Stereological morphometric grids for random sections and 3D volume sets in ImageJ

Session

LST.3 - Advances in Image Processing in Biological Electron Microscopy

Authors

Mr. Aleksandr Mironov (1)

Affiliations

1. University of Manchester

Keywords

Image J, morphometry, stereology

Abstract text

The beauty of unbiased stereological procedures for morphometric quantifications lies in the simplicity of their application based on rigid mathematical framework once the major principles were followed in specimen preparation. Many microscope control and scientific image analysing programs have stereology modules, which can be used to display some type of stereological grid over live or recorded image from the microscope. The user will then have to count how many of grid points or grid lines intersect particular types of objects of interest. Most of stereological software is just a part of multifunctional software packages and, therefore, could incur very significant expenses that can interfere with the widespread use of computerized stereological tools.

One of the well-known exceptions is ImageJ software [1], which is a public domain Java image processing program that was designed with an open architecture that provides extensibility via plugins and macros. 

I have developed a number of macros that covers most of the well-known and widely used estimators for total quantities like volume, surface, length or number of features. 

Grid designs for non-local estimators of volume, surface area and length in random sections include: multipurpose grid with lines and points [2]; Merz grid [3] and multicircles grid that have built-in isotropy; cycloid grid for vertical sections design [4] and unbiased counting frame grid [5] with the disector [6] volume feature. 

For 3D volume sets of stacks of sections my macros include: Cavalieri volume estimator [2], spatial grid surface area estimator [7], length estimator with spherical probes [8] and “unbiased brick” number estimator [9].

Macros for local volume estimators include the nucleator [10] and the rotator [11].

All of the grids have options for random offset, orientation and various densities and include help instructions. The macros display specific grid parameters (such as points/lines per area etc.) in separate window to facilitate further calculations. In combination with Multipoint Tool these grids can be effectively used to count specimen features with stereological tools. Macro format allows to examine and to change the code without previous knowledge of programming language (such as Java). As an additional convenience, these macros are organized in a “tool set” that can be added to the ImageJ toolbar as clickable icons.

All macros and the toolset are licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported License and are freely available at https://github.com/amironov21/ImageJ_stereology

References

[1] Abramoff M.D. et al (2004) Biophotonics International, 11 : p.36.

[2] Gundersen H.J., Jensen E.B. (1987) J. Microsc. 147: p.229.

[3] Howard C.V., Reed M.G. (2005) in “Unbiased Stereology”, ed. C.V. Howard, M.G. Reed,

(Garland Science/BIOS Scientific Publishers, Oxon, 2nd ed.) p.187.

[4] Baddeley A.J. et al (1986) J. Microsc. 142: p.259.

[5] Gundersen H.J. (1977) J. Microsc. 111: p.219.

[6] Sterio D.C. (1984) J. Microsc. 134: p.127.

[7] Cruz-Orive L.M. (1997) J. Microsc. 186: p.93.

[8] Mouton P.R. et al , (2002) J. Microsc. 206: p. 54

[9] Howard C.V., Reed M.G. (2005) in “Unbiased Stereology”, ed. C.V. Howard, M.G. Reed,

(Garland Science/BIOS Scientific Publishers, Oxon, 2nd ed.) p.78

[10] Gundersen H.J. (1988) J. Microsc 151: p. 3

[11] Jensen E.B., Gundersen H.J. (1993) J.Microsc. 170: p. 35