Towards Automated Bioimage Analysis: From Features to Semantics

Abstract

Recent advances in bio-molecular imaging have afforded biologists a more thorough understanding of cellular functions in complex tissue structures. For example, high resolution fluorescence images of the retina reveal details about tissue restructuring during detachment experiments. Time sequence imagery of microtubules provides insight into subcellular dynamics in response to cancer treatment drugs. However, technological progress is accompanied by a rapid proliferation of image data. Traditional analysis methods, namely manual measurements and qualitative assessments, become time consuming and are often nonreproducible. Computer vision tools can efficiently analyze these vast amounts of data with promising results. This paper provides an overview of several challenges faced in bioimage processing and our recent progress in addressing these issues.

Challenges of medical image processing

“…In [11], methods are described to evaluate polymer transport and turnover in fluorescent speckle microscopy (FSM), based on cross correlation and particle flow. Approaches for tracking micro-tubules can be found in [50], where active contours and a hidden Markov model (HMM) are used [39], or in [14] for a speckle-based technique. A global minimization method by simulated annealing for the tracking of fluorescent structures was developed in [47].…”

An Automated Method for Cell Detection in Zebrafish

“…Therefore, the development and evaluation of computerized algorithms that are specifically tailored for accurate and reproducible quantification of microscopic zebrafish images become an increasingly important necessity. While computerized analysis of bioimaging data has been extensively explored (Price et al 1998;Sjostrom et al 1999;Chen et al 2006;Shain et al 1999;Lin et al 2003;Yang and Parvin 2003;Loukas et al 2003;Byun et al 2006;Manjunath et al 2006;Parvin et al 2007), only a few studies of automated quantification for zebrafish images have been reported. Examples of such studies include the image processing system for scoring the schooling behavior of multiple zebrafish (Kato et al 2004), as well as the deformable model that was designed to segment and quantify the embryonic vasculature of zebrafish embryos from microangiography image series (Feng et al 2004).…”