CN116490812A

CN116490812A - Super-resolution detection system and super-resolution detection method

Info

Publication number: CN116490812A
Application number: CN202080107055.6A
Authority: CN
Inventors: 伯恩
Original assignee: MGI Tech Co Ltd
Current assignee: MGI Tech Co Ltd
Priority date: 2020-12-11
Filing date: 2020-12-11
Publication date: 2023-07-25
Also published as: WO2022120853A1

Abstract

A super-resolution detection system (10) and a super-resolution detection method, the super-resolution detection system (10) includes: a light source module (11) for emitting light source light; a light modulator (121) comprising a plurality of micro-mirrors (1211), wherein the micro-mirrors (1211) are divided into a plurality of modulation units, each modulation unit comprises two micro-mirrors (1211), a light modulation module (12) is used for modulating a light source into structural light to be emitted, the structural light can be guided to a sample (20) to be detected so that the sample (20) to be detected emits detection light, the structural light forms stripe light spots on the sample (20) to be detected, and each modulation unit corresponds to a stripe period; an imaging module (13) for acquiring a fringe image based on the detection light and for acquiring a wide-field image; and a controller (14) electrically connected with the light modulator (121) and the imaging module (13) for adjusting the phase of the stripes formed by the structured light and performing super-resolution reconstruction according to the stripe images and the wide-field images, thereby obtaining super-resolution images to obtain biological information of the sample (20) to be detected.

Description

Super-resolution detection system and super-resolution detection method

Technical Field

The invention relates to the field of biochemical information detection, in particular to a super-resolution detection system and a super-resolution detection method.

Background

The gene sequencing technology is a technology for analyzing the sequence of four bases on DNA, and has been widely applied to various research fields of life science and medicine so far, including various genomics, etiology of complex diseases, prenatal diagnosis, drug individuation treatment and the like. The basic method of gene sequencing is as follows: the four bases carry corresponding fluorescent groups through a biochemical method, and the fluorescent groups emit fluorescence with different wavelengths after being excited by lasers with different wavelengths, and the types of the bases are identified through the fluorescence, so that sequencing is realized.

Structured light illumination microscopy (structured illumination microscopy, SIM) technology is a typical wide-field imaging technique, suitable for use in high-fraction fluorescence microscopy imaging systems of genetic sequencers. The implementation of an optical hardware system of the SIM technology is mainly divided based on the adopted devices, and typical devices are as follows: grating-SIM, spatial light modulator (SLM-SIM), digital micromirror device (DMD-SIM), etc. The digital micromirror device (Digital Micromirror Device, DMD) is a Micro-electro-mechanical-system (MEMS) composed of an array of micromirrors coated with highly reflective aluminum films on the surface. A single micromirror is called a Pixel cell (Pixel) that has two states: ON and OFF. The ON state corresponds to the high reflection state and the OFF state corresponds to the no reflection state. The ON and OFF states are achieved by controlling the rotation angle of the mirror (the most common system is a 12 degree deflection angle). The DMD-SIM adopts an electric control mode to realize stripe projection in X/Y directions. The DMD achieves high speed directional switching and fringe phase shifting by switching the micromirrors.

In the traditional mode of adopting the DMD-SIM, more stripe images are required to be acquired, and the time for image acquisition is long, so that the time for processing the images is also prolonged.

Disclosure of Invention

In one aspect, the present invention provides a super-resolution detection system for detecting biological information of a sample to be detected, the super-resolution detection system comprising:

the light source module is used for emitting light source light;

the light modulator comprises a plurality of micro-reflectors, wherein the micro-reflectors are divided into a plurality of modulation units, each modulation unit comprises two micro-reflectors, the light modulation module is used for modulating the light source into structural light to be emitted, the structural light can be guided to the sample to be detected so as to enable the sample to be detected to emit detection light, the structural light forms a stripe light spot on the sample to be detected, and each modulation unit corresponds to a stripe period;

the imaging module is used for acquiring a fringe image according to the detection light and acquiring a wide-field image; and

and the controller is electrically connected with the light modulator and the imaging module, and is used for adjusting the phase of the stripe formed by the structured light and carrying out super-resolution reconstruction according to the stripe image and the wide-field image so as to acquire a super-resolution image and acquire biological information of the sample to be detected.

In another aspect, the present invention provides a super-resolution detection method for detecting biological information of a sample to be detected, the super-resolution detection method including the steps of:

generating structured light, scanning a sample to be detected in a two-step phase shift mode, acquiring a first stripe image and a second stripe image in a first direction, and acquiring the first stripe image in a second direction;

acquiring a wide-field image of the sample to be detected;

setting a preset evaluation index, and acquiring a super-resolution image according to the first stripe image, the second stripe image and the wide field image in an iterative mode so as to acquire biological information of the sample to be detected.

The super-resolution detection system and the super-resolution detection method adopt a two-step phase shift mode, are beneficial to improving the fringe density, increasing the field area and improving the sequencing flux. The scanning mode of two-step phase shift is combined with the super-resolution reconstruction algorithm adopting the iterative mode, so that super-resolution reconstruction is realized on the basis of less acquired images, the image acquisition time is shortened, the processing time of the images is shortened, and the detection cost of biological information is saved.

Drawings

Fig. 1 is a schematic structural diagram of a super-resolution detection system, a sample to be detected and a sequencing chip according to the present embodiment.

Fig. 2 is a schematic diagram of an optical path structure of the super-resolution detection system in fig. 1.

Fig. 3 is a schematic plan view of the optical modulator of fig. 2.

Fig. 4 is a schematic view showing a deflection state of the micromirror in fig. 2.

FIG. 5 is a schematic diagram of a combination of deflection states of micromirrors in each of the modulator cells of the optical modulator of FIG. 2.

FIG. 6 is another schematic diagram of a combination of deflection states of the micromirrors in each of the modulator cells of the optical modulator of FIG. 2.

Fig. 7 is a flowchart of a super-resolution detection method provided in the present embodiment.

Fig. 8 is a detailed flowchart of step S3 in fig. 7.

Fig. 9 is a schematic diagram of a verification process based on a binary bitmap as a reference image.

FIG. 10 is a schematic diagram of a verification process based on a fluorescence image of a sequencing chip as a reference image.

Description of the main reference signs

Super-resolution detection system 10

Light source module 11

Laser 111

Mirror 112

Dichroic mirror 113

First lens 114

Second lens 115

Light modulation module 12

Optical modulator 121

Micro mirror 1211

Total internal reflection mirror 122

Multiple adjusting assembly 123

Third lens 1231

Fourth lens 1232

Imaging module 13

First dichroic mirror 131

Second dichroic mirror 132

Optical filter 133

Fifth lens 134

Imaging device 135

Sample to be measured 20

Sequencing chip 30

Steps S1, S2, S3, S31, S32, S33, S34, S35

The invention will be further described in the following detailed description in conjunction with the above-described figures.

Detailed Description

Referring to fig. 1, a super-resolution detection system 10 according to an embodiment of the invention can be used for detecting biological information of a sample 20 to be detected. The sample 20 to be tested may be a nucleic acid sample (DNA or RNA), a protein, a cell, or the like. In this embodiment, the sample 20 to be measured is a nucleic acid sample, and the biological information may be base sequence information of the sample 20 to be measured.

The sample 20 to be tested is carried on the sequencing chip 30. During the operation of the super-resolution detection system 10, reference light is emitted to the sample 20 to be detected. The sequencing chip 30 and the super-resolution detection system 10 are movably arranged to generate relative motion, so that the sample 20 to be tested and the super-resolution detection system 10 can be driven to generate relative motion. In this embodiment, the sequencing chip 30 is disposed on a carrying platform, and the sequencing chip 30, the sample 20 to be tested and the super-resolution detection system 10 are moved by moving the carrying platform. That is, during the movement of the loading platform, the sequencing chip 30 and the sample 20 to be tested remain stationary. By providing for movement of the sample 20 to be measured relative to the super-resolution detection system 10, reference light can be projected onto different areas of the sample 20 to be measured, a process also referred to as "scanning". Since the field of view of the reference light on the sample 20 to be measured cannot cover the sample 20 to be measured completely, the super-resolution detection system 10 can scan the whole sample 20 to be measured by setting the sample 20 to be measured and the super-resolution detection system 10 to generate relative motion to move the sample 20 to be measured.

In this embodiment, different bases on the sample 20 to be measured are labeled with different fluorescent substances. When the reference light irradiates the sample 20 to be measured, different fluorescent substances are excited to generate fluorescent light of different wavelengths as detection light. The super-resolution detection system 10 is used for acquiring biological information of the sample 20 to be detected according to the detection light.

Referring to fig. 2, solid arrows in fig. 2 indicate the propagation direction of the laser light, and broken arrows indicate the propagation direction of the fluorescence light. The super-resolution detection system 10 comprises a light source module 11, a light modulation module 12, an imaging module 13 and a controller 14, wherein the light source module 11, the light modulation module 12 and the imaging module 13 are respectively and electrically connected with the controller 14.

The light source module 11 includes two lasers 111 for emitting first light and second light, respectively. In this embodiment, the first light and the second light have different wavelengths. For example, one of the two lasers 111 is used to emit red laser light and the other is used to emit green laser light. In other embodiments, the light source module 11 includes other numbers of lasers, and each laser is configured to emit laser light with a different wavelength. The number of lasers may depend on the type of fluorescent substance on the sample 20 to be measured.

The light source module 11 further includes a light combining component for combining the first light and the second light to form light source light. In this embodiment, the light combining component includes a reflecting mirror 112 and a dichroic mirror 113, wherein the laser light emitted from one laser 111 is incident on the dichroic mirror 113 and is transmitted by the dichroic mirror 113, and the laser light emitted from the other laser 111 is reflected by the reflecting mirror 112 to the dichroic mirror 113 and is reflected by the dichroic mirror 113. That is, the laser beams emitted from the two lasers 111 are combined by the dichroic mirror 113 and emitted as light source light.

The light source module 11 further includes a beam expanding assembly. The beam expanding assembly is located on the outgoing path of the light source light, and is used for performing beam expanding treatment on the light source light so as to meet the requirement of the subsequent light path on the field area, which is not described in detail here. In this embodiment, the beam expanding assembly includes a first lens 114 and a second lens 115. The first lens 114 and the second lens 115 cooperate to expand the light source light.

With continued reference to fig. 2, the optical modulation module 12 includes an all-in-one mirror 122, an optical modulator 121, and a multiple adjustment component 123. The total internal reflection mirror 122 is configured to receive the expanded light source light and to project the same to the light modulator 121. The light modulator 121 is used to modulate the received light to produce structured light. The total internal reflection mirror 122 is also used to guide the structured light to the imaging module 13.

Referring to fig. 3, the optical modulator 121 is a digital micromirror device (Digital Micromirror Device, DMD). The light modulator 121 includes a plurality of micromirrors 1211 arranged in the same plane. The plurality of micromirrors 1211 is arranged in a micromirror array comprising a plurality of rows and columns. In this embodiment, each micromirror 1211 is substantially rectangular, and the micromirror array on the light modulator 121 is substantially rectangular.

Referring to fig. 4, each micromirror 1211 can be deflected over a range of angles. In this embodiment, each micromirror 1211 is deflectable about an axis in two opposite directions, and the maximum deflection angles in the two opposite directions are the same, the maximum deflection angles in the two opposite directions being defined as α and- α. During operation of the super-resolution detection system 10, the state of each micro mirror 1211 is controlled by the controller 14 to be a deflection angle α or a deflection angle α. The "ON" state is defined when the deflection angle of each micromirror 1211 is α, and the "OFF" state is defined when the deflection angle of each micromirror 1211 is- α. By adjusting the state of each micromirror 1211 in the light modulator 121, the form and phase of the structured light emitted from the light modulator 121 can be adjusted.

In this embodiment, the structured light emitted from the light modulator 121 has a stripe shape. Structured light may be projected onto sequencing chip 30. When the structured light is projected onto the sequencing chip 30, a light spot including a plurality of parallel stripe shapes is formed on the surface of the sequencing chip 30.

The super-resolution detection system 10 operates in a plurality of detection cycles during operation. At each detection cycle, the super-resolution detection system 10 acquires a plurality of images of the sequencing chip 30. In the same detection period, a plurality of images acquired by the super-resolution detection system 10 are obtained from the structured light of different phases. In this embodiment, the process of changing the phase of the structured light in one detection period is defined as "phase shift".

In a super-resolution system under test, the sequencing chip 30 is photographed using a three-step phase shift detection scheme. That is, in one detection period, three kinds of structured light with different phases are respectively projected onto the sequencing chip 30 in two different directions, and photographing is respectively performed, so that three images can be obtained in each direction, that is, six images can be obtained in total. Biological information of the sample 20 to be tested on the sequencing chip 30 can be obtained according to the six images.

As described above, the plurality of micromirrors 1211 in the light modulator 121 are arranged in a micromirror array including a plurality of rows and a plurality of columns. In the super-resolution system under test, all the micromirrors 1211 in the optical modulator 121 are divided into a plurality of mutually independent modulation units, each of which includes four micromirrors 1211, and the four micromirrors 1211 in the same modulation unit are adjacently arranged. The four micro mirrors 1211 are arranged in the same row in each modulation unit, or the four micro mirrors 1211 are arranged in the same column in each modulation unit.

To achieve three-step phase shifting, the state ("ON" or "OFF") of each micromirror 1211 is controlled in a manner corresponding to one fringe period per modulation unit. That is, one fringe period is represented by four micromirrors 1211.

Referring to fig. 5, fig. 5 shows 8 micro-mirrors 1211 (i.e., two modulation units) to represent two fringe periods. First, by controlling the "ON" and "OFF" states of the respective micromirrors 1211, the structured light emitted from the light modulator 121 is formed into a binary stripe whose light intensity is in the shape of a square wave with a duty ratio of 50%. When the binary stripes are projected onto the surface of the sequencing chip 30 through the multiple adjusting component 123, the binary stripes are changed into sine stripes due to the fact that the numerical aperture of the optical element in the multiple adjusting component 123 is limited, and therefore the effect of low-pass filtering is achieved. Then, the sinusoidal stripes are projected onto the surface of the sequencing chip 30.

As shown in fig. 5, three types of combinations of states of the micromirrors 1211 are provided in each dimming cell: the "on-off" as shown in the (a) diagram, the corresponding fringe phase is 0; the switch is turned on as shown in the graph (b), and the corresponding fringe phase is pi/2; as shown in the graph (c) 'off Guan Kai on', the corresponding fringe phase is pi. The fringes shown in the figures (a), (b) and (c) are projected to the sequencing chip 30 in a time-sharing manner, so that three-step phase shift is realized. In a similar manner, three-step phase shift is realized in the row and column directions by the dimming device 121, and 6 images are acquired in total for the frequency domain super-resolution reconstruction method.

In the above manner, 6 images are acquired in each detection period, and super-resolution reconstruction is required according to the 6 images, which takes a long time. Specifically, the super-resolution detection system 10 provided by the embodiment of the invention adopts a two-step phase shift mode to acquire images and adopts an iterative mode to perform super-resolution reconstruction.

Referring to fig. 6, the present invention proposes a control method using two micro-mirrors 1211 to represent a fringe period, and there are two types of state combination methods of every adjacent four micro-mirrors 1211 (i.e. two adjacent modulation units): as shown in fig. 6 (a), the corresponding stripe phase is 0; the "off switch on" shown in the diagram (b) of fig. 6 corresponds to a stripe phase of pi, thus achieving a two-step phase shift.

Ideally, the period of the sinusoidal fringes projected onto the surface of the sequencing chip 30 should be equal to the resolution of the optical imaging system, which is generally denoted by Li Panju, namely: r=0.61 λ/NA; where λ denotes a wavelength and NA denotes a numerical aperture of the objective lens.

As shown in fig. 3, the number of micro mirrors 1211 in the light modulator 121 is: m N. The field of view obtained by the three-step phase shift method is:the field of view obtained with the two-step phase shift is:obviously, the FOV ₂ ＝4·FOV ₁ . That is, the two-step phase shift approach can increase the field of view by a factor of 4 compared to the three-step phase shift approach.

Referring to fig. 2 again, in the present embodiment, the magnification adjustment component 123 includes a third lens 1231 and a fourth lens 1232. The structural light emitted from the light modulator 121 is incident on the third lens 1231 through the total internal reflection mirror 122, and is then incident on the fourth lens 1232 through the third lens 1231. The third lens 1231 and the fourth lens 1232 are used to collectively adjust the stripe size of the structured light. The third lens 1231 has a focal length f ₃ The fourth lens 1232 has a focal length f ₄ The stripe size of the structured light can be adjusted by adjusting the ratio of the focal lengths. In this embodiment, the fourth lens 1232 is an objective lens. During operation of the super-resolution detection system 10, the focal length of the objective lens is generally kept uniform, so the third lens 1231 with different focal lengths is configured to adjust the focal length ratio between the third lens 1231 and the fourth lens 1232.

The dimensions of each micro mirror 1211 are defined as: Δx=Δy, since four micromirrors 1211 are used to represent a fringe period in the three-step phase shift mode, the fringe period should be equal to the resolution R of the optical imaging system, so the reduction multiple of the fringes formed by the structured light is:(this formula holds when Δx=Δy). In the two-step phase shift approach, two micro-mirrors 1211 are used insteadThe fringe period is equal to the resolution R of the optical imaging system, so the fringe period formed by the structured light is reduced by a factor of:(this formula holds when Δx=Δy). It can be seen that the fringe reduction factor is twice that of the three-step phase shift mode. The fringe density is twice that of the two-step phase shift mode.

On the basis, the stripe formed by the structured light in one detection period has two phases, and two images are required to be respectively acquired in the X and Y directions, so that the number of the acquired images is effectively reduced relative to the three-step phase shift mode, the detection time consumption is reduced, and the time consumption for processing the images is also reduced.

The imaging module 13 includes a first dichroic mirror 131, a second dichroic mirror 132, two filters 133, two fifth lenses 134, and two imaging devices 135.

The first dichroic mirror 131 is located between the third lens 1231 and the fourth lens 1232, transmits the structured light to the fourth lens 1232, and reflects the detection light to the second dichroic mirror 132. The second dichroic mirror 132 is configured to split the received detection light according to different wavelengths, and then guide the split detection light to the two imaging devices 135. One of the filters 133 and one of the fifth lenses 134 are located between the second dichroic mirror 132 and one of the imaging devices 135, and the other of the filters 133 and the other of the fifth lenses 134 are located between the second dichroic mirror 132 and the other of the imaging devices 135. The filter 133 and the fifth lens 134 are used to guide the detection light to the corresponding imaging device 135. The imaging device 135 is used for imaging the detection light generated by the fringe irradiation of the various phases in the aforementioned two-step phase shift method, respectively.

The controller 14 is used for controlling the deflection state of each micro mirror 1211 in the light modulator 121, so as to adjust the stripe shape and the phase of the structured light projected onto the sequencing chip 30, and performing super-resolution reconstruction on a plurality of images acquired by the imaging device 135, so as to acquire the biological information of the sample 20 to be tested.

The present embodiment also provides a super-resolution detection method for detecting biological information of the sample 20 to be detected, which is applied to the super-resolution detection system 10, and specifically, to the controller 14.

Referring to fig. 7, the super-resolution detection method includes the following steps:

step S1, generating structured light, scanning a sample to be detected in a two-step phase shift mode, obtaining a first stripe image and a second stripe image in a first direction, and obtaining the first stripe image in a second direction;

s2, acquiring a wide field image of a sample to be detected;

and step S3, setting a preset evaluation index, and acquiring a super-resolution image according to the first stripe image, the second stripe image and the wide field image in an iterative mode so as to acquire biological information of the sample to be detected.

The method for acquiring the first stripe image and the second stripe image in step S1 is described in detail above and will not be repeated.

Let the surface of the sequencing chip 30 be the reference surface, denoted as I _obj (x, y), (x, y) represents the coordinates of a two-dimensional Cartesian coordinate system. The stripe pattern is as follows: p (P) _Xm (x, y) and P _Yn (X, Y), subscript X, Y denotes the two directions X, Y of the two-dimensional cartesian coordinate system, m, n denote the number of fringe phase shifts in the two directions X or Y, respectively, so there are a total of m+n fringe patterns, namely: { P _X1 (x,y),P _X2 (x,y),…,P _Xm (x,y),P _Y1 (x,y),P _Y2 (x,y),…,P _Yn (x, y) }. For convenience, the m+n stripe patterns herein are collectively referred to as: p (P) _k (x,y)，k∈[1,m+n]。

When the wide-field light scanning sequencing chip 30 is adopted, the super-resolution detection system 10 images the sequencing chip 30 to obtain a low-resolution wide-field image I ₀ (x, y). When the structured light scanning sequencing chip 30 of the stripes is used, the super-resolution detection system 10 collects the low-resolution stripesThe grain image is: i _k (x,y),k∈[1,m+n]。

In step S2, the wide-field image acquisition mode mainly includes two modes:

in one mode, the controller 14 controls all the micromirrors 1211 of the light modulator 14 to be in "ON" state, and the image obtained by the structural light incident ON the sequencing chip 30 is a low-resolution wide-field image I ₀ (x,y)。

Under the condition of ensuring the super-resolution reconstruction effect, the smaller m+n is, the better. In the second mode, since the two-step phase shift method is adopted to scan the sequencing chip 30 in the present embodiment, in one detection period, only 3 (m=2, n=1; or m=1, n=2) original images need to be acquired at least, so that super-resolution reconstruction can be completed. Taking m=2 and n=1 as an example, i.e. two groups of stripes and one group of stripes are respectively projected in the X and Y directions, the acquired stripe images are recorded as: i ₁ (x,y)，I ₂ (x,y)，I ₃ (x, y). From the foregoing, I ₁ (x, y) and I ₂ The fringe phase difference on (x, y) is pi, so I will be ₁ (x, y) and I ₂ (x, y) adding to obtain low resolution wide field image, i.e ₁ (x,y)+I ₂ (x,y)＝I ₀ (x, y). Therefore, when the two-step phase shift mode in this embodiment is adopted, it is unnecessary to intentionally control all the micromirrors 1211 in the optical modulator 14 to be in the "ON" (ON) state, so as to acquire the wide-field image I ₀ (x, y). Therefore, compared with the first mode, the wide-field image is obtained through calculation, and is not obtained through actually projecting the structured light of the wide field, so that the time for obtaining the image is reduced, and the flux is improved.

Referring to fig. 8, step S3 specifically includes:

step S31, taking the wide field image as a preliminary estimation of a super-resolution image;

step S32, constructing an objective function to acquire an objective image, and updating the objective image;

step S33, obtaining a super-resolution image according to the updated target image;

step S34, repeating the step S32 and the step S33, and traversing the first stripe image and the second stripe image; a kind of electronic device with high-pressure air-conditioning system

Step S35, repeating step S32, step S33 and step S34 until the preset evaluation index converges.

The optical transfer function of the super-resolution detection system 10 is defined as: OTF; the point spread function is: PSF. Stripe pattern P _k (x, y) projected onto the sequencing chip 30, the super-resolution detection system 10 acquires a low-resolution fringe image I _k This optical process of (x, y) can be expressed in the frequency domain as:in the airspace, sequencing chip I _obj And stripe pattern P _k Multiplying; in the frequency domain, an optical transfer function OTF and an objective function I _tk Is multiplied by the frequency spectrum of (a).

In step S31 and step S32, the wide field image I is selected ₀ (x, y) as a super resolution sequencing chip 30 (I) _obj ) Preliminary estimation of super-resolution images of (I), i.e _obj ＝I ₀ 。

In step S32, an objective function is constructed: i _tk ＝I _obj ·P _k Then the target image I is updated according to formula (1) _tk Acquiring a first updated image

deconv stands for "deconvolution" operation, which is mainly used to suppress image noise, improve image quality, and increase convergence speed.

In step S33, the first updated imageCarrying in (2) to obtain a second updated image

In step S34, steps S32 and S33 are repeated to traverse all m+n stripe patterns P _k (x,y)，k∈[1,m+n]。

The steps S31 to S34 are regarded as sequentially iterated, and in step S35, a plurality of iterations are performed until the preset evaluation index converges.

In this embodiment, the number of iterations is typically 10 to 50. In this embodiment, the DNA nanospheres are arranged in a square regular pattern and have a specific structure, so that the predetermined evaluation index is a selected structural similarity index (Structural Similarity Index, SSIM) for characterizing the convergence of multiple iterations. Two stripe images, wide field image I, are known ₀ And updated images for each iteration(for simplicity, will beThe superscript "updated" of (1) moves to the subscript position: i _obj-updated ) The SSIM calculation formula is represented by formula (3):

wherein l, c, s respectively compare the images I ₀ And I _obj-update Brightness, contrast, structure. Alpha>0，β>0，γ>0 for adjusting the relative importance of l, c, s, the greater the relative value, the higher the relative importance; typically α=β=γ=1.Andrespectively are images I ₀ And I _obj-updated Average value of (2);andrespectively are images I ₀ And I _obj-update Standard deviation;for image I ₀ And I _obj-update Is a covariance of (c). C (C) ₁ ，C ₂ ，C ₃ Are all constant and are used for maintaining the stability of l, c and s when the denominator is small.

The following verifies the effects of the above-mentioned super-resolution detection system 10 and super-resolution detection method, mainly providing two verification modes: the two-value bitmap is used as a reference image for verification, and the fluorescent image of the sequencing chip is used as a reference image for verification.

Referring to fig. 9 and 10, an ideal reference image I is shown _obj (fig. 9 and 10 (a)) and stripe image P _X1 (first fringe pattern in first direction)Like fig. 9 and 10 (b), P _X2 (second stripe image in first direction, fig. 9 and 10 (c)) and P _Y1 (first stripe image in the second direction, fig. 9 and fig. 10 (d)) are multiplied. Convolving with optical transfer function OTF (fig. 9 and 10 (k)) of the system after two-dimensional Fourier transform, and performing two-dimensional inverse Fourier transform to obtain streak illumination image I ₁ 、I ₂ I ₃ . The above process is a process of acquiring a fringe image. Wide field image I ₀ (fig. 9 and 10 (h)) can be obtained by combining I ₁ +I ₂ Obtained.

According to the above-described super-resolution reconstruction algorithm, a super-resolution image is obtained (fig. 9 and fig. 10 (i)). In the iterative process, SSIM is selected as an evaluation index (fig. 9 and fig. 10 (n)) for characterizing the convergence in the iterative process of the algorithm. As can be seen from fig. 9 and fig. 10 (n): the SSIM of the reference image and the SSIM of the wide-field image have a larger difference, the SSIM of the super-resolution image quickly approaches to the SSIM of the reference image along with the iteration, and the final convergence value is larger than that of the wide-field image, so that the super-resolution effect can be realized by the method.

The original image size of the reference image shown in the diagram (a) in fig. 9 is 1024×1024 pixels, and the iteration time is 12.18s for 50 times; SSIM has converged by 20 iterations, so the 20 iterations of super-resolution reconstruction take 4.87s.

The original image size of the reference image shown in fig. 10 (a) is 256×256 pixels, and the time consumption for 50 iterations is 0.84s; SSIM has converged by 20 iterations, so 20 iterations of super-resolution reconstruction take 0.33s. The use of a graphics processor (Graphics Processing Unit, GPU) or the addition of random access memory (Random Access Memory, RAM) may further reduce time consumption.

By exhibiting the spatial morphology to more intuitively observe the super-resolution reconstruction effect in the spatial domain, in the (o) diagram of fig. 9, since the reference image is a binary lattice, it is necessary to characterize the edge spread function (Edge Spread Function, ESF) and the line spread function (Line Spread Function, LSF) of a single dot diameter position. The left/right axes in fig. 9 (o) compare the ESF and LSF of the reference image ((a) image), the wide-field image ((h) image), and the super-resolution reconstructed image ((i) image), respectively. Looking at the left axis, it can be seen that: because the numerical aperture of the system is limited, the edges of the wide-field image are more gentle and have lower resolution than those of the reference image; the super-resolution reconstructed image has the phenomenon similar to 'ripple wave', but is obviously steeper than the wide-field image, and the resolution is higher, namely the super-resolution effect is realized. Looking at the right axis, it can be seen that: LSFs of the wide-field image, the super-resolution image and the reference image are in a trend of decreasing in sequence; the narrower the LSF, the higher the resolution; since the LSF of the super-resolution image is narrower than that of the wide-field image, the effect of super-resolution is achieved.

In fig. 10 (o), the ESF of the reference image, the ESF of the wide-field image, and the ESF of the super-resolution image can be visually observed. By comparison, it was found that: the two peak points which are obviously distinguished in the reference image cannot be distinguished in the wide-field image due to the limited numerical aperture of the imaging system, but are distinguished again in the super-resolution image, so that the super-resolution effect is realized.

The above is to directly compare the super-resolution effect from the airspace or to evidence the super-resolution effect from the SSIM side; in order to more intuitively compare the super-resolution effect, the reference image (fig. 9 and 10 (a)) and the wide-field image (fig. 9 and 10 (h)) and the super-resolution reconstructed image (fig. 9 and 10 (i)) are subjected to two-dimensional fourier transformation to obtain two-dimensional spectrums (fig. 9 and 10 (j), (l), (m)). The center of the two-dimensional frequency spectrum represents low-frequency space frequency, and the space frequency is gradually increased from the center to the outside; the higher the spatial frequency, the higher the resolution; comparing the two-dimensional spectra (fig. 9 and 10 (j), (l) and (m)) shows that the two-dimensional spectrum of the reference image (fig. 9 and 10 (j)) has the widest range; the two-dimensional spectral range of the super-resolution reconstructed image (m-plot in fig. 9 and 10) is much larger than that of the wide-field image (l-plot in fig. 9 and 10), so that the effect of super-resolution is achieved.

It should be understood that all the method steps described above in this application are not limited to the order of execution of the steps unless specifically indicated. Also, the execution order of the steps is not limited by the order of the reference numerals, for example, the steps S1 and S2, and the reference numerals S1 and S2 are not used to limit that the step denoted by S1 is executed prior to the step denoted by S2, and it should be understood that the super resolution detection method includes both an embodiment in which the step S1 is executed first and then the step S2 is executed and an embodiment in which the step S1 is executed first and then the step S2 is executed.

The super-resolution detection system 10 and the super-resolution detection method provided in this embodiment adopt a two-step phase shift mode, and set two micro-mirrors corresponding to one fringe period, which is beneficial to improving fringe density, increasing field area, and improving sequencing flux.

The scanning mode of two-step phase shift is combined with the super-resolution reconstruction algorithm, so that super-resolution reconstruction is realized on the basis of less acquired images, the image acquisition time is shortened, the image processing time is shortened, and the detection cost of biological information is saved.

It will be appreciated by persons skilled in the art that the above embodiments have been provided for the purpose of illustrating the invention and are not to be construed as limiting the invention, and that suitable modifications and variations of the above embodiments are within the scope of the invention as claimed.

Claims

A super-resolution detection system for detecting biological information of a sample to be detected, the super-resolution detection system comprising:

the light source module is used for emitting light source light;

the light modulator comprises a plurality of micro-reflectors, wherein the micro-reflectors are divided into a plurality of modulation units, each modulation unit comprises two micro-reflectors, the light modulation module is used for modulating the light source into structural light to be emitted, the structural light can be guided to the sample to be detected so as to enable the sample to be detected to emit detection light, the structural light forms a stripe light spot on the sample to be detected, and each modulation unit corresponds to a stripe period;

the imaging module is used for acquiring a fringe image according to the detection light and acquiring a wide-field image; and

and the controller is electrically connected with the light modulator and the imaging module, and is used for adjusting the phase of the stripe formed by the structured light and carrying out super-resolution reconstruction according to the stripe image and the wide-field image so as to acquire a super-resolution image and acquire biological information of the sample to be detected.
The super-resolution detection system as claimed in claim 1, wherein each micro-mirror is defined by dimensions: deltax x Deltay, defining the resolution of the super-resolution detection system as R, and the reduction multiple of the fringe period when the structured light is projected onto the sample to be detected isWhere Δx=Δy.
The super-resolution detection system as claimed in claim 1, further comprising a magnification adjustment assembly comprising a third lens having a focal length f and a fourth lens ₃ The specific focal length f of the fourth lens ₄ ，
The super-resolution detection system as claimed in claim 1, wherein the controller is further configured to control each micro-mirror to reflect the light source light to acquire the wide field image.
A super-resolution detection method for detecting biological information of a sample to be detected, the super-resolution detection method comprising the steps of:

generating structured light, scanning a sample to be detected in a two-step phase shift mode, acquiring a first stripe image and a second stripe image in a first direction, and acquiring the first stripe image in a second direction;

acquiring a wide-field image of the sample to be detected;

setting a preset evaluation index, and acquiring a super-resolution image according to the first stripe image, the second stripe image and the wide field image in an iterative mode so as to acquire biological information of the sample to be detected.
The method of claim 5, wherein the step of obtaining a wide field image of the sample to be detected comprises:

and superposing the first stripe image and the second stripe image in the first direction to acquire the wide field image.
The method according to claim 5, wherein the method is applied to a super-resolution detection system, the super-resolution detection system includes a light source module and a plurality of micro-mirrors, the light source module is used for emitting light source light, and the step of obtaining the wide-field image of the sample to be detected includes:

each micro-mirror is controlled to reflect light source light to acquire the wide field image.
The method of claim 5, wherein the step of setting a preset evaluation index to obtain a super-resolution image based on the first stripe image, the second stripe image and the wide field image in an iterative manner comprises:

taking the wide-field image as a preliminary estimate of a super-resolution image;

constructing an objective function to acquire an objective image, and updating the objective image;

acquiring a super-resolution image according to the updated target image;

repeating the two steps, traversing the first stripe image and the second stripe image; a kind of electronic device with high-pressure air-conditioning system

And repeating the previous step until the preset evaluation index converges.
The method of claim 5, wherein the number of iterations in the step of iterating is 10 to 50.
The method of claim 5, wherein the evaluation index is a selected structural similarity index.