Segmentation and intensity estimation for microarray images with saturated pixels

Images and gpr files were examined using a novel saturation reduction method to determine whether accuracy could be improved by extending dynamic range of saturated pixels Three immunosignatures from human Valley Fever (Coccidiodes) patients and three immunosignatures from human influenza vaccine recipients were examined to test an algorithm that extends the apparent dynamic range of a fluorescence image. These images had several saturated spots at 70PMT and 100% laser power. The program examined the differences between Valley Fever and influenza in terms of standard image processing vs. segmentation and intensity estimation.

本研究采用一种新型饱和度降低方法对图像与gpr文件进行分析，旨在探究通过拓展饱和像素的动态范围是否能够提升分析精度。我们选取3份人类谷热（Valley Fever，Coccidiodes）患者免疫特征样本与3份人类流感疫苗接种者免疫特征样本，用于测试一款可拓展荧光图像表观动态范围的算法。这批图像在70PMT与100%激光功率参数下存在多处饱和斑点。该程序分别基于标准图像处理、图像分割与强度估计两种方案，对谷热与流感样本的免疫特征差异开展了对比分析。