Computerized detection of clustered microcalcifications: evaluation of performance on mammograms from multiple centers.
Computerized detection of clustered microcalcifications: evaluation of performance on mammograms from multiple centers.
复制标题
计算机化检测簇状微钙化：评估多中心乳房 X 光检查的性能。
DOI：
10.1148/radiographics.15.2.7761647
复制
发表时间：
1995
期刊：
影响因子：
--
通讯作者：
Lu,P
中科院分区：
文献类型：
--
作者：
Nishikawa,RM;Doi,K;Giger,ML;Schmidt,RA;Vyborny,CJ;Monnier-Cholley,L;Papaioannou,J;Lu,P
为了研究在不同筛查中心的数字化乳房x线照片中自动检测聚集性微钙化的计算机化方法的性能，作者邀请了118名放射科医生在1993年北美放射学会（RSNA）会议上的科学展览上带来多达5张乳房x线照片。来自14个地点的43张乳房x光片被带到展览现场，在那里它们被数字化和分析。将RSNA病例的分析结果与从两个中心收集的39张乳房x线照片的标准数据库中获得的结果进行比较。该算法在RSNA图像上的检测性能低于在标准数据库上的检测性能。这种较低的性能部分是由于RSNA病例中非常细微的聚集性微钙化的较高比例，以及算法对图像特征（例如，对比度和噪声）的明显依赖，这些特征在中心之间是不同的。作者得出结论，该算法足够稳健和准确，可以进行临床测试。在临床实施时，计算机化方案必须根据每个特定筛查中心的图像特征进行定制，以获得最佳性能。
To investigate the performance of a computerized method for the automated detection of clustered microcalcifications in digitized mammograms from a variety of screening centers, the authors invited 118 radiologists to bring up to five mammograms to their scientific exhibit at the 1993 meeting of the Radiological Society of North America (RSNA). Forty-three mammograms from 14 sites were brought to the exhibit, where they were digitized and analyzed. Results of the analysis on the RSNA cases were compared with those obtained on a standard database of 39 mammograms collected from two centers. The performance of the detection algorithm on the RSNA images was lower than that achieved on the standard database. This lower performance was due in part to the higher fraction of very subtle clustered microcalcifications in the RSNA cases, as well as the apparent dependence of the algorithm on image characteristics (eg, contrast and noise), which varied from center to center. The authors conclude that the algorithm is robust and accurate enough to undergo clinical testing. When it is implemented clinically, the computerized scheme must be customized to the image characteristics at each specific screening center to obtain optimal performance.