Abstract
Feature extraction is a crucial step in most cytometry studies. In this paper a systematic approach to feature extraction is presented. The feature sets that have been developed and used for quantitative cytology at the Laboratory for Biomedical Image Analysis of the GSF as well as at the Center for Image Analysis in Uppsala over the last 25 years are described and illustrated. The feature sets described are divided into morphometric, densitometric, textural and structural features. The latter group is used to describe the eu‐ and hetero‐chromatin in a way complementing the textural methods. The main goal of the paper is to bring attention to the need of a common and well defined description of features used in cyto‐ and histometrical studies. The application of the sets of features is shown in an overview of projects from different fields. Finally some rules of thumb for the design of studies in this field are proposed. Colour figures can be viewed on http://www.esacp.org/acp/2003/25‐1/rodenacker.htm.
Highlights
The visual interpretation of the appearance of cells and tissue parts through a light microscope is at the heart of medical diagnosis as carried out in cytology and histopathology
The main objective when designing the feature extraction approach for a particular study in cytometry or histometry is usually to find a set of features that can discriminate between the different relevant populations of cells/specimens/cases/patients as well as possible
As outlined in the previous sections we propose size and shape, intensity, texture and structure as the main categories of features, a grouping which is based on whether only the spatial, only the densitometric or the combined spatial and densitometric distributions are assessed
Summary
The visual interpretation of the appearance of cells and tissue parts through a light microscope is at the heart of medical diagnosis as carried out in cytology and histopathology. Even worse some times black box machines are used for the measurements where the implemented algorithms are not even known by the author This lack of commonly accepted standards for how different features should be defined and studies documented in such a way that different results can be readily reproduced and compared is probably one of the main reasons why quantitative image analysis based methods have not yet fully penetrated mainline pathology practices. This paper is written in recognition of the problem outlined in the previous paragraphs and can be seen as an attempt to contribute to the method oriented discussion about feature extraction in cytometry It is based on more than 25 years of experience at each of the authors laboratory. Additional information and illustrations can be found under the web-pages of the authors http://www.gsf.de/ILIAD/
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
