Evaluating stability of histomorphometric features across scanner and staining variations: prostate cancer diagnosis from whole slide images
Evaluating stability of histomorphometric features across scanner and staining variations: prostate cancer diagnosis from whole slide images
复制标题
DOI：
10.1117/1.jmi.3.4.047502
复制
发表时间：
2016-10-01
影响因子：
2.4
通讯作者：
Madabhushi, Anant
中科院分区：
文献类型：
--
作者：
Leo, Patrick;Lee, George;Madabhushi, Anant
定量组织形态学（QH）是从数字化组织切片图像中提取计算机特征以预测疾病存在、行为和结果的过程。实验室特定变量（包括染料批次、切片厚度和使用的整个载玻片扫描仪）可能会影响研究中心之间的特征稳定性。我们提出了两个新的措施，消除引起的不稳定性得分和潜在的不稳定性得分，量化跨数据集和数据集内的特征不稳定性。在涉及前列腺癌的用例中，我们检查了可以在整个载玻片图像上检测癌症的QH特征。使用我们的方法，我们发现五个特征家族（图形，形状，共现腺体张量，子图和纹理）在19.7%至48.6%的比较中在数据集之间存在差异，而没有站点变化的预期值为4.2%至4.6%。将所有图像颜色标准化为模板并没有减少不稳定性。在三台扫描仪上扫描相同的34张载玻片表明，Haralick特征受扫描仪变化的影响最大，在62%的比较中不稳定。我们发现，不稳定的功能家庭之间比intrasite分类显着较差。我们的研究结果似乎表明，QH功能应该在不同的网站进行评估，以评估鲁棒性，单独的类区分度不应该代表数字病理学特征选择的基准。(C)2016年，美国光电仪器工程师学会（SPIE）
Quantitative histomorphometry (QH) is the process of computerized feature extraction from digitized tissue slide images to predict disease presence, behavior, and outcome. Feature stability between sites may be compromised by laboratory-specific variables including dye batch, slice thickness, and the whole slide scanner used. We present two new measures, preparation-induced instability score and latent instability score, to quantify feature instability across and within datasets. In a use case involving prostate cancer, we examined QH features which may detect cancer on whole slide images. Using our method, we found that five feature families (graph, shape, co-occurring gland tensor, sub-graph, and texture) were different between datasets in 19.7% to 48.6% of comparisons while the values expected without site variation were 4.2% to 4.6%. Color normalizing all images to a template did not reduce instability. Scanning the same 34 slides on three scanners demonstrated that Haralick features were most substantively affected by scanner variation, being unstable in 62% of comparisons. We found that unstable feature families performed significantly worse in inter-than intrasite classification. Our results appear to suggest QH features should be evaluated across sites to assess robustness, and class discriminability alone should not represent the benchmark for digital pathology feature selection. (C) 2016 Society of Photo-Optical Instrumentation Engineers (SPIE)