TW200424526A - Bioagent automatic detection system - Google Patents

Abstract

A bioagent automatic detection system comprises a detection machine and a data processing device. The detection machine is installed with a rail for a bioagent moving in-and-out of the housing space of the detection machine. The housing space is illuminated by two light sources installed symmetrically and in tilt. A camera is used to take the image of a bioagent. The image is transmitted to a data processing device for image treatment. The invented system calculates the image photo intensity in the photo field measurement region between a measurement line and a control line of the bioagent in order to obtain an average gray value as a graystand, and divides the image of the measurement line into n regions. Next, the system compares the graystandard with the control line portion and the image photo field intensity threshold Tn of each region of the measurement line to determine the detection result.

Description

200424526 V. Description of the invention (l) '-------- »[Technical field to which the invention belongs], the Department of Scholarship ^ related to an automatic detection system for biological reagents, which has the advantages of fast and sensitive detection. This detection system uses a backlight board with a μ lens and an electric contact to match the innovative shirt for this detection system! The processing rule is used to identify the detection result to indicate the result of its reaction. [Previous technology] According to the press, hepatitis C is a type of liver inflammation caused by Hepatitis c virus (HCV). The main route of transmission of hepatitis C is blood and body fluids that enter the body through the skin or mucous membranes. HCV infection that enters the blood later can cause chronic hepatitis. Usually, such chronic hepatitis has no obvious symptoms. L is usually found in a fixed medical examination. Hepatitis C can cause severe liver complications. Patients often die from severe cirrhosis or hepatocellular carcinoma after 15-20 years of infection. After infection with HCV, patients will develop anti-HCV antibodies to fight against invading HCV, and anti-HCV antibodies are the main targets of automatic biological reagent detection systems. = Le ^ r〇tein) is fixed at the position of the test line, and combined with hottein G as a reagent weaving coloring agent, mainly detects the heart in the specimen =, I gM, I g A), as a fast and accurate judgment of human blood Is there a text HCV Wei Bi to help individuals Jimu Yiba individuals and physicians to make early diagnosis. Serum or blood, ^ ^, ^ X blood water is used in the form of Caihuan body, which is tested by centrifugation after blood drawing. For temporary storage, ...- 8. C. If required, the monthly storage must be at -20. C environment. The reagents used in the biological reagent automatic detection system apply immunoassay detection technology and chromatography principles to convert Hcv recombinant proteins (NS3, NS4, NS5,

200424526 V. Description of the invention (2) · 4 Because the sensitivity of the biochemical rapid test reagent is much higher than the traditional test method, this difficulty can be overcome, so the biochemical rapid test reagent has the advantages of fast and high sensitivity. As shown in Figure 1, it is a schematic diagram of a rapid hepatitis test reagent. The biological reagent 8 has an action zone 8 1 for dripping into the blood or body fluid of the subject, and a control line 8 for determining whether the reagent is an effective product. 3 and a test line 8 2 for viewing the results of the biochemical reaction, and the area where the control line 8 3 and the test line 82 are located is collectively referred to as a test area A. If this reagent is a qualified product, a color will appear on the control line 8 3, and the biochemical reaction result after the test serum drips into the action area 81 will appear on the test line 8 2, so as to judge by the reaction color Whether the blood is infected with hepatitis. Because the biochemical reaction uses fluorescence, color arrangement, combination, etc. to judge the results of the reaction, when it is detected visually, general inspectors are prone to deviations due to subjective judgment during visual inspection. Therefore, the machine vision system can be used to identify the arrangement of colors and identify the inspection results. With the development of biochemical technology and the birth of biochips, various biochemical sensors, etc., machine vision technology has become a trend for the identification and detection of biochemical reactions, such as Vic 〇r N. Μ〇r ο ζ ο v Et al. (2001) proposed a method for extracting and measuring array protein wafers using a CCD. Alberto Delgado proposed in 2002 a DNA array microarray inspection system. Joon Myung Song et al. Proposed a method for detecting DNA using CMOS in 2002. And because the test line 8 2 on the biological reagent 8 is usually made by roll coating, it does not have very good uniformity, and the edges often have blurring and uneven concentration. And test line 8 2 must be slanted

200424526 V. Description of the invention (3) 1 ^ Only the light can highlight the detection result, but when the oblique light is irradiated on the test area A of the biological reagent 8, it also causes a problem of gradation change in the detection image, so the reference light field is set. In the measurement area, the light field intensity of the reference light field measurement area must be used as the basis for reference detection. The material of the test strip is to be firmly combined with the specimen, so it causes the surface of rice paper or cloud paper, and the detection image often has the effect of dyeing, which brings noise disturbance. These problems will affect the accuracy of machine vision technology to identify the detection results of biological reagents, and even cause misjudgment. Therefore, how to develop a system that can accurately identify the results of biochemical reactions of biological agents has become the subject of the present invention. [Summary of the Invention] The main object of the present invention is to solve the above-mentioned problems and provide a camera to capture the image of the tested area of the biological reagent, and then cut the image of the test line part of the biological reagent into n areas, so that By comparing the thresholds of each area, it is possible to accurately determine the efficacy of the biological reagent test result. In order to achieve the aforementioned object, the biological reagent automatic detection system of the present invention includes: a testing machine having a containing space therein, and the testing machine having an openable and closable door for biological reagents Enter and exit the accommodating space, and the testing machine has several heat dissipation holes for air circulation; a sliding handle, which is erected in the accommodating space of the testing machine, and a positioning auxiliary rod is provided at the end of the sliding machine. The biological reagent enters and leaves the accommodating space through the slider, and the positioning auxiliary rod provides positioning function; a light source group, which is composed of two light sources, and the two light sources are erected at intervals.

200424526 V. Description of the invention (4) · * At the test area corresponding to the biological reagent above the slider, and the two light sources are arranged in a symmetrically inclined form, and the direction in which the two light sources are inclined is perpendicular to the slide rail. In the direction of extension, radiating fins 贴 are attached to each light source to enhance the cooling effect; a camera is set between the two light sources of the light source group and corresponds to the test area of the biological reagent; more than one fan It is set up in the accommodating space, so as to generate airflow, and the heat in the accommodating space is discharged through several heat dissipation holes; a data processing device is connected with the camera and obtains the camera The biological reagent image is image processed, and the image light field intensity of the light field measurement area between the test line and the control line of the biological reagent is calculated to obtain the average gray value as the threshold value Graystandai * d, and the test line The image is vertically and horizontally divided into n areas, and the threshold value Gray standgnrd is compared with the image light field intensity threshold τn in each area of the I line and the test line. Any area of the image is compared. When the average gray value Grayave is lower than the area threshold Tn, the area is a positive reaction, and the color gray average value Grayp of all positive reaction areas is calculated, and the level of the biochemical reaction concentration is calculated based on the gray average value. If the average gray value Grayave is higher than the regional threshold Tn, the detection result of the biological reagent is determined as a negative reaction. The reagents used in the biological reagent automatic detection system of the present invention apply immunoassay testing technology and chromatography principles to fix HCV recombinant proteins (NS3, NS4, NS5, core protein) at the position of the test line, and the specimen rises to the position of the test line 1 0 ~ 1 5 minutes, color judgment with visual detection system, computer image acquisition, detection results can be known within 0.3 seconds,

200424526 V. Description of the invention (5) * · High sensitivity and easy operation. The above and other objects and advantages of the present invention can be easily understood from the detailed description and accompanying drawings of the selected embodiments below. Of course, the present invention allows some differences in the arrangement or arrangement of other parts, but the selected embodiment is described in detail in this specification and its structure is shown in the drawings. [Embodiment] Please refer to FIG. 1 to FIG. 16. The figure shows the structure of the embodiment of the present invention. This is for illustration only. It is not limited by this structure in patent applications. The biological reagent automatic detection system of the present invention includes a testing machine 1 part of hardware and a data processing device part of software. The testing machine 1 has an accommodating space 1 1, and the testing machine 1 has an openable and closable door 12 for the biological reagent 8 to enter and leave the accommodating space 1 1, and the testing machine 1 has two There are several heat dissipation holes 1 3 on the side for air circulation. · The accommodation space 1 1 of the inspection machine 1 is provided with: a sliding handle 14 which is erected in the accommodation space 1 1 of the inspection machine 1 and a positioning aid is provided at the end of the slide rail 14 The rod 1 4 1 is used for the biological reagent 8 to enter and leave the accommodation space 1 1 through the slider 14 and the positioning auxiliary rod 1 4 1 is used for positioning. A light source group is composed of two light sources 15; in this embodiment, the light source is a yellow light backlight plate; the two light sources 15 are spaced apart from the test area A corresponding to the biological reagent 8 above the slider 14 Place while the two lights

200424526 V. Description of the invention (6) *, the source 15 is symmetrically inclined, and the directions of the two light sources 15 are perpendicular to the extension direction of the slider 14 and separately on each light source 15 Attach heat sink fins 1 5 1 俾 to enhance heat dissipation. A camera 16 is set up between the two light sources 15 of the light source group and corresponds to the test area A of the biological reagent 8. Two fans 17 are set up in the accommodating space 1 1 for the heat dissipation holes 13 on both sides of the testing machine 1 to generate air currents and the heat in the accommodating space 11 is divided into several heat dissipation holes 1 3 discharge. The data processing device 3 is connected to the camera 16 and performs image processing on the biological reagent 8 image obtained by the camera 16, and the light field between the test line 8 2 and the control line 8 3 of the biological reagent 8 Calculate the light field intensity of the measurement area B to obtain the average gray value as the threshold value Gray standard 5 and vertically and horizontally divide the image of the test line 8 2 into π regions C 5 and use the threshold value Graystanda) rd and control Line inspection area D and test line 82 compare the light field intensity threshold value Tn of each area C image. When the average gray value Grayave of any area C image is lower than the area threshold value Tn, the area C is a positive reaction, and Calculate the grayscale average value Grayp of all positive reaction areas, and calculate the level of biochemical reaction concentration based on the average grayscale value. 'If the average grayscale value of Grayave in all areas is higher than the area threshold value τη, the detection result of the biological reagent is determined as Negative reaction. When taking the image of the biological reagent 8 with the camera 16, it is necessary to overcome the noise problem first. The present invention adopts a method of averaging multiple consecutive images to suppress the noise.

Page 11 200424526 V. Description of the invention (7) 1 1

Level where S (X, Y) is the signal value of the image; the noise size of the i-th acquisition of Ni (X, Y) assumes that the overall noise of the i-th acquisition is randomly distributed. The noise value of the overall image may be Represented as the clutter of the image. For example, the L & vei image is taken M times, and the average value is calculated, Υ)

GrayiXj) = —YGray, where Grayi (X, Υ) is the gray value of each single point; M is the derivation of the image size, and it can be obtained that after the image has been acquired M times, the image noise ratio will be Increase V'i times, suppress noise, refer to Figure 6 for reference. Furthermore, since the test line 8 2 on the biological reagent 8 is usually made by rolling application, it does not have very good uniformity, and the edges often have blurring and uneven concentration. In order to overcome this problem, the present invention performs the following operations on the image (as shown in Fig. 7) obtained by the camera 16: The binary image A is etched by B. The operation can be expressed as follows:

Page 12 200424526 V. Description of the invention (8) The mathematical expression of the etching operation of gray image f and b can be expressed as follows: ίβ ^) (β, =, ί +,)-start, (# +, ) Is coincident; (I, / ¾} where Df and Db are the regions of f and b. If degradation becomes a function of a unitary variable, the above formula becomes: (β3.δ) (^) = mini f (s + x) -b (x) \ (s + x) e Df and xe 〇b} where B is a mask with a range of 1 in the range of 6 * 1. Furthermore, when test line 8 2 has a significant reaction phenomenon At this time, the difference in the image of the light field measurement area B between the test line 8 2 and the control line 8 3 can be clearly seen in the image of the test area A of the biological reagent 8 in FIG. 7, and in FIG. 8 It is the form of 3D three-dimensional graphics after inverting the image in Figure 7, but when the test line 8 2 has no obvious reaction phenomenon, the image of the test area A of the biological reagent 8 shown in Figure 9 Therefore, it is difficult to determine the exact position of the test line 82. At this time, the data shown in Figs. 10 and 11 can be obtained through the foregoing calculations, and the exact position of the test line 8 2 can be clearly determined. In order Perform the following tasks: 1 Check whether the control line 8 3 is normal to the control line 8 3 Check whether the control line 8 3 is working properly in the inspection area D. 2 Measure the geometric center position of the control line 8 3 and estimate the test line 8 2 With the light field measurement area A.

200424526 V. Description of the invention (9) * Assuming the geometric center position of the control line 8 3 is P e; the geometric center position of the light field measurement area A is P 丨; the geometric center position of the test line 8 2 is P t; The difference between the geometric center distance between the control line 8 3 and the test line 8 2 is Δd.

Ad

Mi ^ = ^ + T; = 3 · Light field measurement In order to ensure that the external environmental factors are kept fixed at each inspection to reduce the error caused by interference, it is necessary to measure the light intensity of the light field at several specific positions. Test to grasp the degree of change in the test environment. Assume that the average gray value of the light field measurement area B is, and use this value as the reference for the threshold value in the test line 8 2 inspection area. 4 · According to the light field changes, the value of the auto-adjustment threshold should be fixed in theory. However, the error in the light intensity of the light source system and the color drift of the camera's photosensitive chip cause the overall color of the captured image to be biased. The phenomenon of migration occurs. In addition, the light field intensity of the light field measurement area and the test line 8 2 have a nearly constant difference in the light field intensity of each divided area C in the inspection area. As shown in FIG. 12, the area C 1 and the area C 2 There are differences in light field intensity. Therefore, the threshold of the divided area can be automatically adjusted according to the change of the light field to reduce the error caused by the light source. Assume that the threshold value of each divided area in the test line inspection area is; the average gray value of the light field measurement area; the difference in light intensity between the light field measurement area and the divided area is%, ^ = 1 ^ 30;

Page 14 200424526 V. Description of the invention (, ο)-,-T? 2 = Gray ^^ + 5 · The test line for detecting biological reagents In the present embodiment, the test line 8 2 is divided into 30 small tests Area, each small area is set with different thresholds depending on the light field, which is used to check whether the test line 8 2 reacts. In addition, the number of pixels of positive reaction in the segmented area C in the test area is often smaller than the number of pixels of the negative reaction. In order to avoid the average gray value of the segmented area C being misjudged as a negative reaction, the When the gray value is averaged, the gray value of the pixel that may be a positive response will be increased to improve the accuracy of the judgment.

Assuming regional threshold 1, ^ = 1-30; regional average gray value; gray value of pixels smaller than the regional threshold; number of pixels smaller than the regional threshold; pixel gray weight weights smaller than the regional threshold collapse; greater than the area The threshold gray value of the pixel is Gray "x, y !; the number of pixels greater than the regional threshold is Lion_. V (χ5 v) X ^ 4- V Gra (x, 7 j

Gray — = —-—- two-… hemp "+ shirt ~

6 · Judging the reaction result When the reaction color of any divided area in the test area Gra & gt ::; is lower than the area threshold, and η = 1 ~ 30, it is a positive reaction, and all positives are calculated.

Page 15 200424526 V. Description of the invention (11) The average value of the grayscale of the 4 and 1 reaction areas, and the level of the biochemical reaction concentration is calculated based on the average value of the grayscale. If there is no response, it is determined to be negative. Positive reactions: Assume that there are Μί ·· ρ_ ~ segmented regions that show positive reactions; the reaction color of the segmented regions that show positive reactions; regional threshold, π = 1 ^ 30; all positive reactions The average value of the color gray scale of the area; the average gray value of the light field measurement area; the level of the biochemical reaction concentration is; the level interval is 〃.

Grav.

Level

Page 16 200424526 V. Explanation of the invention (12) · a, 2 · Negative reaction (N e g a t i v e): When the test result is. Only the control line shows a color band, it means that the blood of the subject does not contain Anti-HCV antibodies. 3 · Invalid: If there is no color band on the test result or no color band on the control line, it means that the reagent has expired, expired, or operated improperly, and another test is required. The data processing device 3 of the present invention is a personal computer, and the image is received to the computer through a USB2.0 transmission interface (the transmission rate is up to 480Mbps). Using the automatic visual inspection rule proposed by the present invention, the biochemical reaction can be identified in real time to determine whether it is Positive or negative reaction. If it is a positive response, then the level of the reaction concentration is calculated to provide objective and accurate data as an important basis for medical personnel to diagnose hepatitis C. In terms of operation, the user only needs to place the biological reagent 8 to be tested in the slider 1 4 of the testing machine 1, and then press the start detection button in the software of the data processing device 3, and the system will automatically recognize the biochemical reaction. The results are output to a table, and the location of the positive reaction area is marked in the image. As shown in Figure 13 is the result of detecting a negative reaction; as shown in Figures 14 to 16 is the result of detecting a positive reaction. Among them: No. 14; the picture shows the concentration of non-biochemical reaction temple level = 50. Figure 15 shows the state of ㈣ 丨 = 75 in the biochemical reaction concentration level. Figure 16 shows the status of the biochemical reaction concentration level = 100. In summary, the present invention uses a camera 16 to capture the biological agent 8

200424526 V. Description of the invention (13)-x «Measure the image of area A, and then cut the image of the 2 part of the test line 8 of the biological reagent 8 into n areas, and can compare the thresholds of each area to achieve accuracy Judge the efficacy of biological reagent test results. The disclosure of the embodiments described above is used to illustrate the present invention, and is not intended to limit the present invention. Therefore, any change in the value or replacement of equivalent components should still belong to the scope of the present invention. From the above detailed description, those skilled in the art can understand that the present invention can indeed achieve the aforementioned purpose, and it has indeed complied with the provisions of the Patent Law, and filed a patent application.

Page 18 200424526 Simple illustration of the diagram, alas, [Simplified illustration of the diagram]

Figure 1 is a schematic diagram of the appearance of a biological reagent. Figure 2 is a schematic diagram of the configuration of the detection system of the present invention. Figure 3 is a schematic diagram of the structure of the testing machine of the present invention. Figure 4 is a schematic diagram of the structure of the testing machine of the present invention. Figure 5 is a schematic diagram of the biological reagent image detection area of the present invention. Figure 6 is a single-point schematic diagram of the present invention when filtering out noise. Figure 7 is an image of biological reagent 1 obtained by the camera of the present invention. Figure 8 is The three-dimensional coordinate value of the image of the biological reagent 1 in the present invention. FIG. 9 is the image of the biological reagent two obtained by the camera of the present invention. The tenth image is the three-dimensional coordinate value of the image of the biological reagent two in the present invention. The first 11 images are in the present invention. Figures of the binarized numerical value obtained after the binary image of the biological reagent are calculated by binarization. Figure 12 is a schematic diagram of the light field distribution in the present invention. Figure 13 is a display of the negative reaction result of the present invention. The picture shows the day-to-day display of the biochemical reaction concentration level of positive reaction detected by the present invention / = 50 0 state 15 The picture shows the day-to-day display of the biochemical reaction concentration level of positive reaction detected by the present invention iew / = 75

Figure 16 shows the concentration level of the biochemical reaction detected by the present invention. Νβ = 1 〇 〇 State display day surface [Illustration of the drawing number] Detection machine 1 Storage space 1 1 Door 1 2 Heat dissipation hole 1 3

Page 19 200424526

Page 20

Claims (1)

200424526 6. Scope of Patent Application " * 1 · An automatic biological reagent detection system includes: a testing machine, the testing machine has an accommodation space, and the testing machine has a door that can be opened and closed to A biological reagent is provided in and out of the accommodating space, and the detection machine has a plurality of heat dissipation holes for air circulation; a slide holder is erected in the accommodating space of the detection machine, and the end of the slide holder is provided A positioning auxiliary rod is provided for the biological reagent to enter and exit the accommodation space through the slider, and the positioning auxiliary rod provides positioning function; a light source group 'which is composed of two light sources', and the two light sources are spaced apart from the slider. The upper part corresponds to the test area of the biological reagent, and the two light sources are arranged in a symmetrically inclined form, and the direction in which the two light sources are inclined is perpendicular to the extension direction of the slide, and the heat radiation fins are attached to each light source.加强 In order to enhance the heat dissipation effect; a camera is set up between the two light sources of the light source group and corresponds to the test area of the biological reagent; more than one fan is It is set in the accommodating space, and the heat in the accommodating space is exhausted through several heat dissipation holes to generate airflow; a data processing device is connected with the camera and the biological reagent image obtained by the camera Perform image processing, use a mask with a range of 1 in the range of 6 * 1 to perform the etching operation to eliminate the noise of the dyeing effect on the reagent, and perform the following tasks in order; (1) · detect whether the control line is normal to the control line Check whether the control line works normally in the inspection area; 200424526 VI. Patent application scope 'One' (2) · Calculate the geometric center position of the control line, and estimate the position of the test line and the light field measurement area A. Measure the light field between the test line of the biological reagent and the control line. Area image light field intensity calculation to obtain the average gray value as the reference threshold, Gray ^ andard, and the image of the test line is divided vertically and horizontally into n areas, and the reference threshold Graystandard and the control line part and the test line The light field intensity threshold τn of each area image is compared. If the average gray value of any area image is lower than the area threshold value Tn, the area is a positive reaction, and the average gray value of all positive reaction areas is calculated. The value is ^ rayp, and the level of biochemical reaction concentration is calculated according to the average value of gray. The average gray value of all areas on the right is higher than the area threshold. “The detection result of the biological reagent is judged as a negative reaction; the judgment and processing rules are as follows: (1) The threshold value of the light field intensity of each area 7] 1 and the fixed difference between the light field intensity of the light field measurement area is ω η, then the light field intensity threshold τη of each area = the reference threshold G raystandard + ω η, at the data place, the device will calculate the threshold value τη of each area, and then compare the threshold value τ η of each area with the image light field intensity of the test line to reduce the cause The error caused by the light source; (2) the average gray value of each area Grayave that the data processing device wants to obtain; the gray value of the pixels less than the area threshold Gray_ (X, y) 'is less than the area threshold NumlQW; Grayscale weight value ω of the pixel smaller than the regional threshold; Grayhigh (X, y) of the gray value of the pixel larger than the regional threshold; the number of pixels larger than the regional threshold Page 22 200424526 VI. Patent application scope · · · N um high 5 The regional average degree value is r > r? _ 1: Qiao Mu Shi, ·, where Gray ^ Uy) ^, thereby increasing the number of possible positive reactions Pixel gray weight value to improve the accuracy of judgment; (3) The data processing device calculates the biochemical reaction concentration level Level p J \ Gray. The formula is “Noisy” ——— Shi Ding, — “, where ϋ%> 0, Gray ^ < 7: Nump is the number of blocks showing a positive response, and η is the interval of the concentration range of biochemical reactions. 2. The biological reagent automatic detection system according to item 1 of the scope of the patent application, wherein each light source of the light source group is a yellow light backlight board. 3. The biological reagent automatic detection system according to item 1 of the scope of the patent application, wherein both sides of the detection machine have several heat dissipation holes, and the accommodation space is close to the heat dissipation holes on both sides of the test machine, respectively. Equipped with a fan. Page 23