Chen X, Doi K, Katsuragawa S, MacMahon H
Kurt Rossmann Laboratories for Radiologic Image Research, Department of Radiology, University of Chicago, Illinois 60637.
Med Phys. 1993 Jul-Aug;20(4):975-82. doi: 10.1118/1.596979.
In order to implement a computerized scheme for quantitative analysis of interstitial lung disease in chest radiographs in clinical situations, a fully automated method of selecting many square regions of interest (ROIs) in peripheral lung areas are developed. First, the peripheral lung regions are identified, based on the automated detection of lung apices, ribcage edges, and diaphragm. Then a large number of ROIs are selected sequentially by filling in the identified peripheral regions. Finally, those ROIs containing sharp edges are removed based on an edge gradient analysis, for which a gradient-weighted edge orientation histogram is employed. Approximately 200-400 ROIs were automatically selected for each case with this method. The evaluation of using ROC analysis indicated that the automated ROI selection method was effective in quantitative analysis of lung textures in digital chest radiographs.
为了在临床情况下实现胸部X光片中间质性肺病的计算机定量分析方案，开发了一种在肺外周区域自动选择多个方形感兴趣区域（ROI）的全自动方法。首先，基于对肺尖、胸廓边缘和膈肌的自动检测来识别肺外周区域。然后通过填充已识别的外周区域依次选择大量ROI。最后，基于边缘梯度分析去除那些包含锐利边缘的ROI，为此采用了梯度加权边缘方向直方图。使用这种方法，每个病例大约自动选择200 - 400个ROI。使用ROC分析的评估表明，自动ROI选择方法在数字胸部X光片的肺纹理定量分析中是有效的。
