data classified in the third category, only similar bacterial species in the third category are image-analyzed. It is provided with a special classification unit that extracts the difference in the feature amount and classifies it into the fourth category of the subordinate concept of the third category based on the difference, and the estimation unit is based on the data classified by the special classification unit. It is preferable to estimate the type of the dielectric particles based on the above.

Further, in the dielectric particle estimation device according to the present invention, at least a pair of electrodes, a flow path extending in a predetermined direction on the pair of electrodes, and the pair of electrodes are separated from each other while being arranged in the predetermined direction. A dielectric collecting portion having a plurality of slit regions, a pump portion for feeding the sample liquid so as to proceed in the predetermined direction in the flow path, and the sample liquid to be fed. An AC voltage supply unit that supplies an AC voltage of a predetermined frequency to the pair of electrodes so that the dielectric particles cause dielectrophoresis, and a predetermined region in which the plurality of slit regions are arranged in the dielectric collection unit. It is preferable to include an imaging unit for imaging.

Further, in the dielectric particle estimation device according to the present invention, the imaging unit includes a light source unit that outputs illumination light toward the dielectric particles provided in the plurality of slit regions, and the light source unit that outputs the illumination light. A ring slit portion for converting the illumination light into the ring-shaped illumination light, a condensing cylinder portion for condensing the ring-shaped illumination light formed by the ring slit portion, and the condensing cylinder portion. The straight-ahead light and the diffracted light collected by the lens body are aggregated and pass through the lens body portion including the dielectric particles, and the straight-ahead light and the diffracted light aggregated by the lens body portion. It is preferable to have an objective lens portion for forming an image.

Further, in the dielectric particle estimation device according to the present invention, it is preferable that the region connecting the plurality of slit regions between the pair of electrodes is arranged outside the flow path.

Further, in the dielectric particle estimation device according to the present invention, the dielectric particles preferably contain at least one of bacteria, cells, microorganisms, fungi, spores, and viruses.

In the dielectric particle estimation system according to the present invention, the plurality of dielectric particle estimation devices are connected to a plurality of client-side dielectric particle estimation devices and the plurality of client-side dielectric particle estimation devices via a network. The dielectric particle estimation device on the server side includes the dielectric particle estimation device on the server side, and the dielectric particle estimation device extracts the feature amount of the dielectric particles from the image processed by the image processing unit. The dielectric particle estimation device on each client side has a learning unit that performs deep-layer learning based on the feature amounts of the plurality of dielectric particles extracted based on the body particle estimation unit, and the dielectric particle estimation device on each client side is a previously specified bacterial species. The dielectric particle estimation device on the server side estimates the type of the dielectric particle from the inside, and the dielectric particle estimation device on the server side estimates the type of the dielectric particle from the inside including the bacterial species other than the specified bacterial species. do.

According to the present invention, the type of bacteria to be inspected can be estimated.

It is a figure which shows the dielectric particle type estimation system which has the dielectric particle estimation apparatus of embodiment which concerns on this invention. It is a figure which shows the inspection apparatus of the dielectric particle type estimation system of embodiment which concerns on this invention. It is a figure which shows the image pickup part of the inspection apparatus of the dielectric particle type estimation system in the embodiment which concerns on this invention. It is a figure for demonstrating the collection unit in embodiment which concerns on this invention. In the embodiment of the present invention, it is an enlarged view of the wiring area in the electrode film of FIG. 4 (e). In the embodiment of the present invention, it is an enlarged view of the microelectrode part in the region near the end part of a flow path. It is a figure which shows the principle of the inspection method in embodiment which concerns on this invention. It is a figure which shows the state which the noise is removed by the image processing part in the embodiment which concerns on this invention. It is a figure which shows the state which the feature amount is extracted by the feature amount extraction part in the embodiment which concerns on this invention. It is a figure which shows the procedure of estimation using the dielectric particle estimation apparatus in the embodiment which concerns on this invention. It is a figure which shows the state of classifying by the special classification part of the dielectric particle estimation apparatus in the embodiment which concerns on this invention. It is a figure which shows the state of the test by the simple test of the dielectric particle estimation apparatus in the embodiment which concerns on this invention. It is a figure which shows the state of the test by the simple test of the dielectric particle estimation apparatus in the embodiment which concerns on this invention.

Hereinafter, embodiments according to the present invention will be described in detail with reference to the accompanying drawings. In the following, similar elements are designated by the same reference numerals in all drawings, and duplicate description will be omitted. In addition, in the explanation in the text, the reference numerals described earlier shall be used as necessary.

FIG. 1 is a diagram showing a dielectric particle type estimation system 1 according to an embodiment of the present invention. FIG. 2 is a diagram showing an inspection device 2 of the dielectric particle type estimation system 1 according to the embodiment of the present invention. FIG. 3 is a diagram showing an image pickup unit 12 of the inspection device 2 of the dielectric particle type estimation system 1 in the embodiment according to the present invention.

The dielectric particle type estimation system 1 includes a plurality of client-side dielectric particle estimation devices 7, each client-side dielectric particle estimation device 7, and a server-side dielectric particle estimation device 8 connected via a network 9. It is configured to include a plurality of dielectric particle estimation units including. The dielectric particle estimation device 7 on each client side is connected to the inspection device 2. Each inspection device 2 includes an inspection unit 2a, a control device 4, an information processing device 5, and a waste liquid chamber 6.

The dielectric particle type estimation system 1 captures images of dielectric particles such as bacteria obtained by each inspection device 2 that inspects bacteria and the like by utilizing the dielectrophoresis of bacteria and cells in the sample solution (sample). By accumulating and learning, it has a function of estimating the type of bacteria or the like photographed by the inspection device 2.

Each inspection unit 2a is controlled by the control device 4, and the state in which the inspection target such as bacteria is collected is displayed on the information processing apparatus 5.

The inspection unit 2a includes a pump unit 10, an AC voltage supply unit 11, an image pickup unit 12, and a collection unit 3.

The pump unit 10 is composed of, for example, a syringe pump, and includes a drive unit 10a and a sample syringe 10b. The drive unit 10a is configured to include a motor or the like, and is driven and controlled by the control device 4.

The sample syringe 10b is a syringe that holds the sample liquid. A collection unit 3 is connected to the liquid feeding portion of the sample syringe 10b. In the pump unit 10, the sample liquid is appropriately set from the sample syringe 10b by the drive control of the drive unit 10a, and the flow velocity and the flow rate are appropriately set, and the liquid is sent to the collection unit 3. Although the sample is described here as flowing from the syringe side to the waste liquid side, for example, the sample is set on the waste liquid side and operated in the direction of sucking up the syringe to flow the sample from the waste liquid side toward the syringe. , The desired dielectric particles may be captured.

FIG. 4 is a diagram for explaining the collection unit 3 in the embodiment according to the present invention. The collection unit 3 has a flow path 13 through which the sample liquid flows in a predetermined direction (liquid flow direction), and a microelectrode portion 30 provided in the flow path 13. The sample liquid sent from the pump unit 10 passes through the flow path 13 in the collection unit 3 and is discharged into the waste liquid chamber 6. The micro electrode portion 30 is composed of a group of electrodes formed on a micro order.

In the collection unit 3, when the sample liquid is flowing on the microelectrode section 30 in the flow path 13, a predetermined AC voltage is supplied from the AC voltage supply section 11 to the microelectrode section 30. As a result, bacteria and the like to be inspected in the sample liquid undergo dielectrophoresis and are collected by the microelectrode unit 30. The collection unit 3 is an example of a dielectric collection unit that collects dielectric particles such as bacteria in the microelectrode unit 30. The details of the configuration of the collection unit 3 will be described later.

The AC voltage supply unit 11 is composed of, for example, a function generator. The AC voltage supply unit 11 generates an AC voltage having a desired frequency and voltage amplitude under the control of the control device 4, and supplies the AC voltage to the microelectrode unit 30 of the collection unit 3.

The image pickup unit 12 includes an image pickup element such as a CCD image sensor or a CMOS image sensor. The observation method using the image pickup unit 12 is referred to here as a “lens body focusing observation method”.

As shown in FIG. 3, the image pickup unit 12 includes a light source unit 50, a ring slit unit 52, a condensing cylinder unit 54, a lens body unit 56, and an objective lens unit 58.

The light source unit 50 emits illumination light toward the dielectric particles in the slit region Rs, which will be described later.

The ring slit portion 52 is provided between the light source portion 50 and the dielectric particles in the slit region Rs, and makes the illumination light into a ring-shaped illumination light.

The condensing tube portion 54 is provided between the ring slit portion 52 and the dielectric particles in the slit region Rs, collects the ring-shaped illumination light, and emits the illumination light to the lens body portion 56.

The lens body portion 56 is a dielectric particle of the slit region Rs of the collection unit 3, and aggregates the straight light and the diffracted light. The details of the lens body portion 56 will be described later.

The objective lens unit 58 passes the straight light and the diffracted light from the lens body portion 56 to form the image 59.

Here, in order to explain the "lens body focusing observation method", the "modulated contrast observation method" will be described as a comparative example. The modulated contrast observation method is an observation method suitable for observing colorless and transparent cultured cells with good contrast. The modulated contrast observation method is a method of adding contrast to an image by illuminating a sample with oblique illumination light that has passed through a slit and passing a light beam having a refractive index gradient through a "modulation plate" provided on the image forming side. .. If it is a retardation object, it is possible to observe it because the image has contrast even if the specimen is not a lens body.

On the other hand, the "lens body focusing observation method" does not use a modulation plate. Here, the lens body portion 56 refers to a three-dimensional spherical object that refracts light like a lens among phase objects. For example, microorganisms, cells, platelets and the like correspond to it, and the dielectric particles in the slit region Rs function as the lens body portion 56.

As shown in FIG. 3, the oblique illumination light (diffracted light) that has passed through the ring slit portion 52 is amplified by the condensing tube portion 54. The diffracted light collected by the condensing tube portion 54 passes through the sample from various angles, but when the sample is a phase object other than the lens body portion 56, it is only transparent and not visualized as in the case of bright field observation.

Also, if it is not a phase object, light does not pass through it, so it looks black as an image. Only when the phase object is the lens body 56, the aggregated diffracted light is condensed as it passes through the lens body 56 and becomes brighter than the periphery, so that the image appears white.

Since the light is most concentrated in the center of the lens, for example, in the case of a microorganism, the center of a single cell is brighter than the surroundings, and the vicinity of the outer cell wall is darker than the surroundings. This makes it possible to visualize a phase object specialized for the lens body 56 such as a microorganism.

As described above, according to the image pickup unit 12, the direct light and the diffracted light are aggregated and strengthened by the lens body portion 56, so that the sample which is the lens body portion 56 among the phase objects becomes particularly brighter than the background. Further, since the surroundings look particularly dark, it is suitable for observing single cells such as microorganisms, and further, it is suitable for noise processing by the image processing unit 60 and extraction and learning of feature quantities by the learning unit 72 and the like. ing.

The image pickup unit 12 takes an image of a predetermined region in the microelectrode unit 30 of the collection unit 3 and outputs the captured image to the information processing apparatus 5. The image pickup operation of the image pickup unit 12 may be controlled by the control device 4 or may be controlled by the information processing device 5.

The control device 4 controls the operation of the inspection unit 2a, such as feeding the sample liquid by the pump unit 10 and supplying the AC voltage by the AC voltage supply unit 11. The control device 4 includes an internal memory such as a flash memory, and realizes various functions by performing arithmetic processing using various data and the like based on a program stored in the internal memory.

The control device 4 may be configured by a hardware circuit (ASIC, FPGA, etc.) such as an electronic circuit specially designed or a reconfigurable electronic circuit. The function of the control device 4 may be realized by the cooperation of the hardware and the software, or may be realized only by the hardware (electronic circuit).

The information processing device 5 is composed of, for example, a personal computer. The information processing device 5 includes a liquid crystal display and an organic EL display (display unit), and displays an image captured by the image pickup unit 12.

The information processing apparatus 5 includes an internal memory such as a flash memory, and realizes various functions based on a program stored in the internal memory. For example, the information processing apparatus 5 has a function of performing image analysis of the captured image of the imaging unit 12 and measuring the number of objects in a region (slit) corresponding to a predetermined condition in the captured image.

The information processing device 5 may control the imaging operation of the imaging unit 12. Further, the information processing device 5 and the control device 4 may be integrally configured by realizing various functions of the control device 4 in the information processing device 5. The information processing apparatus 5 is an example of the display unit in the present embodiment, and is also an example of the image analysis unit that analyzes the image pickup result of the image pickup unit 12.

The waste liquid chamber 6 is a chamber for storing the sample liquid that has passed through the collection unit 3 of the inspection unit 2a. The waste liquid chamber 6 may be incorporated inside the inspection device 2a.

The configuration of the collection unit 3 will be described with reference to FIGS. 4 to 6. FIG. 4A shows a plan view of the collection unit 3 in the embodiment of the present invention. FIG. 4B shows a sectional view taken along line AA of the collection unit 3 shown in FIG. 4A. As shown in FIG. 4A, the collection unit 3 has a substantially rectangular flat plate shape.

Further, as shown in FIG. 4B, the collection unit 3 includes a cover plate 31, a spacer 32, and an electrode film 33, which are sequentially superposed in the thickness direction.

FIG. 4C shows a plan view of the cover plate 31 in the collection unit 3. FIG. 4D shows a plan view of the spacer 32. FIG. 4 (e) shows a plan view of the electrode film 33.

The cover plate 31 is a flat plate-shaped member formed of, for example, a transparent acrylic plate. As shown in FIG. 4C, the cover plate 31 is provided with two insertion holes 31a and 31b and a notch 31c.

The insertion holes 31a and 31b correspond to the start point and the end point of the flow path 13 in the collection unit 3, respectively (see FIG. 4A). The cutout portion 31c is formed in the cover plate 31 at a position corresponding to the electrode pad 33a on the electrode film 33.

The spacer 32 is, for example, a member made of transparent PET (polyester) tape. The spacer 32 is formed with a rectangular hole 32a corresponding to the flow path 13 and a notch portion 32b having the same shape as the notch portion 31c of the cover plate 31.

The spacer 32 is adhered to the cover plate 31 and the electrode film 33 on each main surface by an adhesive such as 3M (registered trademark) 9996, and the distance between the cover plate 31 and the electrode film 33 (that is, the height of the flow path 13) is high. Is fixed to a predetermined width (for example, 0.1 mm).

The electrode film 33 is a member provided with a microelectrode portion 30 on a transparent film base material such as a PEN (polyethylene naphthalate) film. The micro electrode portion 30 is electrically connected to the electrode pad 33a in the wiring region on the main surface of the electrode film 33. The micro electrode portion 30 and the electrode pad 33a are formed of a metal material such as chromium by, for example, a vapor deposition method or a sputtering method.

The collection unit 3 is electrically connected to the AC voltage supply unit 11 in the electrode pad 33a (see FIG. 2). As shown in FIG. 4A, the electrode pad 33a is exposed by the cutout portions 31c and 32b in a state where the cover plate 31, the spacer 32 and the electrode film 33 are overlapped with each other. Therefore, the collection unit 3 can be easily electrically connected to the AC voltage supply unit 11.

Further, the flow path 13 of the collection unit 3 is formed by the spacer 32 in close contact between the cover plate 31 and the electrode film 33 at predetermined intervals around the rectangular hole 32a. By connecting the sample syringe 10b and the waste liquid chamber 6 to the two insertion holes 31a and 31b located at both ends of the flow path 13 so as to be removable, the sample liquid can be easily passed through the flow path 13. As described above, the collection unit 3 can be easily connected to an electric machine or a flow path, and can be easily thrown away or reused after collecting bacteria and the like in the inspection unit 2a.

FIG. 5 is an enlarged view of a wiring region in the electrode film 33 of FIG. 4 (e) in the embodiment of the present invention. In the present embodiment, the microelectrode portion 30 in the electrode film 33 has two sets of electrode pairs CH1 and CH2. The AC voltage from the AC voltage supply unit 11 is supplied to each of the first and second electrode pairs CH1 and CH2 via the electrode pad 33a. The first and second electrode pairs CH1 and CH2 are formed line-symmetrically at the center line L1. Hereinafter, the first electrode pair CH1 will be described.

The first electrode pair CH1 consists of two electrodes 41 and 42. The electrodes 41 and 42 each have a comb tooth shape arranged at equal intervals. The plurality of protrusions in the comb-teeth shape of the two electrodes 41 and 42 are arranged alternately at predetermined intervals in the liquid flow direction of the flow path 13. Each of the convex portions of the electrodes 41 and 42 extends in a direction intersecting (orthogonal) with the liquid flow direction. The arrangement is the same for the electrodes 41 and 42 of the second electrode pair CH2.

FIG. 6 is an enlarged view of the microelectrode portion 30 in the region Ri near the end of the flow path 13 in the embodiment of the present invention. In the micro electrode portion 30, slit-shaped regions Rs having a predetermined width W2 are formed by slits between the convex portions of the electrodes 41 and 42 on the flow path 13.

As shown in FIG. 6, the plurality of slit regions Rs are arranged in the liquid flow direction between the pair of electrodes 41 and 42. The width W2 of each slit region Rs is set to a predetermined value in, for example, 10 μm to 20 μm. On the other hand, the width W1 of the convex portion of the electrodes 41 and 42 is, for example, 100 μm. The width W2 of the slit region Rs may be set within the range of 0.1 μm to 100 μm.

Further, in the present embodiment, the micro electrode portion 30 and the flow path 13 are set so that the convex portions of the electrodes 41 and 42 protrude from the flow path 13 by a predetermined length Δd, respectively. In other words, the region connecting the plurality of slit regions Rs between the pair of electrodes 41 and 42 on the electrode film 33 is arranged outside the flow path 13.

As a result, the plurality of slit regions Rs arranged in a predetermined direction (liquid flow direction) in which the flow path 13 extends are separated from each other without being connected in the flow path 13. For example, the width W3 (see FIG. 5) of the flow path 13 is 3 mm, and the length Δd protruding from both ends of the electrodes 41 and 42 is set to 0.3 mm. The thickness of the convex portions of the electrodes 41 and 42 is, for example, about 100 nm.

Further, when performing dielectrophoresis of bacteria or the like in the inspection unit 2a, as shown in FIG. 5, an image is taken of a region Rc in which the convex portions of the electrodes 41 and 42 are sequentially arranged from the upstream of the flow path 13 in the micro electrode unit 30. Image is taken by the unit 12.

Here, when the slit regions Rs are connected to each other in the flow path 13, for example, bacteria collected in the slit region Rs on the upstream side are downstream while maintaining the dielectrophoretic force between the electrodes 41 and 42. A situation is assumed in which the slit region Rs on the side is moved. On the other hand, by separating each slit Rs region in the flow path 13 as described above, it is possible to suppress the movement of collected bacteria or the like between the plurality of slit regions Rs.

FIG. 7 is a diagram showing the principle of the inspection method in the embodiment of the present invention. The inspection unit 2a uses dielectrophoresis to collect bacteria and the like to be inspected contained in the sample solution. As shown in FIG. 7A, when an AC voltage having a frequency ω is supplied between the electrodes 41 and 42, the dielectrophoretic force acting on bacteria such as live bacteria and dead bacteria in the sample liquid flowing through the flow path 13. F _DEP is expressed by the following equation.
F _DEP = 2πr ³ ε _m Re [K (ω)] ∇E ² … (1)

In the above equation (1), r is the radius of the dielectric particles such as live and dead bacteria to be inspected, ε _m is the permittivity of the medium of the sample liquid, and E is the strength of the electric field. Re [X] represents the real part of the complex number X. K (ω) is a Clausius-Mossotti factor and is expressed by the following equation.
K (ω) = (ε _p ^* -ε _m ^* ) / (ε _p ^* + 2ε _m ^* )… (2)

In the above equation (2), ε _p ^* (= ε _p + ρ _p / (jω)) is the complex permittivity of the dielectric particles (ε _p is the permittivity of the dielectric particles, and ρ _p is its conductivity. ). Further, ε _m ^* (= ε _m + ρ _m / (jω)) is the complex permittivity of the medium (ρ _m is the conductivity of the medium).

In the above equation (1), when Re [K (ω)]> 0, a positive dielectrophoretic force _FDEP acts on the dielectric particles with respect to the installation direction of the electrodes 41 and 42, and the dielectric particles are the electrodes. It is attracted to the vicinity of 41 and 42 and is attracted to the slits Rs.

On the other hand, when Re [K (ω)] <0, a negative dielectrophoretic force _FDEP acts on the dielectric particles, and the dielectric particles repel the electrodes 41 and 42. Therefore, by appropriately setting the frequency ω, it is possible to selectively adsorb the inspection target to the slit region Rs while eliminating impurities and the like that are not the inspection target.

Next, the client-side dielectric particle estimation device 7 connected to the plurality of inspection devices 2 will be described. The client-side dielectric particle estimation device 7 includes an image processing unit 60, a shape classification unit 62, a growth environment analysis unit 64, an in-shape classification unit 66, a special classification unit 68, and an estimation unit 70. ..

The image processing unit 60 has a function of performing image processing for removing noise from the image captured by the image pickup unit 12. The image processing unit 60 removes a large amount of noise generated by various factors from the image containing the dielectric particles collected by the electrodes 41 and 42 by image processing, and enables more accurate extraction (single extraction). And.

As a parameter for determining whether or not noise is present, for example, it can be determined by luminance, and an image in which noise is removed (right figure in FIG. 8) is formed in the image shown in the left figure of FIG. ..

The shape classification unit 62 classifies the dielectric particles into the first category roughly classified into the types of bacteria based on the shape of the dielectric particles contained in the image processed by the image processing unit 60. Narrow down to one type. The first category, which is a rough type of fungus, is composed of four types: short bacillus type, long bacillus type, fungal type, and small cocci type. For example, the example shown in FIG. 10 shows a procedure for identifying the bacterial species A from the bacteria A to F, and is first narrowed down to a short bacillus form including A, B, and C.

After being classified by the shape classification unit 62, the growth environment analysis unit 64 classifies the dielectric particles into the second category, which is a subordinate concept of the first category, based on the growth environment, and further narrows down the particles. The second category, which is divided by habitat, is composed of aerobic, microaerophile, facultative anaerobic, anaerobic, and obligate anaerobic. For example, in the example shown in FIG. 10, among the short bacillus forms A to C, they are narrowed down to A and B, which are aerobic growth environments.

After being classified by the growth environment analysis unit 64, the classification unit 66 in the same shape performs image analysis of dielectric particles having substantially the same shape in the second category to extract a difference in the feature amount, and based on the difference. It is classified into the third category of the subordinate concept of the second category, and further narrowed down in detail. For example, in the example shown in FIG. 10, among the aerobic short rod bacilli (A, B), when the probability that A is 90% and B is 20% as a discrimination result is high, it is high. , The result of narrowing down to A is output as it is.

When the estimation unit 70 cannot be estimated based on the data classified in the third category, the special classification unit 68 analyzes only similar bacterial species in the third category and extracts the difference in the feature amount. , Classify into the 4th category of the subordinate concept of the 3rd category based on the difference. For example, in the example shown in FIG. 10, when the comparison result in the classification unit 66 in the same shape is, for example, 50% for A and 50% for B, the special classification unit 68 is executed. For example, as shown in FIG. 11, it can be discriminated by performing classification by adhesion pattern, classification by chain case, and the like.

The estimation unit 70 estimates the type of dielectric particles based on the results of classification and comparison by the classification unit 66 in the same shape or the special classification unit 68. The estimation unit 70 estimates a bacterial species from a group of dielectric particles specified in advance.

The dielectric particle estimation device 8 on the server side includes an image processing unit 60, a shape classification unit 62, a growth environment analysis unit 64, an in-shape classification unit 66, a special classification unit 68, and an estimation unit 71. A unit 72 and a simple test unit 74 are provided. Here, since the image processing unit 60, the shape classification unit 62, the growth environment analysis unit 64, the in-shape classification unit 66, and the special classification unit 68 are the same as the dielectric particle estimation device 7 on the client side, they are detailed. The explanation is omitted.

The learning unit 72 has a function of learning images of dielectric particles acquired from a plurality of inspection devices 2 and improving the estimation accuracy of the type of bacteria.

The learning unit 72 extracts the feature amount of the dielectric particles from the image processed by the image processing unit 60. It outputs necessary information such as "area", "average brightness", "major axis / minor axis", and "roundness" as the bacterial feature amount, and makes it possible to measure the statistics and trends of the feature amount. For example, it is possible to extract the feature amount of the image shown in the left figure of FIG. 9 and obtain the statistics shown in the right figure of FIG.

The learning unit 72 performs deep learning based on the features of the plurality of dielectric particles. Specifically, the type of bacteria to be discriminated can be determined by performing noise processing on a large amount of images taken by the image processing unit 72 and learning / analyzing using a large amount of information from which the feature amount has been extracted. You will be able to estimate.

The estimation unit 71 estimates the type of dielectric particles based on the results of classification and comparison by the classification unit 66 in the same shape or the special classification unit 68. Unlike the estimation unit 70 of the dielectric swarming device 7 on the client side, the estimation unit 71 estimates the bacterial species from an unspecified dielectric particle group. That is, the learning unit 72 estimates from the latest data in which deep learning has been performed.

Here, the learning unit 72 can be realized by using CNN (Convolutional Neural Network) of artificial intelligence (AI), and loses the respective positional information and positional relationship with respect to the input image by, for example, convolution and pooling. The feature amount is extracted without any problem, converted into data, classified and output.

Here, "convolution" means using a filter (kernel filter) to convert image data into a feature map without damaging the position information. "Pooling" means to roughly organize the features. That is, in the image input by using CNN, the features are found by convolution, the features are roughly arranged by pooling, and the features are collectively evaluated by honey binding.

As described above, using artificial intelligence (AI), the image of the bacteria tested, the characteristics of the sample tested (objects (eg udon, retort curry, lotion, etc.)), pH (eg 6-7, 3-4, etc.), water activity value (for example, 80% or more, less than 40%, etc.), storage temperature (for example, 10 ° C or less, around 35 ° C, around 50 ° C), and learning while performing analysis. This makes it possible to estimate the bacteria more accurately.

The simple test unit 74 tests the dielectric particle estimation function of the dielectric particle estimation device 7 on the client side and the dielectric particle estimation device 8 on the server side. For example, as shown in FIG. 12A, a learning image and a test image are photographed and collected under the same conditions.

As shown in FIG. 12 (b), a learning model is generated by adjusting the batch size and the number of epochs of the training image, and the learning model created here is used in each test image to determine the conformance rate in FIG. 13 (a). Test as shown in. Whether or not self-other discrimination is possible by the above conformance rate test is calculated as what percentage for each test image. A test is carried out so that the result of the self-other discrimination shows a clear difference as shown in FIG. 13 (b) and is passed.

Subsequently, the operation of the dielectric particle type estimation system 1 having the above configuration will be described. First, the operation of each inspection device 2 connected to the client-side dielectric particle estimation device 7 via the network 9 will be described, and then the dielectric particle type estimation system including the client-side dielectric particle estimation device 7 will be described. 1 will be described.

In each inspection device 2, the inspection section 2a is controlled by the control device 4, and the microelectrode section in the flow path 13 from the AC voltage supply section 11 while flowing the sample liquid from the pump section 10 to the flow path 13 of the collection unit 3. An AC voltage having a predetermined frequency is supplied to 30, and a positive dielectrophoretic force is applied to the inspection target.

Next, the region Rc in which the slit regions Rs are lined up is imaged in the microelectrode unit 30. At this time, the frequency ω of the AC voltage is set so that the positive dielectrophoretic force acts according to the inspection target, and the voltage amplitude of the AC voltage and the flow velocity in the flow path 13 are appropriately controlled. This makes it possible to selectively collect various inspection targets.

For example, by frequency control, it is possible to switch whether or not to distinguish between live bacteria (active bacteria) and dead bacteria (damaged bacteria) in bacteria. FIG. 7A shows an example in which both live and dead bacteria are collected. For example, by supplying an AC voltage to the electrodes 41 and 42 at a frequency of ω = 100 MHz, a positive dielectrophoretic force is applied to both live and dead bacteria, and both live and dead bacteria are subjected to other. It can be collected separately from the ones.

FIG. 7B shows an example in which live bacteria are selectively collected. In the example shown in FIG. 7 (b), as shown in FIG. 7 (a), the frequency ω is increased to 3 MHz after the operation at the frequency ω = 100 MHz. Then, while the positive dielectrophoretic force acts on the live bacteria, it does not act on the dead bacteria. As a result, only live bacteria excluding dead bacteria can be adsorbed on the slit region Rs.

The inspection device 2 uses a "lens body focusing observation method", which is different from the phase difference observation method which is a general method. The lens body condensing observation method utilizes the fact that the condenser lens is eliminated and the "diffracted light" and "straight light" of the illumination light are aggregated and appear brighter than the background only when the phase object is the lens body. This is a method of observing only the lens body 56 of an object with a contrast of light and dark.

Further, in the inspection device 2, there is a ring slit portion 52 between the condensing cylinder portion 54 and the light source portion 50, and the light in which the illumination light is formed into a ring is applied to the sample. Since the straight light and the diffracted light are aggregated and strengthened by the lens body 56, the sample which is the lens body 56 among the phase objects becomes particularly brighter than the background. In addition, since the surrounding area looks particularly dark, it is suitable for observing single cells such as microorganisms, and is suitable for accumulating and learning a large amount of images.

The dielectric particle estimation device 7 on the client side can perform image processing on a large amount of images taken by the image processing unit 60 to discriminate the bacteria to be discriminated. This has a remarkable effect that the bacteria can be estimated with higher accuracy. Further, the dielectric particle estimation device 7 on the client side is limited to estimating from the bacterial species specified in advance, but a wider range can be obtained by using the dielectric particle estimation device 8 on the server side as well. It has a remarkable effect that the bacterial species can be estimated in.

1 Dielectric particle type estimation system, 2 Inspection device, 2a Inspection unit, 3 Collection unit, 4 Control device, 5 Information processing device, 6 Waste liquid chamber, 7 Client side dielectric particle estimation device, 8 Server side dielectric Particle estimation device, 9 network, 10 pump section, 10a drive section, 10b sample syringe, 11 AC voltage supply section, 12 imaging section, 13 flow path, 30 microelectrode section, 31 cover plate, 31a, 31b insertion hole, 31c cut Notch, 32 spacer, 32a rectangular hole, 32b notch, 33 electrode film, 33a electrode pad, 41, 42 electrodes, 50 light source, 52 ring slit, 54 condenser tube, 56 lens body, 58 objectives Lens unit, 59 images, 60 image processing unit, 62 shape classification unit, 64 growth environment analysis unit, 66 same shape classification unit, 66 dielectric particle estimation unit, 68 special classification unit, 70 estimation unit, 71 estimation unit, 72 Learning department, 74 simple test department.

Claims (7)

A dielectric particle estimation device having a dielectric particle estimation unit that acquires an image of the dielectric particles captured by causing dielectric migration of the dielectric particles contained in the sample liquid and estimates the type of the dielectric particles. And,
The dielectric particle estimation unit is
An image processing unit that removes noise from the image captured by the image pickup unit and performs image processing to recognize and extract the corresponding portion of the dielectric particles.
A shape classification unit that classifies the dielectric particles into a first category that is roughly classified based on the shape of the dielectric particles contained in the image that has been subjected to the image processing.
After being classified by the shape classification unit, the growth environment analysis unit further classifies the dielectric particles into the second category based on the growth environment.
After being classified by the growth environment analysis unit, the image analysis of the dielectric particles having substantially the same shape in the second category is performed to extract the difference in the feature amount, and based on the difference, the subordinate of the second category. A classification unit within the same shape that classifies into the third category of the concept,
An estimation unit that estimates the type of the dielectric particles based on the data classified by the classification unit within the same shape, and an estimation unit.
A dielectric particle estimator comprising. In the dielectric particle estimation device according to claim 1,
When the estimation unit cannot be estimated based on the data classified in the third category, only similar bacterial species in the third category are image-analyzed to extract the difference in the feature amount, and the difference is determined. It is equipped with a special classification unit that classifies into the fourth category of the subordinate concept of the third category based on the above.
The estimation unit is a dielectric particle estimation device characterized by estimating the type of the dielectric particles based on the data classified by the special classification unit. In the dielectric particle estimation device according to claim 1 or 2.
Dielectric trapping having at least a pair of electrodes, a flow path extending in a predetermined direction on the pair of electrodes, and a plurality of slit regions between the pair of electrodes that are aligned in the predetermined direction and separated from each other. Shubu and
A pump unit that feeds the sample liquid so as to travel in the predetermined direction in the flow path, and a pump unit.
An AC voltage supply unit that supplies an AC voltage of a predetermined frequency to the pair of electrodes so that the dielectric particles in the sample liquid to be sent cause dielectrophoresis.
In the dielectric collecting unit, an image pickup unit that images a predetermined region in which the plurality of slit regions are lined up, and an image pickup unit.
A dielectric particle estimator comprising. In the dielectric particle estimation device according to claim 3,
The image pickup unit is
A light source unit that outputs illumination light toward the dielectric particles provided in the plurality of slit regions, and a light source unit.
A ring slit portion for converting the illumination light output from the light source portion into a ring-shaped illumination light, and a ring slit portion.
A condensing tube portion for condensing the ring-shaped illumination light formed by the ring slit portion, and a condensing tube portion.
A lens body portion formed by aggregating the straight light and diffracted light collected by the condensing cylinder portion and including the dielectric particles, and a lens body portion.
An objective lens unit for forming an image by passing the straight light and the diffracted light aggregated by the lens body unit.
A dielectric particle estimator characterized by having. In the dielectric particle estimation device according to claim 3 or 4.
A dielectric particle estimation device characterized in that a region connecting the plurality of slit regions between the pair of electrodes is arranged outside the flow path. The dielectric particle estimation device according to any one of claims 3 to 5.
The dielectric particle estimator comprising at least one of a bacterium, a cell, a microorganism, a fungus, a spore, and a virus. A plurality of dielectric particle estimation devices according to any one of claims 3 to 6 are provided.
The plurality of dielectric particle estimators include a plurality of client-side dielectric particle estimators, a server-side dielectric particle estimator connected to the plurality of client-side dielectric particle estimators via a network, and the like. Including
The server-side dielectric particle estimator is
Deep learning is performed based on the feature amounts of a plurality of the dielectric particles extracted based on the dielectric particle estimation unit that extracts the feature amount of the dielectric particles from the image processed by the image processing unit. Has a learning department to do
The dielectric particle estimation device on each client side estimates the type of dielectric particles from the bacterial species specified in advance, and determines the type of dielectric particles.
The server-side dielectric particle estimation device is a dielectric particle estimation system characterized in that it estimates the type of dielectric particles from among bacterial species other than the specified bacterial species.

Images