Please use this identifier to cite or link to this item:
標題: Indirect Immunofluorescence影像細胞自動切割
Automatic Indirect Immunofluorescence Image Cell Segmentation
作者: 詹永寬
關鍵字: 資訊科學軟體;技術發展
以HEp-2細胞螢光染色之間接免疫螢光法(Indirect immunofluorescence, IIF),常被用來觀察抗核抗體(Anti-nuclear antibody)之變異,以診斷自我免疫疾病。抗核抗體(Anti-nuclear antibody)檢驗對抗體實體可進行大範圍的檢測,並以所引發螢光型態的不同,來診斷疾病的有用標記。因此發展一套自動檢測系統,針對IIF image裏之細胞中的螢光型態進行辨識,協助醫檢人員以便能正確地診斷出疾病來,是有其必要性的,且此自動檢測系統並不受限於須具有醫檢專業相關背景之人員才能使用。自動檢測系統首先從對ANA檢測的IIF image中切割出細胞的區域,並辨識出細胞區域中的螢光型態。故如何從IIF image中切割出細胞的區域,是一個值得研究之重要課題。因此本計劃將提出一自動細胞切割系統,以便能有效地從IIF image中切割出細胞來。本計劃提出color selector來將一張彩色IIF image轉成灰階影像;run length enhancer以去除細胞周圍的螢光暈圈;adaptive filter來將細胞內部紋路與坑洞抹平;gradient computing method有效的計算出影將中邊線的梯度。這些技術也能被應用到其他類型的影像切割上。本計劃也利用watershed and distance transform techniques,來對重疊物件進行分割。依螢光分布狀況,IIF images大略可被分成六種。因不同種類影像之特性差距滿大,故本計劃也提議一rough classifier,來對IIF image進行分類,並對不同類別IIF image,採用不同的叁數值來對其進行細胞切割。

Indirect immunofluorescence (IIF) with HEp-2 cells has been frequently used for the detection of antinuclear autoantibodies (ANA) in systemic autoimmune diseases. The ANA testing allows to scan a broad range of autoantibody entities and to describe them by distinct fluorescence patterns. Automatic inspection for fluorescence patterns in IIF image may assist physicians, without relevant experience, in making correct diagnosis. The Automatic inspection system first severs the cells from an IIF image and then recognize what fluorescence pattern are on the cells. How to segment the cells from an IIF image is essential in developing an automatic inspection system for ANA testing. This proposal will present an efficient method for automatically detecting and locating the cells with fluorescence pattern in an IIF image.This proposal proposes a color selector to transform a color IIF image into a gray-level image, a run length enhancer to suppress the hazy fluorescent halo surrounding the cells, an adaptive filter to smooth the surfaces and to fill in the holes on the cells, gradient computing method effectively to compute the pixel gradients of an image. These proposed techniques also can be applied to segment the objects from other kinds of images. Furthermore, watershed and distance transform techniques are used to split overlapping objects. Since the properties of the six patterns of IIF images are extremely different, the proposed segmentation method adopts different parameter values in segmenting the cells from distinct IIF images. A rough classifier is proposed to decide which set of the parameter values should be employed to segment the objects from distinct IIF images.
其他識別: NSC99-2221-E005-022
Appears in Collections:資訊管理學系

Show full item record

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.