201100797 六、發明說明： 【發明所屬之技術領域】 本發明是有關於一種用於微粒定量方法，特別是指一 種用於微流體晶片之微粒定量方法。 【先前技術】 我們的生活週遭存在著數目龐大且複雜的環境微生物（ 如細菌、黴菌等），其中許多具有嚴重影響人類健康之特殊 〇 致病性或衛生性。獲得傳統微生物檢測結果報告通常需耗 時數天，因此對於檢驗結果通常無法獲得即時性資訊。 近年來’隨著微流體晶片技術的不斷突破，具分選、 • 捕捉濃縮等功能之微流體晶片亦陸續被開發出來，目前這 類微流體晶片使用時，雖然具有捕捉濃縮之功能，但是對 於所捕捉縮之微粒（生物微粒或非生物微粒）的定量，則 通常無法立即在同一微流體晶片上完成，有些研究是將分 選並濃縮之微粒置於另一計數晶片中，在計數晶片之微流 〇 道内以電場或流場控制的方式，將粒子聚焦成排列成一束 地通過其檢測區域，然後利用檢測區域之阻抗變化或螢光 的強度變化來計數微粒。此方法雖可達到計數之目的，但 因流體速度無法太快，以至於其定量時間相當長，此外， 以阻抗與螢光的差異來檢測，需克服環境的背景訊號，才 能得到具代表性的結果。 【發明内容】 目此，本發明之目的，即在提供-種可快速對微流體 晶片收集之微粒進行定量的方法。 3 201100797 於疋，本發明用於微流體晶片之微粒定量方法，微流 體晶片設有-微流道，且微粒在連續流體帶動下於微流 道中之預定區域聚集堆疊成微粒團簇，該微粒定量方法包 含以下步驟：(a)俯視或仰視擷取微流體晶片中之微粒困簇 影像；（b)分析微粒團簇影像各單位畫素之透光性，而計 算出微粒團簇各單位畫素的堆疊高度，並由各單位畫素之 堆叠同度與單位畫素面積大小估算出微粒團鎮之總體積； 及（Ο由微粒團I總體積、單—微粒體積、流體流速與微 粒聚集時間’定量出微粒濃度。 本發月之力效.透過操取已被集中之微粒團簇影像， 且依據微㈣影像各單位畫素之透光性估算各單位畫素之 微粒堆疊高度的影像分析方法，可於微流體晶片進行微粒 分選捕捉過程中’即時進行微粒之定量，可大幅縮短微粒 定量所需的時間，且可應用於高流速之微流體晶片。 【實施方式】 、有關本發明之前述及其他技術内容、特點與功效，在 m參考圖式之一個較佳實施例的詳細說明中，將可 清楚的呈現。 、 个贫明用於微流體〜彳双视疋置 法的較佳實施例’可用以定量出微流體晶片3所聚集濃彳 之微粒4 ;農度’在本實施财，該微流體晶片3為三維 電泳微流體晶片，作音价。士 仁實施時不以此為限，該微粒4可以, 生物微粒’例如破、&必4、a 細胞或微生物，或非生物微粒，f 如乳膠微粒。 201100797 该微流體晶另3具有一微流道31，及分別設置於微流 道31中之分選電極32與捕捉電極33，該微流道31具有— 分選段311與多數連通於分選段311末端之捕捉段312，該 等分選電極32是沿分選段311長度方向間隔排列地上下對 稱設置於分選段311上、下側，該等捕捉電極33是分別沿 該專捕捉段3 12長度方向間隔排列地上下對稱設置於該等 捕捉段3 12上、下側，且該等上下對稱之分選電極32與捕 C) 捉電極33可分別被通以預定頻率之交流電，而分別於分選 段311與捕捉段312中產生吸引或排斥具預定介電特性之微 粒4的介電泳力❶ 如圖1〜4所示，該微粒定量方法包含以下步驟： • 步驟（一）捕捉聚集微粒4 ^將含有預定微粒4之檢體 混合於預定體積之介電泳液（圖未示）中，並以預定之流 速將含有微粒4之介電泳液注入分選段311中，以連續介^ 泳液持續將微粒4輸送通過該等分選電極32，使該等微粒 〇 4在料分選電極32之介電泳力的作用下，依據其介電特 性，隨著介電泳液的流動被分離至預定之捕捉段312中， 並分別被捕捉段312中之捕捉電極33的介電泳力捕捉撞住 丄且在持續流經之介電泳液”動下，於料捕捉電極Μ 前方開始聚集堆疊成微粒團簇4〇。 步驟（二）擷取微粒團簇40影像。於微粒4開始聚集 —預定時間後，以影像擷取裝置（圖未示）俯視操取該捕 捉312位於該等捕捉電極33前方區域的微粒團鎮仙影 5 201100797 步驟（二）建立二維灰階柱狀分布圖。以三維影像分 析方式，依據微粒團簇影像各單位畫素之透光性，將各單 位畫素之影像轉換成灰階值的三維柱狀圖（如圖3所示）， 理論上，影像的顏色愈深其灰階值則愈小，而影像顏色愈 淺其灰階值則愈大，在微流道中，微粒團簇堆疊的高度愈 高，則透光性愈差，故其影像顏色愈深，並將影像顏色最 深時的高度設定等同於微流道的高度（/〇。接著，將上述所 得二維柱狀分布圖影像的灰階值反轉（如圖4所示），使得 微粒4堆疊的高度與其灰階值成正比，亦即影像顏色越深 表示微粒4堆疊高度越高。 步驟（四）估算微粒4濃度。依據步驟（三）反轉後 之三維柱狀分布圖的高度與每單位畫素之面積，由下式（1 )估算出二維柱狀分布圖之總體積，亦即微粒團簇4〇之總 體積，其中’ 為被捕捉電極捕捉之微粒團簇4〇的總體 積，X、7為單位畫素之長、寬尺寸，A微微流道之高度， Zmax為單位畫素之最大灰階值，Li為每單位畫素之灰階值。 V^ = {XxY)x^{h!L^)L· (1) 於计算出微粒團簇40之總體積後，便可再以下式（2 )什算出微粒初始濃度，其中，c為微粒的濃度，^為每 個微粒4體積’ 為流體之流量，r為微粒4()被捕捉的時 間’ /β為檢體樣本被稀釋的倍率。 c = [(Vtrap /Vp)/(FR X Γ)]χ β ( 2 ) 如圖5〜1〇所示，以下是以乳膠微粒之定量為例說明本 發明微粒40定量方法之測試結果。 201100797 在以下測試例中’流體速度分別為0.75，1_〇〇和1.25 μιη/s仙量刀別為0.6，0.8和1.0 μΐ/ηΰη ,施加於該等捕捉 電極33之電壓為2〇 Vpp，頻率為1〇 MHz，介電泳液為去 離子水。將配製濃度為1·7*107 particles/ml，粒徑為2 μιη 2乳膠微粒導引至微流道之捕捉段312後，捕捉擋止於該 等箭頭狀之捕捉電極33前方，每隔二分鐘擷取其乳膠微粒 堆積構成微粒團簇40的影像，觀察其變化，分別如圖5、7 〇 9所不，並將前述三個微粒團簇影像經三維影像灰階值轉 換成一維灰)¾柱狀分布圖，如圖68、1 〇所示經換算之 後可得到其微粒的總體積與濃度。 如圖11、12所示，在不同的流體速度條件下，被捕捉 f政粒的數目是呈線性的成長，並且其微粒的數量與流速是 成正比。濃度隨時間變化的曲線，在乳膠微粒堆積初期， 因為乳膠微粒在微流道尚未達到飽和的狀態，所以計算的 、结果會比原始的漢度較高。但在5分鐘之後，充填密度已 接近飽和，因此濃度就也趨於穩定，經估算後，微粒濃度 約為2.0*107 particles/mi，與原始的配置濃度丨mo7 particles/ml十分接近，準確率相當高。 在本實施例中，是以仰視方式操取微流體晶片3令之 微粒團簇40影像進行分析處理與估算微粒4濃度，但實施 時，亦可改採由上往下的方式，俯視擷取微流體晶片3中 t微粒團鎮40影像進行分析與估算，同樣可達到本發明之 目的。 綜上所述，透過操取已被集中之微粒團箱40影像進行 7 201100797 三維影像分析，且依據微團簇影像各單位晝素之透光性建 立三維灰值柱狀圖，使得微粒團簇影像之各單位畫素的灰 階值與微粒4堆疊高度成正比的影像分析方法，可於微流 體晶片3進行微粒4分選捕捉過程中，即時進行微粒4之 定量，定量準確度高，且可大幅縮短微粒定量所需的時間 ，並可應用於低流速或高流速之微流體晶片3中，此定量 方法對於食品或飲用水中的細菌量化具有相當不錯的發展 潛力。因此’確實可達到本發明之目的。 惟以上所述者，僅為本發明之較佳實施例而已，當不 能以此限定本發明實施之範圍，即大凡依本發明申請專利 範圍及發明說明内容所作之簡單的等效變化與修飾，皆仍 屬本發明專利涵蓋之範圍内。 【圖式簡單說明】 圖1是本發明用於微流體晶片之微粒定量方法之一較 佳實施例搭配使用之微流體晶片的俯視示意圖； 圖2是該較佳實施例之步驟流程圖； 圖3是該較佳實施例進行微粒定量時所擷取之微粒團 簇的三維灰階柱狀分布圖； 圖4是類似圖3之視圖’說明三維灰階柱狀圖經灰階 值反轉後之視圖，並說明灰階值與微粒堆疊高度成正比的 情況； 圖5是該較佳實施例於微流體晶片開始捕捉聚集微粒 時（1分鐘時）所擷取之微粒團簇影像； 圖ό是圖5微粒團簇影像轉換成之三維灰階值柱狀分 201100797 布圖； 圖7是類似圖5之視圖，說明已捕捉聚集微粒3分鐘 時的微粒團簇影像； 圖8是類似圖6之視圖，是圖7微粒團簇影像轉換成 之二維灰階值柱狀分布圖； 圖9是該類似圖5之視圖，說明已捕捉聚集微粒5分 鐘時的微粒團簇影像； 圖1 〇是類似圖6之視圖，是圖9微粒團簇影像轉換成 之二維灰階值柱狀分布圖； 圖11是該較佳實施例針對微流體晶片於不同時間所捕 捉之乳膠微粒所估算微粒團簇體積的曲線圖；及 圖12是該較佳實施例針對微流體晶片於不同時間所捕 捉之乳膠微粒所估算之微粒濃度的曲線圖。 201100797 【主要元件符號說明】 X ·» N « X ·> ； …微流體晶片 3 2' …· ·分選電極 1 * ¢. Φ * ΤΪ . …微流道 3 3 “… ……捕捉電極 3 11… •…分選段 4 X » V <· K . ……微粒 3 12… …·捕捉段 10201100797 VI. Description of the Invention: [Technical Field] The present invention relates to a method for quantifying particles, and more particularly to a method for quantifying particles for microfluidic wafers. [Prior Art] There are a large number of complex environmental microorganisms (such as bacteria, molds, etc.) around our lives, many of which have special pathogenic or hygienic properties that seriously affect human health. It usually takes several days to obtain reports of traditional microbiological test results, so it is often impossible to obtain real-time information for test results. In recent years, with the continuous breakthrough of microfluidic wafer technology, microfluidic wafers with sorting, capture and concentration functions have been developed. Currently, such microfluidic wafers have the function of capturing and enriching, but for The quantification of captured microparticles (bioparticles or non-biological particles) is usually not immediately possible on the same microfluidic wafer. In some studies, the sorted and concentrated microparticles are placed in another counting wafer. The microfluidic channel is controlled by electric field or flow field to focus the particles in a bundle through the detection area, and then use the impedance change of the detection area or the intensity change of the fluorescence to count the particles. Although this method can achieve the purpose of counting, the fluid velocity cannot be too fast, so that the quantitative time is quite long. In addition, the difference between the impedance and the fluorescence is detected, and the environmental background signal must be overcome to obtain a representative image. result. SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a method for quantifying particles collected by a microfluidic wafer. 3 201100797 In the present invention, the microparticle wafer is provided with a microfluidic channel, and the microfluidic wafer is provided with a microchannel, and the microparticles are aggregated in a predetermined region in the microchannel under a continuous fluid to form a cluster of particles, the microparticles. The quantitative method comprises the steps of: (a) drawing a cluster image of the particles in the microfluidic wafer in a plan view or a bottom view; (b) analyzing the light transmittance of each unit pixel of the particle cluster image, and calculating the unit of the particle cluster The stacking height of the element, and the total volume of the particle group is estimated by the stacking degree of each unit pixel and the unit pixel area; and (the total volume of the particle group I, the single-particle volume, the fluid flow rate and the particle aggregation Time 'quantizes the particle concentration. This month's force effect. Estimate the particle stack height of each unit of pixels by taking the image of the cluster of particles that have been concentrated, and based on the light transmittance of each unit of micro (4) image pixels. Analytical method, which can quantitatively quantify particles during microparticle sorting and capture, can greatly shorten the time required for particle quantification, and can be applied to high flow rate microfluidics [Embodiment] The foregoing and other technical contents, features and effects of the present invention will be apparent from the detailed description of a preferred embodiment of the present invention. A preferred embodiment of the fluid ~ 彳 double viewing method can be used to quantify the concentrated particles 4 concentrated by the microfluidic wafer 3; the agricultural level, the microfluidic wafer 3 is a three-dimensional electrophoretic microfluidic wafer, For the implementation of Shiren, Shiren is not limited to this, the particles 4 can be, biological particles 'such as broken, & must 4, a cells or microorganisms, or non-biological particles, such as latex particles. 201100797 The microfluidic crystal The other 3 has a microchannel 31, and a sorting electrode 32 and a trapping electrode 33 respectively disposed in the microchannel 31. The microchannel 31 has a sorting section 311 and a plurality of capturing sections 312 connected to the end of the sorting section 311. The sorting electrodes 32 are vertically and symmetrically arranged on the upper and lower sides of the sorting section 311 at intervals along the length direction of the sorting section 311. The trapping electrodes 33 are vertically symmetrically arranged along the longitudinal direction of the dedicated capturing section 3 12, respectively. Set to The upper and lower sides of the segment 3 12 are captured, and the upper and lower symmetrical sorting electrodes 32 and the catching electrodes 33 are respectively passed through an alternating current of a predetermined frequency to generate an attraction or a difference in the sorting section 311 and the capturing section 312, respectively. Excluding the dielectrophoretic force of the microparticles 4 having a predetermined dielectric property ❶ As shown in FIGS. 1 to 4, the microparticle quantification method comprises the following steps: • Step (1) capturing the aggregated particles 4 ^ mixing the specimen containing the predetermined microparticles 4 a predetermined volume of dielectrophoresis liquid (not shown) is injected into the sorting section 311 at a predetermined flow rate to inject the microparticles 4 through the sorting electrode 32 in a continuous flow. The particles 〇4 are separated into the predetermined capturing segment 312 by the flow of the dielectrophoretic solution according to the dielectric properties of the microparticles 在4 under the dielectrophoretic force of the material sorting electrode 32, and are respectively captured by the segment 312. The dielectrophoretic force of the trap electrode 33 captures and collides with the dielectrophoresis liquid, and starts to aggregate and form a particle cluster 4〇 in front of the material capture electrode Μ. Step (2) captures the image of the particle cluster 40. After the particles 4 begin to gather for a predetermined period of time, the image capturing device (not shown) is used to plan the capture of the particles 312 located in the area in front of the capturing electrodes 33. Columnar distribution map. According to the three-dimensional image analysis method, according to the light transmittance of each unit pixel of the particle cluster image, the image of each unit pixel is converted into a three-dimensional histogram of gray scale values (as shown in FIG. 3), theoretically, the image is The darker the color, the smaller the grayscale value, and the lighter the image, the higher the grayscale value. In the microchannel, the higher the height of the particle cluster stack, the worse the light transmittance, so the image color is higher. Deep, and set the height of the image at the darkest level to the height of the microchannel (/〇. Then, reverse the grayscale value of the resulting two-dimensional columnar image (as shown in Figure 4), so that the particles 4 The height of the stack is proportional to its grayscale value, that is, the darker the image color, the higher the stacking height of the particles 4. Step (4) Estimate the concentration of the microparticles 4. According to the height of the three-dimensional columnar profile after the reversal of step (3) With the area per unit pixel, the total volume of the two-dimensional columnar distribution map, that is, the total volume of the particle clusters 4〇, is estimated by the following formula (1), where 'the particle clusters captured by the trapped electrode 4〇 The total volume, X, 7 is the length and width of the unit pixel , the height of the A microfluidic channel, Zmax is the maximum grayscale value of the unit pixel, and Li is the grayscale value per unit pixel. V^ = {XxY)x^{h!L^)L· (1) After calculating the total volume of the particle clusters 40, the initial concentration of the particles can be calculated by the following formula (2), where c is the concentration of the particles, ^ is 4 volumes per particle 'flow rate of the fluid, and r is the particle 4 () The captured time ' / β is the magnification at which the sample is diluted. c = [(Vtrap /Vp) / (FR X Γ)] χ β ( 2 ) As shown in Fig. 5 to 1 ,, the following is a test result of the quantitative method of the microparticle 40 of the present invention by taking the quantitative amount of the latex particles as an example. 201100797 In the following test cases, the 'fluid speeds were 0.75, 1_〇〇 and 1.25 μηη/s cents were 0.6, 0.8 and 1.0 μΐ/ηΰη, and the voltage applied to the capture electrodes 33 was 2〇Vpp, The frequency is 1 〇 MHz, and the dielectrophoresis solution is deionized water. After the preparation of the concentration of 1·7*107 particles/ml and the particle size of 2 μη 2 of the latex particles are guided to the capture section 312 of the microchannel, the capture is blocked in front of the arrow-shaped capture electrodes 33, every second Minutes take the image of the latex particles stacked to form the particle cluster 40, observe the change, as shown in Figure 5, 7 〇9, and convert the three particle cluster images into three-dimensional gray value by three-dimensional image. The 3⁄4 columnar profile, as shown in Figure 68, 1 〇, can be converted to the total volume and concentration of the particles. As shown in Figures 11 and 12, at different fluid velocity conditions, the number of captured f-regimes grows linearly, and the number of particles is proportional to the flow rate. The curve of concentration as a function of time, in the initial stage of latex particle accumulation, because the latex particles have not reached saturation in the microchannel, the calculated result is higher than the original Han. However, after 5 minutes, the packing density is close to saturation, so the concentration tends to be stable. After estimation, the particle concentration is about 2.0*107 particles/mi, which is very close to the original concentration 丨mo7 particles/ml. Quite high. In this embodiment, the microfluidic wafer 3 is used to perform the analysis and processing of the microparticles 40 image in a bottom view manner, and the concentration of the microparticles 4 is estimated. However, when implemented, the top-down mode may be adopted, and the image may be taken from the top to the bottom. The analysis and estimation of the t-particle cluster 40 image in the microfluidic wafer 3 can also achieve the object of the present invention. In summary, the 3 201100797 3D image analysis is performed by taking the image of the concentrated particle cluster 40, and the 3D gray value histogram is established according to the light transmittance of each unit of the micro cluster image, so that the particle cluster is made. The image analysis method in which the gray scale value of each unit pixel of the image is proportional to the stack height of the microparticle 4 can accurately quantify the microparticle 4 during the microparticle 4 sorting and capturing process, and the quantitative accuracy is high, and The time required for particle quantification can be drastically shortened and can be applied to microfluidic wafers 3 with low flow rates or high flow rates, which have considerable development potential for quantification of bacteria in food or drinking water. Therefore, the object of the present invention can be achieved. The above is only the preferred embodiment of the present invention, and the scope of the invention is not limited thereto, that is, the simple equivalent changes and modifications made by the scope of the invention and the description of the invention are All remain within the scope of the invention patent. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a top plan view of a microfluidic wafer used in conjunction with a preferred embodiment of the method for quantifying microfluidic wafers of the present invention; FIG. 2 is a flow chart of the steps of the preferred embodiment; 3 is a three-dimensional gray-scale columnar distribution diagram of the particle clusters taken in the microparticle quantification of the preferred embodiment; FIG. 4 is a view similar to FIG. 3, illustrating that the three-dimensional gray-scale histogram is inverted by gray scale values. a view, and illustrates a case where the gray scale value is proportional to the height of the particle stack; FIG. 5 is an image of the particle cluster captured by the microfluidic wafer when the microfluidic wafer starts to capture the aggregated particles (at 1 minute); Figure 5 is a three-dimensional gray-scale value column of 201100797 layout; Figure 7 is a view similar to Figure 5, showing the image of the particle cluster captured when the aggregated particles are captured for 3 minutes; Figure 8 is similar to Figure 6 The view is a two-dimensional gray-scale columnar map converted from the particle cluster image of FIG. 7; FIG. 9 is a view similar to FIG. 5, illustrating the image of the particle cluster when the aggregated particles have been captured for 5 minutes; Is similar to the view of Figure 6, is Figure 9 micro The particle cluster image is converted into a two-dimensional gray scale columnar map; FIG. 11 is a graph of the estimated particle cluster volume of the latex particles captured by the microfluidic wafer at different times in the preferred embodiment; and FIG. It is a graph of the particle concentration estimated for the latex particles captured by the microfluidic wafer at different times in the preferred embodiment. 201100797 [Description of main component symbols] X ·» N « X ·>; ...microfluidic wafer 3 2' ... · sorting electrode 1 * ¢. Φ * ΤΪ . ... microchannel 3 3 "... ...... capture electrode 3 11... •... Sorting section 4 X » V <· K . . . Particle 3 12... ... capture section 10