CN1385700A - Sperm fluorescent staining computer detection system - Google Patents

Abstract

There is sperm fluorescence staining computer detecting system as well as its application to the sperm detection, and it uses the image card controlled by the computer to realize the sperm image which is under the microscope or take a random and continuous short of the sperm, then performs identification, correlative operation and motional parameter calculation in order to ascertain the sperm motility rate and livability; its fluorescence staining method can make the live sperm display green in the fluorescence, the dead ones display red and the rest of any secretion and impurity all not colorated, and it resolves the difficult problem that the detecting system is not easy to get rid of the impurity and analyze the sperm shape, and it can not only detect and analyze the sperm, but also be applied to the study on the venereal disease aspect and the diagnosis of the pathology aspect.

Description

Sperm fluorescent staining computer detection system

The invention belongs to biomedical Digital image technology field, particularly the dynamic image Visual calculation machine testing system of sperm fluorescent dyeing.

Sound attitude analyzing and testing to sperm has following many methods at present:

One, on microscope with eyes direct-views, its dynamic perfromance is difficult to obtain quantitatively, can only draw several "+" qualitatively and number be evaluated.

Two, utilize multiple exposure technique, though more better to its dynamic perfromance than above-mentioned, can obtain several indexs, speed is very slow, and data volume is little, and low precision also is difficult to adapt to clinical needs.

Three, utilize light scattering method, this method sample rate is fast, and analytical calculation is also very fast, but the detected parameters that directly obtains seldom, so be unsuitable for clinical.

Four, utilize computing machine and digital image technology to be equipped with the formed visual at random check and analysis device of the micro-corresponding software package that videotapes fast.The utility model patent (96222683.1) that is entitled as " moving (quiet) attitude image detection of sperm and the visual and management devices of medical record image " that the applicant finishes in advance is exactly this class device.This device is made of microimaging parts, image acquisition parts, MICROCOMPUTER PROCESSING parts, image storage and medical record management (VCD) parts, demonstration print member, constant temperature dust proof member.The principal feature of this device is sampling (candids photograph), it is fast to handle, data volume is big, and visual voltinism is strong, and objectivity is good, the precision height, function is complete, the parameter of surveying many, be the comparatively ideal device of present semen efamination analysis, clinical, scientific research institution's need.But this device still has the some shortcomings part, and one adopts black and white camera (CCD), and there is difference in the image that the testing staff is observed under mirror, judges to be difficult for accurately; They are two years old, this image card can only be operated under the lower state of microcomputer clock frequency, lower-speed state, the microcomputer that is mated is up to 586/166 with it, and the internally cached and External Cache of CPU must be closed, just can make the image card operate as normal, cause the waste of computer resource, and processing speed be slower.

The purpose of this invention is to provide a kind of method that semen sample is dyeed, and the computer detection system that designs based on this method, in order to improving accuracy in detection, make it have that sound attitude self check degree of testing the speed is fast, resolution is high, precision is high, objectivity reaches the strong characteristics of function well, make it be more suitable for demand clinical, teaching and scientific research.

The sperm that the present invention proposes is from analytic system, comprises the microimaging parts that the still image random shooting with test sample gets off; The picture signal that said microimaging parts are sent is discerned to give handling and is stored, and further the signal of maturation is carried out a series of calculation processing and result's output and carries out the teaching of interactive medical record, the MICROCOMPUTER PROCESSING parts of academic exchange; Image stores and medical record management (VCD) parts, show print member, to test sample heat, insulation and dustproof constant temperature dust proof member; Said microimaging parts are made up of micro-, high-resolution color CCD; Said image acquisition component is to constitute by dynamic or stationary singnal are captured with pretreated black and white and coloured image card: said MICROCOMPUTER PROCESSING parts are to be made of 586 microcomputers and multimedia unit thereof; Said demonstration print member is made up of cromogram monitor and printer; Said constant temperature dust proof member is made up of organic glass seal closure and air conditioner type warmer; Also comprise the still image dedicated analysis parts of forming by full-color colour atla and corresponding software.

Area of computer aided semen analysis system is the new technology that nearly of growing up in 20～30 years detects the semen parameter, it carries out organic combination with microimaging technology and computer system, the birth of this technology, solved the cumbersome deficiency that waits of manual inspection seminal fluid, can be objective apace seminal parameters calculated and go out the result, be subjected to being engaged in the welcome of researchers such as reproductive medicine, andrology by computer printout.Through nearly 30 years clinical application, find that also there are some unsolved technical matterss in the area of computer aided semen analysis instrument (CASA) that uses at present, for example, in image processing procedure, how accurately to discern sperm, how to get rid of in the seminal fluid except that sperm, to some other cells, as leucocyte, the erroneous judgement of red blood cell and reproduction cell that comes off and impurity etc., existing gray scale recognition methods is used as sperm identification to these compositions probably, and the too high illusion of sperm concentration appears, sometimes real sperm fails again to identify in the image, may cause density on the low side.In addition, can not to distinguish motionless sperm be extremely or work for traditional gray scale recognition methods.One of task of the present invention just provides a kind of technical scheme that addresses the above problem, and this technical scheme comprises sperm fluorescent dyeing and at the Computerized analysis system of sperm fluorescent dyeing sample.

Fluorescence staining of sperm method of the present invention has been used two kinds of reagent, is referred to as reagent A, reagent B respectively.The compound method of reagent A is: first preparation pH3～7.8, and the damping fluid of preferred pH4～6, again with the acridine orange fluorchrome of this damping fluid preparation 0.001～0.5%, preferred 0.01%, and it is stand-by to put it in-20 to-30 ℃ of low temperature refrigerators preservation.Reagent A also can adopt physiological saline or distilled water to substitute the acridine orange fluorchrome solution of damping fluid preparation 0.001～0.5%.Reagent B adopts the B liquid in the Molec.Probe.Eugene.Oregon.USA life or death staining of sperm reagent.The step of dyeing is: after the fresh semen liquefaction, draw 100 μ l seminal fluid and put into clean tube, abundant mixing was placed 5 minutes in 37 ℃ behind adding reagent A and the reagent B.Also can add reagent A earlier in 37 ℃ of placements 5 minutes, and then add reagent B, place 5 minutes in 37 ℃.Draw the dyeing seminal fluid of 10 μ l, add in the makler plate, put microscopically and detect.

Detection principle of the present invention:

Shown in Fig. 2 block diagram, utilize the quick candid photograph function that image card had under the computer control, the moving image or the still image of the sperm under realization adjusts the telescope to one's eyes are taken at random continuously, the image sequence that is obtained is temporarily stored in image card and the computer storage unit, image card and computing machine are discerned computing to each width of cloth graphical analysis then, extract the location parameter of sample in the image and give record, again according to the location parameter in these different images, obtain the incidence relation (promptly carrying out related computing) between each location parameter, solve the track that sample moves by this this Relation Parameters, calculate the parameters of sample motion at last from these tracks.

The present invention has following characteristics:

1. can distinguish sperm anyway, the rubescent look fluorescence of Necrospermia.Sperm jaundice green fluorescence alive, do not influence the propulsion speed of the density of sperm in the stoste and the sperm of living, staining procedure is simple, grasp easily, A liquid phase in sperm fluorescent dyeing reagent A provided by the invention and the existing Molec.Probe.Eugene.Oregon.USA life or death staining of sperm reagent relatively, it is low to have cost, low price, the advantages such as slowly, not influencing observing effect of preserving easily, decay.

2. adopt colour imagery shot, coloured image card, resulting like this image is a full-color image, and especially aspect the still image processing, its superiority is the most outstanding, makes the testing staff of institute can obtain observing identical image down with mirror, and it is more accurate to judge.Make the MICROCOMPUTER PROCESSING result more accurate because the object of surveying and impurity essential distinction is arranged on color.

3. said image card almost can be complementary with all at present common on the market microcomputers, and is compatible good, makes full use of the computer resource arithmetic speed and greatly improves.

4. adopt two software packages, a special disposal dynamic image, one is used for handling still image specially.In static software package, added stain smear has been observed; Analyze deformity and differentiate and organize the function of embryo aspect.Like this software package is divided into two, acts in accordance with the division of their functions and duties, can make result more accurate, and expand the basis of laying for function from now on.

Application of the present invention is very wide, as artificial propagation department of the andrology's Urology Surgery in the hospital, male sterility clinical laboratory, family planning, test-tube baby and animal etc.

Brief Description Of Drawings:

Fig. 1 is an embodiments of the invention general construction synoptic diagram.

Fig. 2 is a testing process block diagram of the present invention.

Fig. 3 is an embodiment of the invention software system structure synoptic diagram.

Fig. 4 is an embodiment of the invention software systems operational flow diagram.

Fig. 5 is embodiment of the invention sperm motility stack sequence synoptic diagram.

Fig. 6 is an embodiment of the invention sperm identifying synoptic diagram.

Fig. 7 is an embodiment of the invention dual threshold method synoptic diagram.

Fig. 8 is embodiment of the invention chain code direction definition figure.

Fig. 9 is an embodiment of the invention sperm profile synoptic diagram.

Figure 10 concerns synoptic diagram for embodiment of the invention position.

Figure 11 is an embodiment of the invention sperm motility direction definition synoptic diagram.

Figure 12 is an embodiment of the invention sperm motility curve synoptic diagram.

Figure 13 is the related computing block diagram of the embodiment of the invention.

Figure 14 is overlapping detection of the embodiment of the invention and cancellation process flow diagram.

Figure 15 is an embodiment of the invention sperm trace routine general diagram.

The equipments overall structure of the embodiment of the invention comprises as shown in Figure 1 with the lower part:

1), microimaging parts: microscope 1, CCD2;

2), image acquisition component: full-color image card (being installed in the computing machine 3);

3), microcomputer parts: 586 microcomputers and multimedia components 3 thereof;

4), display unit: colour picture monitor 4, printer 5;

5), constant temperature dust proof member: heating arrangement and organic glass seal closure (not shown) thereof;

6), special-purpose sample box and worktable (not shown);

7), video recorder 6.

Now that the function description of each minute is as follows:

Show the emblem shooting part:

The microimaging parts are made up of microscope and high resolving power, the highly sensitive colour imagery shot (CCD) of band shooting, and its function is that the sound attitude image random shooting that will be placed on the test sample in the sample box gets off to send into image acquisition component in order to handling.

Image acquisition component mainly is by the image card burden, and its function is the image that CCD sends to be discerned to give handling store, and further the signal of maturation is delivered to computing machine and carry out overall treatment.

The MICROCOMPUTER PROCESSING parts are made up of peripheral hardwares such as computing machine, monitor, printers, its function is to process processing by the original signal that volume software is brought image card comprehensively, by display and printer all parameters are shown and printout, or original signal is for good and all stored in order to check in the future and careful research by video recorder, multimedia technology.

Display unit is actual to be made up of the computer monitor and the colour picture monitor of singly joining, and computer monitor is that the signal of being responsible in gatherer process and the calculating process shows, aims at patient in order to watch the characteristic of disease of controlling oneself and singly join colour picture monitor.

The constant temperature parts are formed by adding gentle heat-preserving equipment, heat is constantly euthermic hot blast to be blasted in the sealing stay-warm case by hot hair dryer, this room conditioning formula thermostatic mode also is one of advantage of the present invention, homogeneous heating, the influence that not changed by the extraneous air-flow stream of people and room temperature, effect is very remarkable.

Special-purpose sample box is must orchestration spare in the parts of the present invention, for guaranteeing that the individual layer sampling sample cell degree of depth is 10 μ m.Worktable also is according to the special adapted of parts situation.

Multimedia components is made up of CD-ROM drive, sound blaster, decompress(ion) card and audio amplifier, its function: the one, and for the medical worker carries out the science and education activity, the 2nd, carry out exchange of technology etc. in order to show medical record.

Above relevant hardware has been done brief introduction, be described further again with regard to the principle of its parts and the function of software below.The operation of whole software and structured flowchart are as shown in Figure 3 and Figure 4;

Realize that sperm detects except hardware devices such as microcomputer image card, microimaging automatically, software also is important ingredient.Except that considering needed pattern-recognition and mathematical operation algorithm, also consider the convenience of clinical use in software systems, easy to learn, dirigibility is good, and adaptable characteristics have adopted the output input manager module of necessity that menu mode occurs in the software.

What specify is: this software is higher to the power of holding of mistake, in most cases, mistake (the operation wrong knowledge of software to occurring in the system, the mistake of input and output and system itself) can discern automatically, and the execution of interrupt routine automatically, show error message, all these is to realize by the administration module of makeing mistakes in the system.

The brief introduction of present embodiment sperm trace routine is as follows:

(1) image obtaining and storing.

Native system mainly comprises following four parts to the trace routine of sperm as shown in Figure 2: shooting of image (obtaining) and storage, and the identification of sperm, related computing and kinematic parameter calculate.Its ultimate principle is to look like to describe the actual sample motion conditions of (refer to sperm, hereinafter be referred to as sample) by a series of ordered graph, as shown in Figure 5 (following mainly is that example describes in detail with the sperm).

Consecutive image shown in Figure 4 is the expression sperm is moved to position 5 by position 1 a track sequence.

Utilize the peculiar function of selected coloured image plate to realize the continuous candid photograph and the storage of image easily, now very crucial is the identification that how to realize exactly sperm.

(2) identification of sperm

From superposition sequence chart 6 as seen, utilize consecutive image to judge the motion conditions of sperm, at first can Guan Jian problem be identify sperm A and sperm B from image exactly, this belongs to the problem of computer patterns identification, head it off will relate to the denoising of image and handle, image binaryzation (going background) is handled, the extraction of Sperm characteristics and coupling (pattern-recognition) or the like problem in the image, and it concerns that block diagram is as shown in Figure 5.

1, the calculating of local background

Owing in seminal fluid, except sperm is arranged, also deposit impurity, the inequality of light path in the microoptic in addition, make the image that records present the inconsistent phenomenon of background light and shade, therefore take a unified threshold value pattern to go background process just very difficult with the view picture figure that is obtained, some often concrete threshold values only can realize the good separation of a part of image background.

At this problem, we have adopted the method for dynamic threshold, image is carried out the trend binary conversion treatment, just according to the mean value of topography inside, calculate the threshold value that we need, make binary-state threshold change, so just can make whole image can both realize the good separation of sperm and background with the variation of local gray level.The concrete method of calculating local background's mean value is: existing 800 * 600 image segmentation are become the small images of 8 * 8 or 4 * 4 pixels, to the mean value U of its gray scale of image calculation of each fritter: $\mathrm{Ui}=\frac{1}{64}\underset{k=1}{\overset{8}{\mathrm{\Σ}}}\underset{I=1}{\overset{8}{\mathrm{\Σ}}}U(k,L)$ Or $\mathrm{Ui}=\frac{1}{16}\underset{K=1}{\overset{4}{\mathrm{\Σ}}}\underset{I=1}{\overset{4}{\mathrm{\Σ}}}U(K,1)$

U wherein _I(k, 1) is the gray-scale value of (k, 1) pixel in the i piece subimage block, U ₁Be the average pixel gray-scale value of i piece subimage, according to U ₁Size, we just can determine the binary-state threshold size of corresponding i piece subimage block.

2, the calculating of binary-state threshold

In the ordinary course of things, from the sperm motility image that microscopically obtained, the gray level of sperm usually changes within the specific limits, and the head color is more shallow often, and the afterbody color is more shallow, and the middle part color of head is more shallow, and gray level is higher; All around color is darker, and gray level is low.In addition, because The noise, it also is at random that the gray level of sperm body itself changes, and particularly the variation of oxen and horses sperm gray level is more even.In order to realize the better binary conversion treatment effect, the present invention has adopted the dual threshold method in the binaryzation process, promptly adopt two threshold values respectively dark the and light-colored part of face palladium to be carried out binary conversion treatment, and the use of two threshold values is that certain restriction relation is arranged, as shown in Figure 7.The disposal route of this dual threshold has the function of background separation and the dual processing of noise cancellation.

3, denoising place and binaryzation computing

Because the interference of various factors and the restriction of image observation condition, the image that is obtained signal to noise ratio (S/N ratio) often is very low, the source of its noise usually is in the seminal fluid to be contained is granular piece of tissue impurity, do not eliminate in the simple binaryzation computing of this impurity with the separation of background, the existence of these noises has influenced the uniform continuity of smart image, there is identification to bring very big difficulty to sperm, so the present invention has introduced the noise removing processing in the binaryzation process of image.

Because noise presents graininess more, compare with the yardstick of sperm, its size is generally less, and its gray scale also has bright, dark gradually changeable unlike the gray scale of sperm image, still utilize these characteristics that image is carried out denoising to handle, the method of handling is: image to be processed is scanned, to each picture element that is scanned (k, i), detect its gray-scale value U (k, 1) size of each point gray-scale value and all around is if satisfy condition: 1, the gray-scale value U of this point (k, 1)＜threshold value T ₁, the gray-scale value U of this point (k, 1)＜T ₂And has any gray-scale value＜T in 4 of the upper and lower, left and right at least ₁The time, then this is the picture element of sperm image, otherwise the picture element that this point is not the sperm image (for background dot or noise spot).

Being formulated above-mentioned processing procedure is respectively:

The first step: if U (k, 1)＜T1, then point (k, 1) may be noise or sperm picture point, if U (k, 1)＞T ₁, then point (k, 1) may be background dot or sperm figure.

Second step: the point to all U (k, 1)＜T1 detects

If U (k+1,1)＞T ₂And U (k-1,1)＜T ₂And U (k, 1-1)＜T ₂, then point (k, 1) is a noise spot.

The 3rd step: all are confirmed as the picture element (k ', 1 ') of sperm picture point

If U (k '-1,1 ')＜T2, then point (k '+1,1 ') is sperm picture point (marginal point).

U(k’-1，1’)＜T2........(k’-1，1’)......

U(k’，1’+1)＜T2........(k’，1’+1)......

U(k’，1’-1)＜T2........(k’，1’-1)......

The 4th step: whether all sperm picture point are all found out.

Being then to finish, is not then to repeat third step.

Through after such processing, the granularity noise in the image has been eliminated, and the image of sperm has also guaranteed uniform continuity preferably because of noise is eliminated.In addition, the employing owing to threshold method also makes the profile of sperm image that more complete reservation has been arranged.

4, Boundary Extraction

Through after the binary conversion treatment, next step is how isolated profile in the image (these profiles are the profile of sperm a bit, and some may not be the sperm profile) to be carried out the border to follow the tracks of (Boundary Extraction just) at image.Promptly obtain needed contour curve by the point of the grey scale change in the tracking image, these contour curves are foundations that we carry out sperm shape recognition and the extraction of sperm location parameter.

The method of Boundary Extraction has multiple, and of the present invention is direction chain code method, and the direction definition of chain code as shown in Figure 8.

Adopt the right hand to touch the wall criterion in the time of enthusiastically, in tracing process, establish that current will to search the direction that obtains be direction, and previous direction be before-direction, when getting different values for before-direction, the direction of sequential search to be:

Last direction	The direction of searching
		????1	????2.1.8.7.6
????2	????4.3.2.1.8.7
		????3	????4.3.2.1.8
????4	????6.5.4.3.2.1
		????5	????6.5.4.3.2
????6	????8.7.6.5.4.3
		????7	????8.7.6.5.4
????8	????2.1.8.7.6.5

The selection of initial direction is relevant with concrete scan mode, is from bottom to top scan mode because we adopted, so initial direction is made as before-direction=2.

5, the calculating of location parameter

Fig. 9 is the sperm profile that the present invention obtains with the border tracking mode, to this profile, we need select in a certain position of sperm certain a bit to calculate the location parameter that sperm moves, because the head of sperm, represented the key substance place of sperm life on the one hand, on the other hand, the change of shape of sperm head in motion process changes much smaller, much obvious than afterbody.

In real work, the selection of position utilization asks the method for average to obtain, just location point (C _x, C _y) can obtain by following two formulas: $\mathrm{Cx}=\frac{1}{2}(X\max +X\min )$ $\mathrm{Cy}=\frac{1}{2}(Y\max +Y\min )$

Wherein Xmax and Xmin, Ymax and Ymin have represented the minimum and maximum boundary coordinate value that is obtained in the sperm profile tracing process respectively.In order to guarantee this position, parameter can drop in the scope of sperm head, we have taked two measures, the first, select rational binary-state threshold, make image after the process binary conversion treatment, the image of sperm head has obtained complete reservation, and the afterbody image not necessarily all keeps, and only keeps near near the image the head, to show the difference with other impurity images.The second, when implementing the border tracking,, and ask its sperm portion and tail position by the curvature of calculating tracker wire each point.Its tail position is carried out part deletion, mainly is that head border and afterbody local boundary are followed the tracks of and carried out to central points.

6, the statistics of quantity

After image is finished the border tracking, just determined the quantity of the sperm sample of gathering, how many bar tracker wires are arranged, just there are what sperm samples collected.Utilize the strong cumulative calculation function of computing machine, its quantity what can obtain soon.But be noted that the sperm count that has in the sperm count that utilizes video camera (CCD) to gather microscopically and the practical field of view compares, there is certain difference, the size of this species diversity is relevant with the actual motion speed of sperm, and this can only determine the size of its difference by a large amount of experiments.

Three, related computing

According to preceding described, the situation of sperm motility is to describe by a series of orderly images, each width of cloth in this sequence image is carried out sperm identification computing respectively, just can obtain a series of sperm location parameter group (these parameter group also are orderly), each group parameter is all one by one corresponding to the width of cloth sperm image in the orderly image sequence, obviously, also be to have certain mutual relationship between this each group location parameter.Though divide next step displacement corresponding to sperm on each location parameter in first group as each location parameter in second group, found this corresponding relation, also just found displacement curve.So-called related computing also is to seek the computing of respectively organizing this corresponding relation between the location parameter.

The core of related computing is three basic norms: distance criterion, speed criterion and direction criterion.

1, distance criterion

By shown in Figure 10, the location point at sperm a place (is promptly write down for A in the position (parameter) of a certain sperm a among the figure (1) with parameter A.Also can mention in the argumentation below " position A ", promptly represent by that represented location point of parameter A), the adjacent successive image of figure (1) wherein has B for figure (2) ₁, B ₂And B ₃Position A among three positions and the figure (1) is close, so B ₁, B ₂, B ₃May the pairing position of which parameter be the pairing the next position of sperm a? (we claim that this position is a predicted position), the roughly upper limit according to the image time interval in the continuous shooting process and sperm motility speed, near the A of position, determine a zone, for example, with position A is the center, and r is the border circular areas (shown in figure (2)) that radius is drawn.In the successive image, drop on the location point within this zone, as B ₁, B ₃The position accounts for, and then may become the predicted position point of sperm a, and position B ₂Then be not the predicted position point of sperm a, this definite mode is distance criterion, and it is divided into location parameter to be processed may predicted position point parameter and nonanticipating location point parameter two big classes.As describing then be with mathematic(al) representation:

If { Ai|i=1,2......., I} is a certain location parameter group, ai|i=1, and 2......I} by correspondence detect the set of sperm, { Bi|j=1,2...., J} is that { r is called the differentiation radius of distance criterion to parameter group for the adjacent subsequent parameter group of Ai}, r=shooting time interval of delta t x sperm speed limit Vmax.

Distance criterion can be described as:

For { the some elements A among the Ai} _kKAk ∈ 1,2....I}, computed range d=D (A _k, B _j) ‖ j=1,2...., J,

If d＜r, then B _jBe sperm a _IPossible predicted position point parameter,

If d＞r, then B _jBe sperm a _INonanticipating location point parameter,

If D (A _k, B _j) expression position A _kWith position B _jBetween distance.

2, speed criterion

When to judge when occurring a plurality of possible sperm predicted position in the radius r institute restricted portion, (as the B of figure (2) among Figure 10 ₁And B ₃The position), first transient motion speed under may predicted position is predicted, got wherein that possible position of velocity variations minimum is the predicted position of sperm, Here it is speed decision criteria.

If it is that speed is V that sperm a has just _A, corresponding to predicted position B ₁Can obtain corresponding instantaneous velocity is V _B3=D (A ₁B ₃)/Δ t is so the predicted position B of sperm a is: B=Bi, and Bi ∈ { B1, B3}, V _A-V _Bi=min (V _A-V _Bi, V _A-V _B3).

3, direction criterion

At first we make definition to the directivity of sperm motility, that regulation sperm moves in the plane of delineation and has 16 kinds of different direction of motion, and as shown in figure 11: for each direction of motion, we have also stipulated the value of a numeral as this direction.When sperm remained static, we are defined as the direction of motion value was " 16 ".

So-called direction criterion is: after having determined the predicted position of a certain sperm a according to the speed criterion, the sperm motility direction value under this predicted position is calculated.Compare with the original direction of motion value of sperm a if calculate the direction value of gained, change within a certain scope, then this pre-position really is the predicted position of sperm a, otherwise changes not within a certain scope, and this position is not a sperm a institute predicted position just.Being formulated this direction criterion is:

If the original direction of motion of sperm is S _A, be S by the direction of motion of predicted position gained that B calculates _B, the change threshold of setting direction of motion again is Δ S, so have again:

If | SB-SA≤Δ S| or | (SB-SA) mod16|≤Δ S sets up, and then position, position S is desired predicted position, otherwise is not predicted position just.

4, minimum distance criterion is when employing speed criterion and direction criterion can not produce the definite predicted position of sperm a, and we have selected minimum distance criterion as the last basis of determining the sperm predicted position again.So-called minimum distance criterion is also promptly: the position of known sperm a is A, and possible the predicted position of also knowing sperm a is gathered and is: and B1=1,2.....L}, to all I values, computed range D (A1B1|L=1 ... predicted position .L).The predicted position B of sperm a is: B=Bi|D (A1B1)=min[D (A1B1)]

Making above criterion basis for establishing is: the sperm of a proper motion, under the influence that does not have extraneous factor, saltus step by a relatively large margin all can not take place in the spot speed of sperm motility and instantaneous direction, the curve movement that the characteristics of this non-saltus step show as sperm is a level and smooth curve, as shown in figure 12.

When the motion of sperm is subjected to ectocine and bumps, or when sperm is in up-set condition, variation by a relatively large margin all will take place in the movement velocity of sperm and its direction, and at this moment the instantaneous velocity of sperm motility is often less, and this also is the reason that we adopt minimum distance criterion.

5, the process flow diagram of related computing

Preceding four problems are several criterions of having introduced in the related computing, will provide the process flow diagram of related computing below, as shown in figure 13.

From process flow diagram, as seen, after utilizing above-mentioned several criterion that the relevance of sperm motility is calculated, also to carry out the lap position and detect and lap position cancellation computing result of calculation.This is because after the judgement through speed criterion and direction criterion, each sperm has all had its predicted position of celebrating mutually, and that these predicted positions might occur is identical (promptly two or＜2820880＞0 erroneous judgements and the overlapping phenomenon of appearance).The disposal route of overlapping detection and lap position cancellation computing is such, as shown in figure 14.

After through plyability and lap position cancellation processing, the relevance computing that sperm motility detects just is through with, and next is exactly to have calculated the parameter index of sperm motility.

Four, kinematic parameter calculates

In related operational flowchart shown in Figure 13, a location parameter rearrangement step is arranged, the introducing of this ordered steps, make between each position parameter group corresponding relation performance for the sequence number correspondence between each parameter (for example, first group number one parameter is corresponding to second group number one parameter, number one parameter corresponding to the 3rd group, ...., number one parameter corresponding to the N group), this makes that we can be according to location parameter and this simple corresponding relation, depict the track of sperm motility easily, calculate relevant kinematic parameter.If we are at known location parameter (through the parameter after the related calculation process) set: { A ₁(j) }, I=1 wherein, 2........N is the group number of location parameter, j=1,2.....J is the number sequence number of the location parameter in every group, also the i.e. sequence number of corresponding sperm.Obviously, this location parameter set description J sperm { a _j| j=1,2....., the movement locus of J}.Because we arrange according to sequence number at the known relation of respectively organizing between the location parameter, so the series of parameters of relevant sperm motility is also described according to the determined relation of sequence number j.

(1) track of sperm motility

With location point A _i(j) variation with I couples together, and has just obtained desired sperm a _jMovement locus.Obviously, we can obtain J bar sperm motility geometric locus altogether to the variation of j=J with j=1.

(2) linear velocity of sperm motility: $V\left(j\right)\left[\underset{j=1}{\overset{N-1}{\mathrm{\Σ}}}\left(D(\mathrm{Ai}\left(j\right),A_{i+1}\left(j\right))\right)\right]/\left[(N-1)\mathrm{\Δt}\right],j=\mathrm{1,2}......J$

This linear velocity has reflected the displacement of the lines of sperm motility and the ratio of time.

(3) space rate of sperm motility: V (j) D (A _I(j), A _n(j))/[(N-1) Δ t], j=1,2......, J

This space rate has reflected the intensity that sperm advances along a certain direction.Space rate is big, shows that this sperm advances soon, and space rate is little, illustrates that the sperm pace is little.

(4) sperm motility line average velocity and straight line average velocity $V=\frac{1}{J}\underset{J=1}{\overset{J}{\mathrm{\Σ}}}V\left(j\right)$

V _LineThe average case that has reflected sperm motility speed. $V=\frac{1}{J}\underset{J=1}{\overset{J}{\mathrm{\Σ}}}V\left(j\right)$

V _DirectlyThe situation of on average advancing that has reflected sperm motility.

(5) sperm motility linear speed difference and space rate variance ${\mathrm{\&delta;}}_{\mathrm{<figure-callout\; id="2"\; label="V"\; filenames="A0110669000211.png"\; state="\{\{state\}\}">V</figure-callout>}}^2=\frac{1}{J}\underset{j=1}{\overset{J}{\mathrm{\&Sigma;}}}{(V\left(j\right)-V)}^2$ ${\mathrm{\&delta;}}_{\mathrm{<figure-callout\; id="2"\; label="V"\; filenames="A0110669000211.png"\; state="\{\{state\}\}">V</figure-callout>}}^2=\frac{1}{J}\underset{j=1}{\overset{J}{\mathrm{\&Sigma;}}}{(V\left(j\right)-V)}^2$

Here replace sample correction variance with second-order moment around mean, as long as sperm number J wherein is enough big.

(6) the motility rate number percent of sperm motility:

The threshold speed V of definition sperm motility _r, as the movement velocity V of sperm _Line＞V _rThe number of sperm account for a few percent of the sperm sum of being checked, the motility rate number percent of the sperm motility of promptly being known as.

(7) sperm concentration (being the sperm concentration of sample solution)

Sperm concentration=(the sample volume V that sperm sum C/ is observed) * (dilution ratio d)-1.

The above present embodiment device of introducing obtaining from image until the output of sound attitude testing result.By Figure 15 provide also overall picture for example in the drawings content of (1) part be included in Fig. 5, Fig. 6, Figure 10 and Figure 13; (2) Bu Fen content is included in Fig. 7, Fig. 8 and Fig. 9; (3) Bu Fen content comprises Figure 11, Figure 12 and Figure 14.

The most basic difference of the present invention and pick-up unit in the past is: 1, all change color system on the hardware into, microscope is by the original homemade import microscope that enlargement ratio is higher, performance is stronger that changes into.2, this device has on hardware the big change for improving the function of static analysis, has done bigger replenishing on software again, makes whole device be divided into two software packages.

Fluorescence staining of sperm embodiment of the present invention

1. the preparation of reagent: with the acridine orange fluorchrome solution of the damping fluid of pH5 preparation 0.01%, and put it into preserve in-20 ℃ of low temperature refrigerators stand-by.Reagent B adopts the B liquid in the Molec.Probe.Eugene.Oregon.USA life or death staining of sperm reagent.

2. Ran Se step: get seminal fluid to be checked, treat that fresh semen liquefaction back draws 100 μ l seminal fluid and put into clean tube, add behind 1 μ l reagent A and the 1 μ l reagent B fully mixing, placed 5 minutes in 37 ℃, the dyeing seminal fluid of drawing 10 μ l adds in the makler plate, puts microscopically and detects.

Fluorescent staining method of the present invention, make sperm alive under fluorescence excitation, be green, Necrospermia takes on a red color, all the other secretions and impurity are not painted, by the software recognition methods, the accuracy of detection is improved greatly again, the function of its detection is widened, particularly the morphological analysis to sperm provides convenience and accuracy, makes the automatic checkout system of sperm be able to basic improvement.

Characteristics of the present invention:

One, utilize decoration method, under fluorescence excitation, make sperm alive be green, Necrospermia takes on a red color, and any impurity in the seminal fluid, secretion, tissue block are not painted, and do not affect again sperm motility, so just solved the puzzlement of detection system maximum, how decontamination only detects one of the biggest problem of sperm;

Two, owing to only have chromospermism, also provide very big convenience to the morphological analysis of sperm like this, this has just solved two of the biggest problem in the detection system;

Three, because sample with different colors, has also utilized a kind of identification that its accuracy of identification is improved greatly in software analysis identification, thereby make the operation of its detection system more simple and practical accurately;

Four, because anyway the sperm color is clearly demarcated, make examining report one screen complete, for the motility rate of sperm, survival rate movement locus etc. very clear;

Five, because should there be larger change in the hardware and software of native system mutually, make the application function of native system that larger broadening be arranged, not only can detect the analysis seminal fluid, but also can be for the research of venereal disease aspect, the diagnosis of pathology aspect, this also be conventional method can't be obtained, that is to say that application is more wide.

Claims (6)

1. sperm fluorescent staining computer detection system comprises the microimaging parts that the sound attitude image random shooting with test sample gets off; The picture signal that said microimaging parts send is discerned the pre-service storage, and further the signal of maturation is delivered to the image acquisition parts in education department's machine; Finishing the random signal that image card is brought under the control of software carries out a series of calculation process and result's output and carries out the teaching of interactive medical record, the MICROCOMPUTER PROCESSING parts of academic exchange; Image is stored and medical record management (VCD) parts, show print member, it is characterized in that: the image card under said system computerizeds control is realized adjusting the telescope to one's eyes down taking continuously at random through the sperm motility image or the still image of fluorescent dye, and the image sequence that is obtained is temporarily stored in image card and the computer storage unit, image card and computing machine carry out the identification of sperm to each width of cloth image then, related computing and kinematic parameter calculate, obtain the incidence relation between each location parameter, Relation Parameters solves the track of sample motion thus, calculate the parameters of sample motion at last from these tracks, thereby determine the motility rate of sperm, survival rate also outputs to display with it, storer or printer.

2. as the said sperm fluorescent staining computer detection system of claim 1, it is characterized in that: the identification of said sperm, it is the method that adopts dynamic threshold, image is carried out trend binary conversion treatment and calculating, be about to the small images that existing 800 * 600 image segmentation become 8 * 8 or 4 * 4 pixels, to the mean value U of its gray scale of image calculation of each fritter: $\mathrm{Ui}=\frac{1}{64}\underset{k=1}{\overset{8}{\mathrm{\&Sigma;}}}\underset{I=1}{\overset{8}{\mathrm{\&Sigma;}}}U(k,L)$ Or $\mathrm{Ui}=\frac{1}{16}\underset{K=1}{\overset{4}{\mathrm{\&Sigma;}}}\underset{I=1}{\overset{4}{\mathrm{\&Sigma;}}}U(K,1)$

U wherein _I(k, 1) is the gray-scale value of (k, 1) pixel in the i piece subimage block, U ₁Be the average pixel gray-scale value of i piece subimage, according to U ₁Size, determine the binary-state threshold size of corresponding i piece subimage block; Image to be processed is scanned, and (k i), detects its gray-scale value U (k, 1) and the size of each point gray-scale value all around thereof, according to condition: the gray-scale value U of this point (k, 1)＜threshold value T to each picture element that is scanned ₁, the gray-scale value U of this point (k, 1)＜T ₂And has any gray-scale value＜T in 4 of the upper and lower, left and right at least ₁Judge gray-scale value＜T _IThe time, then this is the picture element of sperm image, otherwise the picture element that this point is not the sperm image.

3. want 1 to ask said sperm fluorescent staining computer detection system as right, it is characterized in that: said related computing is the computing that corresponding relation between the location parameter is respectively organized in a kind of searching, carries out according to distance criterion, speed criterion, direction criterion;

Distance criterion

For { the some elements A among the Ai} _kKAk ∈ 1,2....I}, computed range d=D (A _k, B _j) ‖ j=1,2...., J,

If d＜r, then B _jBe sperm a _IPossible predicted position point parameter,

If d＞r, then B _jBe sperm a _INonanticipating location point parameter,

{ Ai|i=1 wherein, 2......., I} is a certain location parameter group, ai|i=1, and 2......I} by correspondence detect the set of sperm, { Bj|j=1,2...., J} is that { r is called the differentiation radius of distance criterion to parameter group for the adjacent subsequent parameter group of Ai}, r=shooting time interval of delta t x sperm speed limit Vmax;

The speed criterion

When occurring a plurality of possible sperm predicted position in the radius r institute restricted portion, first transient motion speed under may predicted position is predicted that getting wherein, that possible position of velocity variations minimum is the predicted position of sperm, be B=Bi, and Bi ∈ { B1, B3}, V _A-V _Bi=min (V _A-V _Bi, V _A-V _B3), wherein a is a sperm, V _AFor just being speed, B is a predicted position, V _B3=D (A ₁B ₃)/Δ t is an instantaneous velocity;

The direction criterion

If | SB-SA≤Δ S| or | (SB-SA) mod16|≤Δ S sets up, and then position, position S is desired predicted position, otherwise is not predicted position just

S wherein _ABe the original direction of motion of sperm, S _BBe the direction of motion by predicted position gained that B calculates, Δ S for the change threshold of setting direction of motion again is.

4. want 1 to ask said sperm fluorescent staining computer detection system as right, it is characterized in that: said kinematic parameter calculates and comprises

(1) track of sperm motility

With location point A _i(j) variation with i couples together, and obtains the movement locus of desired sperm aj, obtains J bar sperm motility geometric locus altogether with j=1 to the variation of j=J;

(2) linear velocity of sperm motility: $V\left(j\right)\left[\underset{j=1}{\overset{N-1}{\mathrm{\&Sigma;}}}\left(D(\mathrm{Ai}\left(j\right),A_{i+1}\left(j\right))\right)\right]/\left[(N-1)\mathrm{\&Delta;t}\right],j=\mathrm{1,2}......J$

(3) space rate of sperm motility: V (j) D (A _I(j), A _n(j))/[(N-1) Δ t], j=1,2......, J

(4) sperm motility line average velocity and straight line average velocity $V=\frac{1}{J}\underset{J=1}{\overset{J}{\mathrm{\&Sigma;}}}V\left(j\right)$ $V=\frac{1}{J}\underset{J=1}{\overset{J}{\mathrm{\&Sigma;}}}V\left(j\right)$

(5) sperm motility linear speed difference and space rate variance ${\mathrm{\&delta;}}_V^2=\frac{1}{J}\underset{j=1}{\overset{J}{\mathrm{\&Sigma;}}}{(V\left(j\right)-V)}^2$ ${\mathrm{\&delta;}}_V^2=\frac{1}{J}\underset{j=1}{\overset{J}{\mathrm{\&Sigma;}}}{(V\left(j\right)-V)}^2$

5. sperm fluorescent dyeing method is characterized in that using following reagent A: reagent A is that the concentration with the damping fluid of pH3～7.8 or physiological saline or distilled water preparation is 0.001～0.5% acridine orange fluorchrome.

6. want 1 to ask said sperm fluorescent staining computer detection system as right, it is characterized in that: said system also comprise to test sample heat, insulation and dustproof constant temperature dust proof member.

Images