WO2019000740A1 - 微柱凝胶卡凝集检测结果识别系统及血型分析仪 - Google Patents

微柱凝胶卡凝集检测结果识别系统及血型分析仪 Download PDF

Info

Publication number
WO2019000740A1
WO2019000740A1 PCT/CN2017/107930 CN2017107930W WO2019000740A1 WO 2019000740 A1 WO2019000740 A1 WO 2019000740A1 CN 2017107930 W CN2017107930 W CN 2017107930W WO 2019000740 A1 WO2019000740 A1 WO 2019000740A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
area
bright area
module
bright
Prior art date
Application number
PCT/CN2017/107930
Other languages
English (en)
French (fr)
Inventor
闫晓磊
吴锋
沙利烽
Original Assignee
苏州长光华医生物医学工程有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 苏州长光华医生物医学工程有限公司 filed Critical 苏州长光华医生物医学工程有限公司
Publication of WO2019000740A1 publication Critical patent/WO2019000740A1/zh

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/80—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving blood groups or blood types or red blood cells
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/8483—Investigating reagent band

Definitions

  • the present invention belongs to the technical field of medical devices, and relates to a microcolumn gel card agglutination detection result recognition system and a blood type analyzer.
  • clinical blood testing generally adopts a simple detection method with clear results and stable results, mainly using biochemical gel filtration technology and centrifugation technology and immunological antigen-antibody specific reaction products.
  • concentration of the gel controls the size of the gel gap so that the gap allows only free red blood cells to pass, thereby separating free red blood cells and aggregated red blood cells.
  • the detection principle is: When the antigen and antibody react, the blood cells are agglutinated, and the agglutination block cannot pass through the gel gap, but remains in the upper layer of the gel tube, and is positive. The unaggregated blood cells are centrifuged and passed through the gel gap and deposited at the bottom of the gel tube, which is a negative reaction.
  • the result of agglutination reaction in a microcolumn gel card microtube is generally obtained by using a camera to collect an experimental result image, and then automatically processing the image, and then performing a result according to the processed image. Automatic judgment.
  • images of the reaction area of a fixed size in the microtube are directly intercepted on the acquired image for analysis, and then the judgment result is given according to the judgment standard of the blood agglutination reaction.
  • the criteria for judging the blood agglutination reaction are shown in Fig. 1.
  • the detection results are divided into “yang 4", “yang 3", “yang 2", “Yang 1”, “weak agglutination”, “mixed agglutination”, “hemolytic” and negative. Therefore, image analysis needs to accurately determine the position and approximate relative area of the agglutination reactants in the microtubules.
  • the above method of directly intercepting the image of the reaction area of the fixed size in the microtube is only applicable to a microcolumn gel card of a fixed size, and when the width of the microcolumn gel card used is different, the intercepted reaction area is It is not comprehensive, or the interception area is too large, which may affect the final analysis result and the case of misjudgment.
  • the present invention provides a microcolumn gel card agglutination detection result recognition system, which is suitable for various micro-column gel cards of different specifications, and effectively improves the accuracy of the identification of agglutination detection results. .
  • a microcolumn gel card agglutination detection result recognition system comprising:
  • an image acquisition module configured to collect an image of a microcolumn gel card after agglutination reaction and centrifugation
  • a micro tube image intercepting module configured to intercept a microtube image in the microcolumn gel card image
  • a target area acquiring module configured to acquire a target area to be identified in the micro tube image, where the target area is an image area within the inner wall of the micro tube;
  • an effective reactant region capture identification module for capturing an effective reactant region within the target domain, identifying a location and a relative area of the captured effective reactant region
  • a standard setting module configured to preset a judgment standard by a user
  • a judgment output module configured to determine and output an agglutination detection result of the microtube.
  • the micro tube image capturing module includes a clipping frame and a clipping frame control unit; the intercepting frame control unit is configured to control a length, a width, and a position of the clipping frame; and the intercepting frame is configured to intercept an image. .
  • the target area obtaining module includes:
  • an image conversion unit configured to convert the color image into a grayscale image
  • a binarization processing unit for binarizing the grayscale image
  • a fitting line unit for performing linear fitting or curve fitting according to boundary pixel points identified by the binarization map, and drawing an inner wall boundary line of the microtube;
  • an image deletion unit configured to delete an image.
  • the effective reactant region capture identification module comprises:
  • a bright area search and identification unit configured to find and identify a position of the bright area in the micro tube
  • a sub-bright area search and identification unit configured to find and identify the bright area in the micro The position and relative area within the tube
  • the bright area is an image area in which the gray value in the target area is greater than or equal to a first threshold; the second bright area is a gray level value in the target area is greater than or equal to a second threshold, and is smaller than the first a threshold image area; the first threshold is greater than the second threshold.
  • the first threshold gray value is in a range of 108-128; and the second threshold gray value is in a range of an average gray value of the target area image ⁇ 10.
  • the criterion determined by the standard setting module is:
  • the bright area is distributed on the top of the micro tube and there is no sub-bright area, then it is judged as "yang 4";
  • the bright area is distributed on the top of the micro tube and has a sub-bright area, and is judged as "yang 3";
  • the bright region is distributed at the bottom of the microtube, the sub-bright region and the sub-bright region is distributed in the middle and lower part of the microtube, and is judged as "yang 1";
  • the bright region is distributed at the bottom of the microtube, the sub-bright region and the sub-bright region is close to the bright region of the bottom, and is judged as "weak agglutination";
  • the bright region is distributed in all regions of the microtube, and is judged as "hemolytic";
  • a bright region is distributed at the bottom of the microtube and there is no sub-bright region, and it is judged as "negative".
  • the determining output module comprises:
  • the coarse judging unit performs a rough classification judgment on the result according to the judging criterion preset by the standard setting module according to the position of the bright area identified by the bright area search and recognition unit;
  • a fine judging unit based on the coarse classification, determining a position and a relative area of the sub-bright area identified by the sub-bright area search and recognition unit, and comparing with a criterion determined by the standard setting module, Performing a fine classification judgment on the result, and obtaining a result;
  • an output unit configured to output a result determined by the fine judgment unit.
  • the output of the output unit further includes an original image acquired by the image acquisition module.
  • the result confirmation module is further included, and the result output by the judgment output module needs to be manually confirmed.
  • the present invention also discloses a blood type analyzer for identifying a system using the above-described microcolumn gel card agglutination detection result.
  • Microcolumn gel card agglutination detection result recognition system Microcolumn gel card agglutination detection result recognition system:
  • FIG. 1 is a schematic view showing an embodiment of a microcolumn gel card agglutination detection result identification system according to the present invention
  • FIG. 2 is a schematic view showing another embodiment of the microcolumn gel card agglutination detection result identification system of the present invention
  • FIG. 3 is a schematic diagram of a microtube image of a predetermined criterion and a corresponding result
  • FIG. 4 is a schematic diagram of operation of a micro tube image capture module on a software window
  • FIG. 5 is an image of a captured microtube
  • FIG. 6 is a binarized diagram of the microtube image taken in FIG. 5; [0048] FIG.
  • FIG. 7 is an image of a target area
  • FIG. 8 is a schematic view of a bright area identified in the image of the target area of FIG. 7.
  • A bright area
  • B sub-bright area
  • the microcolumn gel card agglutination detection result recognition system of the present invention comprises: an image acquisition module, configured to collect a microcolumn gel card image obtained by agglutination reaction and centrifuged; microtubule image An intercepting module, configured to intercept a microtube image in the image of the microcolumn gel card; and a target area acquiring module, configured to acquire a target area to be identified in the image of the micro tube, wherein the target area is an inner wall of the micro tube Within An image region; an effective reactant region capture recognition module, configured to capture an effective reactant region in the target domain, and identify a location and a relative area of the captured effective reactant region; a standard setting module, configured for the user to preset a judgment criterion; and a judgment output module for judging and outputting the agglutination detection result of the microtube.
  • the micro tube image intercepting module comprises a truncation frame 1 and a truncation frame control unit 2, as shown in FIG. 4 is a schematic diagram of operation of the micro tube image intercepting module on a software window, and the truncated frame control unit 2 can control the adjustment interception.
  • the length, width and position of frame 1 are adapted to different specifications of the microcolumn gel card, so that the intercepting frame 1 can intercept the complete image of each microtube 3 of the microcolumn gel card, and ensure the reaction column of the gel column in the microtube 3 All intercepted to
  • the target area is an image area within the inner wall of the peripheral portion of the inner wall of the microtube, that is, the gel column reaction area in the microtube 3, as shown in FIG.
  • the target area acquisition module includes: an image conversion unit, a binarization processing unit, a fitting line drawing unit, and an image deletion unit.
  • the image conversion unit is configured to convert the color image into a grayscale image
  • the binarization processing unit is configured to process the grayscale image by binarization
  • the fitting line drawing unit is configured to perform the boundary pixel according to the binarization map. Linear fitting or curve fitting, and drawing the inner wall boundary line of the microtube; image deletion unit for deleting the image.
  • the micro tube image intercepted by the truncation frame is first converted into a gray image by using an image conversion unit, and the gray image is binarized by a binarization processing unit.
  • pixel points belonging to the inner wall of the micro tube are identified on the binarized image, and linear fitting or curve fitting is performed on the pixel points, for example, a linear fitting method is used to fit the boundary line of the straight body inner wall, and the curve is utilized.
  • the fitting method fits the boundary line of the inner wall of the tube at the bottom of the curve, and traces the boundary line of the inner wall of the microtube which is fitted above.
  • the image deletion unit is used to delete the image of the periphery of the boundary line of the microtube inside the above-mentioned fitting.
  • the target area that needs to be analyzed as shown in Figure 7.
  • the grayscale conversion, binarization processing and fitting method of the system image automatically identify the inner wall boundary of the microtube, and delete the interference image of the outer wall boundary of the microtube without the analysis target, thereby obtaining more complete and more accurate image information. Ensure the accuracy of subsequent analysis and judgment to obtain more accurate identification results; also make the system suitable for the identification of agglutination detection results of various micro-column gel cards, and improve the system in different specifications of the analyzer. Versatility.
  • Embodiment 2 On the basis of Embodiment 1, the target area is identified using an effective reactant area capture identification module.
  • the effective reactant region capture identification module comprises: a bright region search and recognition unit, configured to find and identify a position of the bright region in the microtube; and a sub-bright region search and recognition unit, configured to find and identify the bright region in the micro The position and relative area within the tube.
  • the bright area is an image area in which the gray value in the target area is greater than or equal to the first threshold, that is, an area in which the red blood cells in the micro tube are concentrated, as in the area eight in FIGS. 3, 7, and 8. According to experience, the first threshold gray value is generally taken in the range of 108-128.
  • the sub-bright area is an image area in which the gray value in the target area is greater than or equal to the second threshold and is smaller than the first threshold, that is, the area in which the red blood cells are scattered in the micro tube, as in the area of FIG. 3, the first threshold is greater than the second threshold.
  • the second threshold gradation value is then taken within a range of an average gradation value of ⁇ 10 of the target area image. The system can automatically perform value debugging within the above range, and determine an optimal value of the first threshold and the second threshold applicable to the image according to the actual gray value distribution of each recognized image.
  • the judging output module judges and analyzes the recognition result according to the judgment standard preset by the standard setting module. And output the recognition result.
  • the preset criterion of the standard setting module is:
  • the bright area is distributed on the top of the micro tube and there is no sub-bright area, then it is judged as "yang 4";
  • the bright area is distributed on the top of the micro tube and has a sub-bright area, and is judged as "positive 3";
  • the bright region is distributed at the bottom of the microtube, the sub-bright region and the sub-bright region is distributed in the middle and lower part of the microtube, and is judged as "yang 1";
  • the bright region is distributed at the bottom of the microtube, the sub-bright region and the sub-bright region is close to the bright region of the bottom, and is judged as "weak agglutination";
  • the bright areas are distributed on the top and bottom of the micro tube, and there is no sub-bright area, and it is judged as "mixed agglutination";
  • the bright region is distributed in all regions of the microtube, and is judged as "hemolytic";
  • a bright region is distributed at the bottom of the microtube and there is no sub-bright region, and it is judged as "negative".
  • the judging output module includes: a rough judging unit, searching for the position of the bright region identified by the recognizing unit according to the bright region A, and performing coarse classification judgment on the result according to the judging criterion preset by the standard setting module;
  • the fine judgment unit based on the above-mentioned rough classification, searches for the position and the relative area of the sub-bright area identified by the recognition unit according to the sub-bright area B, and compares the judgment criteria preset by the standard setting module, and performs fine classification judgment on the result.
  • the result is obtained; the output unit is used to output the result judged by the fine judgment unit.
  • the target region shown in FIG. 7 is identified by the effective reactant region capture recognition module, and the bright region search recognition unit recognizes that the bright region A is located at the top of the gel column in the microtube, and judges the rough judgment of the output module.
  • the unit initially determines that it is "yang 4" or "yang 3"
  • the fine judgment unit considers that the bright area in the target area is located at the top of the micro tube, and there is no bright area, and the reaction result is "yang 4". And output the result using the output unit.
  • the coarse judgment unit initially judges whether it is "positive 1", “weak agglutination” or "negative” after comparing the above criteria. Then, based on the identified position and relative area of the sub-bright area B, it is judged which one of the reaction results specifically belongs to.
  • the fine judgment unit considers that the "bright area is located at the bottom of the microtube and there is no sub-bright area" in the target area, and the reaction result is "negative”; if the sub-bright area B is found And identifying that the bright area B has a relatively large relative area and is distributed in the middle and lower part of the micro tube, the fine judgment unit considers that the bright area is located at the bottom of the micro tube, the sub-bright area and the sub-bright area are distributed in the micro area.
  • the reaction result is "yang 1"; if the relative area of the sub-bright area B is identified to be small and close to the bright area A at the bottom, the fine judgment unit considers that the "bright area is distributed in the target area" At the bottom of the tube, there is a bright area with a sub-bright area and a sub-bright area near the bottom. The result of the reaction is "weak agglutination".
  • the result output by the output unit may further include outputting an original image acquired by the image acquisition module.
  • the discrimination judgment performed in two steps further improves the accuracy of the result recognition.
  • the result confirmation module is added to the embodiment 1 or 2, and the result output by the above judgment output module needs to be manually confirmed. For example, after judging that the output module determines the result, the judgment result is outputted in the form of a pop-up window or other forms, and the result confirmation is manually performed, and the system ends the identification work after the confirmation. After manual confirmation, system misjudgment caused by instrument failure or other reasons is avoided.
  • the present invention also discloses a blood type analyzer, which adopts the above-mentioned microcolumn gel card agglutination detection result recognition system, and effectively improves the accuracy of blood group analysis.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Immunology (AREA)
  • Hematology (AREA)
  • Engineering & Computer Science (AREA)
  • Chemical & Material Sciences (AREA)
  • Molecular Biology (AREA)
  • Analytical Chemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Pathology (AREA)
  • Urology & Nephrology (AREA)
  • General Physics & Mathematics (AREA)
  • Biochemistry (AREA)
  • Biomedical Technology (AREA)
  • Physics & Mathematics (AREA)
  • Medicinal Chemistry (AREA)
  • Food Science & Technology (AREA)
  • Microbiology (AREA)
  • Biotechnology (AREA)
  • Cell Biology (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)
  • Investigating Or Analysing Biological Materials (AREA)

Abstract

一种微柱凝胶卡凝集检测结果识别系统，包括：图像采集模块、微管图像截取模块、目标区域获取模块、有效反应物区域捕获识别模块、标准设定模块和判断输出模块。该系统通过二值化处理单元和拟合描线单元自动拟合出微管内壁边界线，获取微管内壁边界线以内的反应区域作为分析的目标区域，避免了区域以外的图像部分干扰分析判断，有效提高了识别的准确性，同时使得该系统适用于各种不同规格的微柱凝胶卡，提高了该系统在不同规格的分析仪中的通用性；对分析的目标区域内有效反应物区域进行亮度位置和相对面积相结合的分类判断，再进行粗分类和细分类的对照分析，提高了结果识别的准确性。

Description

微柱凝胶卡凝集检测结果识别系统及血型分析仪 技术领域
[0001] 本发明属于医疗器械技术领域， 涉及一种微柱凝胶卡凝集检测结果识别系统及 血型分析仪。
背景技术
[0002] 目前临床血液检测普遍采用操作简单、 结果明确、 稳定的卡式检测法， 主要是 利用生物化学凝胶过滤技术和离心技术及免疫学抗原抗体特异性反应相结合的 产物， 通过调节凝胶的浓度来控制凝胶间隙的大小， 使其间隙只能允许游离的 红细胞通过， 从而使游离的红细胞和聚集的红细胞分离。 检测原理是： 当抗原 、 抗体反应， 血细胞发生凝集， 离心吋， 凝集块不能通过凝胶间隙， 而留在凝 胶管的上层， 则呈阳性反应。 未凝集的血细胞离心吋可通过凝胶间隙， 而沉积 在凝胶管的底部， 则呈阴性反应。
[0003] 在自动化的血液分析仪中， 对微柱凝胶卡微管内的凝集反应结果的判断一般是 采用相机采集实验结果图像， 再自动软件对图像进行处理， 然后根据处理后的 图片进行结果自动判断。 在现有的血液分析系统中， 都是在采集的图像上直接 截取微管内固定大小的反应区域图像进行分析， 然后根据血液凝集反应的判断 标准给出判断结果。 血液凝集反应的判断标准如图 1所示， 根据微柱凝胶卡微管 内的凝集反应物的位置和相对面积将其检测结果分为"阳 4"、 "阳 3"、 "阳 2"、 "阳 1"、 "弱凝集"、 "混合凝集"、 "溶血 "和阴性等。 所以图像分析需要精准地判断出 微管内凝集反应物的位置和大概的相对面积。 但是上述直接截取微管内固定大 小的反应区域图像进行分析的方法， 仅适用于某一固定尺寸的微柱凝胶卡， 当 使用的微柱凝胶卡的宽度不同吋， 则其截取的反应区域就不全面， 或者是截取 区域过大， 从而有可能影响了最终的分析结果， 出现误判的情况。
技术问题
[0004] 为解决上述问题， 本发明提出了一种微柱凝胶卡凝集检测结果识别系统， 适用 于各种不同规格的微柱凝胶卡， 同吋有效提高了凝集检测结果识别的准确性。 问题的解决方案
技术解决方案
[0005] 微柱凝胶卡凝集检测结果识别系统， 包括：
[0006] 图像采集模块， 用于采集经过凝集反应并离心完成的微柱凝胶卡图像；
[0007] 微管图像截取模块， 用于截取所述微柱凝胶卡图像中的微管图像；
[0008] 目标区域获取模块， 用于获取所述微管图像中需要识别的目标区域， 所述目标 区域为所述微管内壁以内的图像区域；
[0009] 有效反应物区域捕获识别模块， 用于在所述目标域内， 捕获有效反应物区域， 识别所捕获的有效反应物区域的位置和相对面积；
[0010] 标准设定模块， 用于用户预先设定判断标准；
[0011] 判断输出模块， 用于判断和输出所述微管的凝集检测结果。
[0012] 优选地， 所述微管图像截取模块包括截取框和截取框控制单元； 所述截取框控 制单元用于控制所述截取框的长度、 宽度和位置； 所述截取框用于截取图像。
[0013] 优选地， 所述目标区域获取模块包括：
[0014] 图像转换单元， 用于将彩色图像转换成灰度图像；
[0015] 二值化处理单元， 用于二值化处理灰度图像；
[0016] 拟合描线单元， 用于根据二值化图识别出的边界像素点进行线性拟合或曲线拟 合， 并描出所述微管的内壁边界线；
[0017] 图像刪除单元， 用于刪除图像。
[0018] 优选地， 所述有效反应物区域捕获识别模块包括：
[0019] 明亮区域査找识别单元， 用于査找并识别出该明亮区域在所述微管内的位置； [0020] 次明亮区域査找识别单元， 用于査找并识别出该次明亮区域在所述微管内的位 置及相对面积；
[0021] 所述明亮区域为所述目标区域内灰度值大于或等于第一阈值的图像区域； 所述 次明亮区域为所述目标区域内灰度值大于或等于第二阈值， 并小于第一阈值的 图像区域； 所述第一阈值大于所述第二阈值。
[0022] 优选地， 所述第一阈值灰度值在 108-128范围内； 所述第二阈值灰度值在所述 目标区域图像的平均灰度值 ±10的范围内。 [0023] 优选地， 所述标准设定模块预设的判断标准为：
[0024] 明亮区域分布于微管顶部且没有次明亮区域， 则判断为"阳 4";
[0025] 明亮区域分布于微管顶部且有次明亮区域， 则判断为"阳 3";
[0026] 没有明亮区域只有次明亮区域， 则判断为"阳 2";
[0027] 明亮区域分布于微管底部、 有次明亮区域且次明亮区域分布于微管中下部， 则 判断为 "阳 1";
[0028] 明亮区域分布于微管底部、 有次明亮区域且次明亮区域靠近底部的明亮区域， 则判断为"弱凝集"；
[0029] 明亮区域分布于微管的顶部和底部， 没有次明亮区域， 则判断为"混合凝集"；
[0030] 明亮区域分布于微管的全部区域， 则判断为"溶血"；
[0031] 明亮区域分布于微管底部且没有次明亮区域， 则判断为"阴性"。
[0032] 优选地， 所述判断输出模块包括：
[0033] 粗判断单元， 根据所述明亮区域査找识别单元识别出的明亮区域的位置， 对照 所述标准设定模块预设的判断标准， 对所述结果进行粗分类判断；
[0034] 细判断单元， 在所述粗分类的基础上， 根据所述次明亮区域査找识别单元识别 出的次明亮区域的位置和相对面积， 对照所述标准设定模块预设的判断标准， 对所述结果进行细分类判断， 得出结果；
[0035] 输出单元， 用于输出所述细判断单元判断得出的结果。
[0036] 优选地， 所述输出单元输出的结果还包括所述图像采集模块采集的原始图像。
[0037] 优选地， 还包括结果确认模块， 需要人工对所述判断输出模块输出的结果进行 确认。
[0038] 本发明还公幵了一种血型分析仪， 使用上述的微柱凝胶卡凝集检测结果识别系 统。
[0039] 本发明的有益效果： 微柱凝胶卡凝集检测结果识别系统：
[0040] (1) 通过二值化处理单元和拟合描线单元自动拟合出微管内壁边界线， 获取 微管内壁边界线以内的反应区域作为分析的目标区域， 避免了区域以外的图像 部分干扰分析判断， 有效提高了识别的准确性， 同吋使得该系统适用于各种不 同规格的微柱凝胶卡的凝集检测结果识别， 提高了该系统在不同规格的分析仪 中的通用性。
[0041] (2) 对分析的目标区域内有效反应物区域进行亮度位置和相对面积相结合的 分类判断， 再进行粗分类和细分类的对照分析， 进一步提高了结果识别的准确 性。
发明的有益效果
对附图的简要说明
附图说明
[0042] 下面结合附图和实施例对本发明进一步说明。
[0043] 图 1是本发明微柱凝胶卡凝集检测结果识别系统实施例示意图；
[0044] 图 2是本发明微柱凝胶卡凝集检测结果识别系统另一实施例示意图；
[0045] 图 3为预设的判断标准的类别与相应结果的微管图像示意图；
[0046] 图 4为微管图像截取模块在软件窗口上的运行示意图；
[0047] 图 5为截取的微管图像；
[0048] 图 6为图 5中截取的微管图像的二值化图；
[0049] 图 7为目标区域图像；
[0050] 图 8为在图 7目标区域图像中识别出的明亮区域示意图。
[0051] 附图标记： 1、 截取框； 2、 截取框控制单元； 3、 微管。
[0052] A、 明亮区域； B、 次明亮区域。
具体实施方式
[0053] 现在结合附图对本发明作进一步详细的说明。 这些附图均为简化的示意图， 仅 以示意方式说明本发明的基本结构， 因此其仅显示与本发明有关的构成。
[0054] 实施例 1
[0055] 如图 1所示， 为本发明的微柱凝胶卡凝集检测结果识别系统， 包括： 图像采集 模块， 用于采集经过凝集反应并离心完成的微柱凝胶卡图像； 微管图像截取模 块， 用于截取所述微柱凝胶卡图像中的微管图像； 目标区域获取模块， 用于获 取所述微管图像中需要识别的目标区域， 所述目标区域为所述微管内壁以内的 图像区域； 有效反应物区域捕获识别模块， 用于在所述目标域内， 捕获有效反 应物区域， 识别所捕获的有效反应物区域的位置和相对面积； 标准设定模块， 用于用户预先设定判断标准； 和判断输出模块， 用于判断和输出所述微管的凝 集检测结果。
[0056] 其中， 微管图像截取模块包括截取框 1和截取框控制单元 2， 如图 4所示为该微 管图像截取模块在软件窗口上的运行示意图， 截取框控制单元 2可以控制调节截 取框 1的长度、 宽度和位置， 以适应不同规格的微柱凝胶卡， 使截取框 1能截取 到微柱凝胶卡各微管 3的完整图像， 确保微管 3内凝胶柱反应区域全部被截取到
[0057] 上述目标区域为剔除了微管内壁外围部分的内壁以内的图像区域， 即上述微管 3内凝胶柱反应区域， 如图 7所示。 目标区域获取模块包括： 图像转换单元、 二 值化处理单元、 拟合描线单元和图像刪除单元。 其中， 图像转换单元用于将彩 色图像转换成灰度图像； 二值化处理单元用于二值化处理灰度图像； 拟合描线 单元用于根据二值化图识别出的边界像素点进行线性拟合或曲线拟合， 并描出 所述微管的内壁边界线； 图像刪除单元， 用于刪除图像。 在该实施例中， 为了 获取上述目标区域， 首先利用图像转换单元将上述截取框截取的微管图像转换 成灰度图像， 并利用二值化处理单元将该灰度图像进行二值化处理， 获得二值 化图像， 如图 6所示。 然后在该二值化图像上识别出属于微管内壁的像素点， 对 这些像素点进行线性拟合或曲线拟合， 如利用线性拟合方法拟合出直线的管身 内壁边界线， 利用曲线拟合方法拟合出曲线的管底内壁边界线， 并描出上述拟 合的微管内壁边界线， 最后再利用图像刪除单元将上述拟合描出的微管内壁边 界线外围的图像刪掉， 获得需要分析的目标区域， 如图 7。 该系统图像灰度转换 、 二值化处理及拟合方法等， 自动识别出微管内壁边界， 并将微管内壁边界外 围不是分析目标的干扰图像刪掉， 获得更全更精确的图像信息， 确保后续分析 判断的准确性， 以获得更加精准的识别结果； 也使得该系统适用于各种不同规 格的微柱凝胶卡的凝集检测结果识别， 提高了该系统在不同规格的分析仪中的 通用性。
[0058] 实施例 2 [0059] 在实施例 1的基础上， 采用有效反应物区域捕获识别模块对目标区域进行识别 。 该有效反应物区域捕获识别模块包括： 明亮区域査找识别单元， 用于査找并 识别出该明亮区域在微管内的位置； 次明亮区域査找识别单元， 用于査找并识 别出该次明亮区域在微管内的位置及相对面积。 上述明亮区域为目标区域内灰 度值大于或等于第一阈值的图像区域， 即微管内红细胞集中的区域， 如图 3、 7 和 8中的区域八。 根据经验， 该第一阈值灰度值一般在 108-128范围内取值。 次明 亮区域为目标区域内灰度值大于或等于第二阈值， 并小于第一阈值的图像区域 ， 即微管内红细胞散落的区域， 如图 3中的区域^ 上述第一阈值大于第二阈值 ， 该第二阈值灰度值则在目标区域图像的平均灰度值 ±10的范围内取值。 系统可 自动在上述范围内进行取值调试， 根据每一次识别的图像的实际灰度值分布情 况， 确定适用于该图像的第一阈值和第二阈值的最优值。
[0060] 在目标区域内利用上述有效反应物区域捕获识别模块识别出明亮区域 A和次明 亮区域 B之后， 再利用判断输出模块对照标准设定模块预设的判断标准， 对识别 结果进行判断分析， 并输出识别结果。
[0061] 该实施例中， 标准设定模块预设的判断标准为：
[0062] 明亮区域分布于微管顶部且没有次明亮区域， 则判断为"阳 4";
[0063] 明亮区域分布于微管顶部且有次明亮区域， 则判断为"阳 3";
[0064] 没有明亮区域只有次明亮区域， 则判断为"阳 2";
[0065] 明亮区域分布于微管底部、 有次明亮区域且次明亮区域分布于微管中下部， 则 判断为 "阳 1";
[0066] 明亮区域分布于微管底部、 有次明亮区域且次明亮区域靠近底部的明亮区域， 则判断为"弱凝集"；
[0067] 明亮区域分布于微管的顶部和底部， 没有次明亮区域， 则判断为"混合凝集"；
[0068] 明亮区域分布于微管的全部区域， 则判断为"溶血"；
[0069] 明亮区域分布于微管底部且没有次明亮区域， 则判断为"阴性"。
[0070] 如下表， 相应结果的微管图像如图 3所示。
[]
Figure imgf000009_0001
[0071] [0072]上述判断输出模块包括： 粗判断单元， 根据明亮区域 A査找识别单元识 别出的明亮区域的位置， 对照上述标准设定模块预设的判断标准， 对结果进行 粗分类判断； 细判断单元， 在上述粗分类的基础上， 根据次明亮区域 B査找识别 单元识别出的次明亮区域的位置和相对面积， 对照标准设定模块预设的判断标 准， 对结果进行细分类判断， 得出结果； 输出单元， 用于输出细判断单元判断 得出的结果。
[0072] 例如， 利用有效反应物区域捕获识别模块对图 7所示的目标区域进行识别， 明 亮区域査找识别单元识别出了明亮区域 A位于微管内凝胶柱顶部， 则判断输出模 块的粗判断单元在对照上述预设的判断标准之后， 初步判断其为 "阳 4"或"阳 3" 而次明亮区域査找识别单元在该目标区域内没有査找到次明亮区域， 则细判断 单元便认为该目标区域内的明亮区域位于微管顶部， 且没有明亮区域， 其反应 结果为 "阳 4"， 并利用输出单元输出该结果。
[0073] 如果识别到明亮区域 A分布于微管底部， 则粗判断单元在对照上述判断标准之 后， 初步判断其为 "阳 1"、 "弱凝集"或"阴性"。 然后再根据识别出的次明亮区域 B 的位置和相对面积大小， 对照判断其具体属于哪一种反应结果。 如果没有査找 到次明亮区域 B， 则细判断单元便认为该目标区域内 "明亮区域位于微管底部， 且没有次明亮区域"， 其反应结果为 "阴性"； 如果有査找到次明亮区域 B， 且识 别出该次明亮区域 B相对面积较大， 分布于微管中下部， 则细判断单元便认为该 目标区域内"明亮区域位于微管底部、 有次明亮区域且次明亮区域分布于微管中 下部"， 其反应结果为 "阳 1"; 如果识别出次明亮区域 B相对面积较小， 且靠近底 部的明亮区域 A， 则细判断单元便认为该目标区域内 "明亮区域分布于微管底部 、 有次明亮区域且次明亮区域靠近底部的明亮区域"， 其反应结果为 "弱凝集"。
[0074] 上述输出单元输出的结果还可以包括输出图像采集模块采集的原始图像。
[0075] 通过明亮区域的粗分类判断和次明亮区域的细分类判断， 这样分两步进行的识 别判断， 进一步提高了结果识别的准确性。
[0076] 实施例 3
[0077] 如图 3所示的实施例 3， 在实施例 1或 2的基础上增加了结果确认模块， 需要人工 对上述判断输出模块输出的结果进行确认。 例如， 在判断输出模块判断出结果 之后， 以弹出窗口的形式或其他形式输出该判断结果， 并要求人工进行结果确 认， 确认之后系统才结束本次识别工作。 经过人工确认， 避免了因仪器故障或 其他原因导致的系统误判的情况发生。
[0078] 实施例 4
[0079] 本发明还公幵了一种血型分析仪， 采用上述的微柱凝胶卡凝集检测结果识别系 统， 有效提高了血型分析的准确性。
[0080] 以上述依据本发明的理想实施例为启示， 通过上述的说明内容， 相关工作人员 完全可以在不偏离本项发明技术思想的范围内， 进行多样的变更以及修改。 本 项发明的技术性范围并不局限于说明书上的内容， 必须要根据权利要求范围来 确定其技术性范围

Claims

权利要求书
微柱凝胶卡凝集检测结果识别系统， 其特征在于， 包括：
图像采集模块， 用于采集经过凝集反应并离心完成的微柱凝胶卡图像 微管图像截取模块， 用于截取所述微柱凝胶卡图像中的微管图像； 目标区域获取模块， 用于获取所述微管图像中需要识别的目标区域， 所述目标区域为所述微管内壁以内的图像区域；
有效反应物区域捕获识别模块， 用于在所述目标域内， 捕获有效反应 物区域， 识别所捕获的有效反应物区域的位置和相对面积； 标准设定模块， 用于用户预先设定判断标准；
判断输出模块， 用于判断和输出所述微管的凝集检测结果。
根据权利要求 1所述的微柱凝胶卡凝集检测结果识别系统， 其特征在 于： 所述微管图像截取模块包括截取框和截取框控制单元； 所述截取 框控制单元用于控制所述截取框的长度、 宽度和位置； 所述截取框用 于截取图像。
根据权利要求 1所述的微柱凝胶卡凝集检测结果识别系统， 其特征在 于： 所述目标区域获取模块包括：
图像转换单元， 用于将彩色图像转换成灰度图像；
二值化处理单元， 用于二值化处理灰度图像；
拟合描线单元， 用于根据二值化图识别出的边界像素点进行线性拟合 或曲线拟合， 并描出所述微管的内壁边界线；
图像刪除单元， 用于刪除图像。
根据权利要求 1所述的微柱凝胶卡凝集检测结果识别系统， 其特征在 于： 所述有效反应物区域捕获识别模块包括：
明亮区域査找识别单元， 用于査找并识别出该明亮区域在所述微管内 的位置；
次明亮区域査找识别单元， 用于査找并识别出该次明亮区域在所述微 管内的位置及相对面积； 所述明亮区域为所述目标区域内灰度值大于或等于第一阈值的图像区 域； 所述次明亮区域为所述目标区域内灰度值大于或等于第二阈值， 并小于第一阈值的图像区域； 所述第一阈值大于所述第二阈值。
[权利要求 5] 根据权利要求 4所述的微柱凝胶卡凝集检测结果识别方法， 其特征在 于： 所述第一阈值灰度值在 108-128范围内； 所述第二阈值灰度值在 所述目标区域图像的平均灰度值 ±10的范围内。
[权利要求 6] 根据权利要求 4所述的微柱凝胶卡凝集检测结果识别系统， 其特征在 于： 所述标准设定模块预设的判断标准为：
明亮区域分布于微管顶部且没有次明亮区域， 则判断为"阳 4";
明亮区域分布于微管顶部且有次明亮区域， 则判断为"阳 3";
没有明亮区域只有次明亮区域， 则判断为"阳 2";
明亮区域分布于微管底部、 有次明亮区域且次明亮区域分布于微管中 下部， 则判断为"阳 1";
明亮区域分布于微管底部、 有次明亮区域且次明亮区域靠近底部的明 亮区域， 则判断为"弱凝集"；
明亮区域分布于微管的顶部和底部， 没有次明亮区域， 则判断为"混 合凝集"；
明亮区域分布于微管的全部区域， 则判断为"溶血"；
明亮区域分布于微管底部且没有次明亮区域， 则判断为"阴性"。
[权利要求 7] 根据权利要求 4所述的微柱凝胶卡凝集检测结果识别系统， 其特征在 于： 所述判断输出模块包括：
粗判断单元， 根据所述明亮区域査找识别单元识别出的明亮区域的位 置， 对照所述标准设定模块预设的判断标准， 对所述结果进行粗分类 判断；
细判断单元， 在所述粗分类的基础上， 根据所述次明亮区域査找识别 单元识别出的次明亮区域的位置和相对面积， 对照所述标准设定模块 预设的判断标准， 对所述结果进行细分类判断， 得出结果； 输出单元， 用于输出所述细判断单元判断得出的结果。 [权利要求 8] 根据权利要求 7所述的微柱凝胶卡凝集检测结果识别系统， 其特征在 于：
所述输出单元输出的结果还包括所述图像采集模块采集的原始图像。
[权利要求 9] 根据权利要求 1-8任一项所述的微柱凝胶卡凝集检测结果识别系统， 其特征在于： 还包括结果确认模块， 需要人工对所述判断输出模块输 出的结果进行确认。
[权利要求 10] —种血型分析仪， 其特征在于： 使用如权利要求 1-9任一项所述的微 柱凝胶卡凝集检测结果识别系统。
PCT/CN2017/107930 2017-06-28 2017-10-27 微柱凝胶卡凝集检测结果识别系统及血型分析仪 WO2019000740A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710507496.5 2017-06-28
CN201710507496.5A CN107389957B (zh) 2017-06-28 2017-06-28 微柱凝胶卡凝集检测结果识别系统及血型分析仪

Publications (1)

Publication Number Publication Date
WO2019000740A1 true WO2019000740A1 (zh) 2019-01-03

Family

ID=60334126

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/107930 WO2019000740A1 (zh) 2017-06-28 2017-10-27 微柱凝胶卡凝集检测结果识别系统及血型分析仪

Country Status (2)

Country Link
CN (1) CN107389957B (zh)
WO (1) WO2019000740A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110530867A (zh) * 2019-10-11 2019-12-03 江苏贝索生物工程有限公司 卡式微试管反应杯及利用该反应杯的全自动试管检测法
CN110780080B (zh) * 2019-11-08 2024-07-12 安邦（厦门）生物科技有限公司 一种血型分析仪及血型分析方法
CN112485457B (zh) * 2020-11-18 2023-04-18 天津德祥生物技术股份有限公司 一种检测结果识别方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103543277A (zh) * 2013-09-13 2014-01-29 中国科学院苏州生物医学工程技术研究所 一种基于灰度分析与种类识别的血型结果识别算法
CN105678752A (zh) * 2015-12-30 2016-06-15 合肥天一生物技术研究所 一种基于固定参数分割微柱小管的血型识别方法
CN105866444A (zh) * 2016-05-03 2016-08-17 江阴力博医药生物技术有限公司 MNSs血型系统检测试剂卡及其制备方法
CN106199014A (zh) * 2016-06-29 2016-12-07 中山生物工程有限公司 一种基于血型卡的血型判别方法
CN106290921A (zh) * 2016-08-03 2017-01-04 中山生物工程有限公司 一种基于微孔膜的血型检测卡、血型检测系统

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5872860A (en) * 1997-03-13 1999-02-16 Ortho Diagnostic Systems, Inc. Calibration cassette for use in calibrating an automated agglutination reaction analyzing system
CN102680716A (zh) * 2012-06-07 2012-09-19 北京金豪制药股份有限公司 ABO/RhD血型抗原检测试剂卡及其制备
CN103984939B (zh) * 2014-06-03 2017-07-04 爱威科技股份有限公司 一种样本有形成分分类方法及系统
CN105403714A (zh) * 2015-10-21 2016-03-16 中山市生科试剂仪器有限公司 一种判图仪的软件控制方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103543277A (zh) * 2013-09-13 2014-01-29 中国科学院苏州生物医学工程技术研究所 一种基于灰度分析与种类识别的血型结果识别算法
CN105678752A (zh) * 2015-12-30 2016-06-15 合肥天一生物技术研究所 一种基于固定参数分割微柱小管的血型识别方法
CN105866444A (zh) * 2016-05-03 2016-08-17 江阴力博医药生物技术有限公司 MNSs血型系统检测试剂卡及其制备方法
CN106199014A (zh) * 2016-06-29 2016-12-07 中山生物工程有限公司 一种基于血型卡的血型判别方法
CN106290921A (zh) * 2016-08-03 2017-01-04 中山生物工程有限公司 一种基于微孔膜的血型检测卡、血型检测系统

Also Published As

Publication number Publication date
CN107389957A (zh) 2017-11-24
CN107389957B (zh) 2018-10-16

Similar Documents

Publication Publication Date Title
JP6858243B2 (ja) 試料容器の容器キャップを識別するためのシステム、方法及び装置
WO2019000740A1 (zh) 微柱凝胶卡凝集检测结果识别系统及血型分析仪
JP2019504997A (ja) 側方多視点から試料を定量化するように構成された方法及び装置
JP6453564B2 (ja) 検出装置
CN103735253A (zh) 一种基于移动终端的中医舌象分析系统及方法
CN108845153A (zh) 一种粒子分析系统及方法
Sarkar et al. Malaria detection from RBC images using shallow convolutional neural networks
JP6422824B2 (ja) 検体前処理方法及び検体前処理装置
CN114332058A (zh) 基于神经网络的血清质量识别方法、装置、设备及介质
WO2019000741A1 (zh) 微柱凝胶卡凝集检测结果识别方法
US7702172B2 (en) Particle extraction for automatic flow microscope
CN112798678A (zh) 基于血清的新型冠状病毒感染快速检测方法
CN110567900B (zh) 试样反应中抗原过量的判断方法、装置及光学检测系统
CN110243823B (zh) 基于支持向量机的小儿佝偻病自动筛查装置及其使用方法
CN115452796B (zh) 一种基于表面增强拉曼散射成像的免疫层析试条t线的高灵敏识别方法
CN111398138A (zh) 一种干式血液细胞分析装置的光学检测系统及方法
CN1275188C (zh) 光干扰条件下有效指纹信号的检出和残留指痕的甄别方法
JP2020118596A (ja) 生体試料検出装置
US11511279B2 (en) Automatic correction of tube label information
CN116993677A (zh) 一种微柱凝胶卡凝集检测结果识别方法
CN115015189A (zh) 一种荧光分析仪、分析系统及试管识别方法
JP7407282B2 (ja) 管アセンブリタイプを識別する装置および方法
Abhishek et al. An enhanced algorithm for the quantification of human chorionic gonadotropin (hCG) level in commercially available home pregnancy test kits
JPH08210970A (ja) 粒子凝集パターン判定装置
Manasa et al. An automated algorithm for the quantification of hCG level in novel fabric-based home pregnancy test kits

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17916249

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17916249

Country of ref document: EP

Kind code of ref document: A1