Gourt Home::Gourt :: Stereology
Related Topics:Stereology
Stereology is a spatial version of sampling theory. The word "stereology" was coined in May 1961 by the founding fathers of the International Society for Stereology to describe the set of methods that allow a 3-D interpretation of structures based on observations made on 2-D sections. Stereology is therefore often referred to as the science of estimating higher dimensional information from lower dimensional samples. For example, measurements of the area fraction of a certain grain type made on a planar section cut through an opaque petrological sample will reveal the actual 3-D grain volume fraction.
Classically, stereological techniques involved mathematical or computer modeling of the structures under investigation and applying statistical procedures. These methods (which usually assume randomness of the structure under investigation) are now referred to as "model based stereology". These methods are especially popular in materials science, metallurgy and petrology where shapes of e.g. crystals may be modelled.
Since the early 1970s a new branch of stereology have evolved, where the needed randomness is obtained by random spatial sampling. These methods are referred to as "design based stereology" and provide unbiased estimates depending on very few and weak assumptions about the investigated structure. These methods have gained increasing popularity in the biomedical sciences—especially in lung-, kidney-, bone-, cancer- and neuro-science—as these methods do not depend on an exact geometric model of the investigated structure. Many of these applications are directed toward determining the number of elements in a particular structure. The application of stereology is based upon determining the density of the object within the space of interest. The product of the density and the volume of the space within which the objects lie provides an estimate of object number. An experimental concern is that the distribution of objects within the space of interest may not be uniform. Thus, the sample, although random, may not yield a result that is representative of the true density.
An ongoing field of stereological research is the prediction of estimator variance—i.e. the precision of the unbiased stereological estimates. In general, the newest stereological methods based upon systematic, uniformly random sampling of the region of interest are very efficient and need remarkably smaller sample size than methods based upon simple random sampling. The concern in biological systems, where the distribution of objects may not be uniform, is that the density of objects within the regions sampled may not be representative of the density within the volume as a whole.
The primary scientific journals for stereology are "Journal of Microscopy" and "Image Analysis & Stereology" (ex Acta Stereologica).
References
External links
- International Society for Stereology
- Journal of Microscopy
- Image Analysis & Stereology
- Stereology Information Center
Commercial Products
"Stereology"