Review of Segmentation for Medical image analysis

標籤：sem   均值   photo   方差   tom   html   extra   重疊   row   

成像方法：X射線，CT,MRI,SPECT,PET等分割的定義： Image segmentation is a procedure for extracting the region of interest (ROI) throughan automatic or semi-automatic process【1】. 應用： border detection in angiograms of coronary冠狀動脈血管造影, surgical planning, simulation of surgeries, tumor detection and segmentation腫瘤檢測與分割, brain development study, functional mapping, blood cells automated classification, mass detection in mammograms, image registration, heart segmentation and analysis of cardiac images。

分割方法（4類）：

1）region-based methods, Here we explain two most popular regionbased approaches: thresholding and region growing。1,1）閾值法缺點：沒有考慮映像的空間資訊，導致雜訊敏感
局部閾值法（基於局部的均值方差資訊）和Otsu閾值化（找最優全域閾值，極小化類內方差）1.2）地區生長法，一種互動分割方法，會產生hole或不連通地區2）clustering methods, 2.1）K-means2.2）Fuzzy c-means2.3）EM演算法 3）classifier methods（模式識別）, k近鄰（KNN,非參數）和極大似然（參數），缺點沒有利用空間資訊，訓練資料需要人工分割。  4）hybrid methods.4.1）Gruph cut4.2） 結果評價：Dice Similarity Index（DSI）度量自動與人工分割的重疊程度。 實驗資料：   參考文獻：  【1】Norouzi, A., Rahim, M.S.M., Altameem, A., Saba, T., Rad, A.E., Rehman, A., Uddin, M., 2014. Medical Image Segmentation Methods, Algorithms, and Applications. IETE Technical Review 31, 199-213.

Review of Segmentation for Medical image analysis