【信息技术】【2010.06】【含源码】图像处理在医学诊断中的应用

本文为美国俄勒冈大学（作者：KRISTINE A. THOMAS）的硕士论文，共67页。

图像处理是提高疾病诊断可靠性和可重复性的有力工具。在病理学家的手中，图像处理提供来自组织学图像的定量数据，这些数据补充了专家们目前使用的定性数据。本论文提出一种分析苏木精与曙红(H&E)染色组织学切片的数字化影像的新方法，用以检测及量化炎性多形核白细胞，以协助评估胎盘的急性炎症程度，从而作为影像处理在辅助治疗中的应用。

本文提出了一种新的阈值分割和形状分析方法。最重要的贡献是H&E染色组织学切片的自动颜色阈值选择算法，这是迄今为止公开的唯一的无监督方法。

Image processing is a powerful tool forincreasing the reliability and reproducibility of disease diagnostics. In thehands of pathologists, image processing provides quantitative data fromhistological images which supplement the qualitative data currently used byspecialists. This thesis presents a novel method for analyzing digitized imagesof hematoxylin and eosin (H&E) stained histology slides to detect andquantify inflammatory polymorphonuclear leukocytes to aid in the grading ofacute inflammation of the placenta as an example of the use of image processingin aid of diagnostics. Methods presented in this thesis include segmentation, anovel threshold selection technique and shape analysis. The most significantcontribution is the automated color threshold selection algorithm for H&E stained histology slides which is the only unsupervised method published todate.

原文及完整源码下载地址：
http://page2.dfpan.com/fs/0lc5j2021b29216efd7/

更多精彩文章请关注微信号：