Estimation of the directional distribution of spatial fibre processes using stereology and confocal scanning laser microscopy

Fibrous structures like polymers, glass fibres, muscle fibres and capillaries are important components of materials and tissues. A spatial fibre process is the union of smoothly curved or linear one‐dimensional features of finite length, arranged in an unbounded three‐dimensional reference space according to some random mechanism. Design‐based stereology was combined with confocal scanning laser microscopy to study samples of fibre‐reinforced composites, which were considered as realizations of not necessarily isotropic fibre processes. The methods enable the unbiased estimation of the intensity and of the directional distribution of spatial fibre processes from arbitrarily directed pairs of registered parallel optical sections a known distance apart. The directions of fibres sampled by a frame of observation on the reference plane are estimated from the coordinates of the intersection points of the fibres with both optical planes using confocal scanning laser microscopy. The true directional distribution of the fibre process is estimated by weighting each measured direction by the reciprocal of its chance of being sampled, which can be inferred from the data. The concept of complete directional randomness for uniformly and independently distributed spatial directions is introduced. The cumulative distribution function of the angular distances between different directions and other exact relations are derived for complete randomness of vectorial and axial directions. A Monte Carlo method is constructed to test spatial fibre processes, whose fibres have negligibly small curvature, for complete directional randomness. Confocal scanning laser microscopy was used to study the angular distribution of glass fibres in a polymer composite which was subjected to increasing hydrostatic extrusion. The hypothesis of complete directional randomness had to be rejected for all samples with 1% probability of error. The directional distribution was of the bipolar type, with the principal axis directed parallel to the axis of extrusion. Progressive stretching of the material increased the degree of anisotropy of the glass fibres. Although presented for an application in polymer physics, the methods are general and may also be applied in biological investigations.

[1]  H. Gundersen,et al.  The isector: a simple and direct method for generating isotropic, uniform random sections from small specimens , 1992 .

[2]  H. Gundersen,et al.  Notes on the estimation of the numerical density of arbitrary profiles: the edge effect , 1977 .

[3]  H. J. G. GUNDERSEN,et al.  Some new, simple and efficient stereological methods and their use in pathological research and diagnosis , 1988, APMIS : acta pathologica, microbiologica, et immunologica Scandinavica.

[4]  E R Weibel,et al.  Recent stereological methods for cell biology: a brief survey. , 1990, The American journal of physiology.

[5]  D. Stoyan Stereological determination of orientations, second-order quantities and correlations for random spatial fibre systems , 1985 .

[6]  L. Devroye Non-Uniform Random Variate Generation , 1986 .

[7]  Nicholas I. Fisher,et al.  Statistical Analysis of Spherical Data. , 1987 .

[8]  K. Kanatani Stereological determination of structural anisotropy , 1984 .

[9]  I. Ward Optical and Mechanical Anisotropy in Crystalline Polymers , 1962 .

[10]  A. Gokhale Utility of the horizontal slice for stereological characterization of lineal features , 1993 .

[11]  Werner Nagel,et al.  Stationäre räumliche Faserprozesse und ihre Schnittzahlrosen , 1980, J. Inf. Process. Cybern..

[12]  M. Giné,et al.  Invariant Tests for Uniformity on Compact Riemannian Manifolds Based on Sobolev Norms , 1975 .

[13]  D. Stoyan,et al.  The pair correlation function for point and fibre systems and its stereological determination by planar sections * , 1985, Journal of microscopy.

[14]  K. Mardia Statistics of Directional Data , 1972 .

[15]  T. Mattfeldt,et al.  Second‐order stereology of benign and malignant alterations of the human mammary gland , 1993, Journal of microscopy.

[16]  J. E. Hilliard,et al.  Determination of Structural Anisotropy , 1967 .

[17]  B. Heller Circular Statistics in Biology, Edward Batschelet. Academic Press, London & New York (1981), 371, Price $69.50 , 1983 .

[18]  L. M. Cruz Orive,et al.  On the estimation of particle number. , 1980, Journal of microscopy.

[19]  A. Gokhale Estimation of length density Lv from vertical slices of unknown thickness , 1992 .

[20]  T. Mattfeldt,et al.  Estimation of surface area and length with the orientator , 1990, Journal of microscopy.

[21]  T. W. Anderson,et al.  Tests for randomness of directions against equatorial and bimodal alternatives , 1972 .

[22]  T. Mattfeldt,et al.  Orthogonal triplet probes: an efficient method for unbiased estimation of length and surface of objects with unknown orientation in space , 1985, Journal of microscopy.

[23]  C. Vyvyan Howard,et al.  Estimating the length of a bounded curve in three dimensions using total vertical projections , 1991 .

[24]  Ewald R. Weibel,et al.  Stereological Analysis of Anisotropic Structures using Directional Statistics , 1985 .

[25]  Viktor Beneš,et al.  A practical approach to the stereology of anisotropic structures , 1989 .

[26]  R. E. Miles The importance of proper model specification in stereology , 1978 .

[27]  Hans Jørgen G. Gundersen,et al.  Second-order stereology , 1990 .

[28]  Kanatani Ken-Ichi DISTRIBUTION OF DIRECTIONAL DATA AND FABRIC TENSORS , 1984 .

[29]  B. Möginger,et al.  Determination of the weighting function g(βi,r,vf) for fibre orientation analysis of short fibre-reinforced composites , 1991 .

[30]  Charles L. Tucker,et al.  Stereological measurement and error estimates for three-dimensional fiber orientation , 1992 .

[31]  H. J. G. Gundersen,et al.  The new stereological tools: Disector, fractionator, nucleator and point sampled intercepts and their use in pathological research and diagnosis , 1988, APMIS : acta pathologica, microbiologica, et immunologica Scandinavica.

[32]  G. S. Watson Statistics on Spheres , 1983 .

[33]  D. C. Sterio The unbiased estimation of number and sizes of arbitrary particles using the disector , 1984, Journal of microscopy.

[34]  Arun M. Gokhale,et al.  Unbiased estimation of curve length in 3‐D using vertical slices , 1990 .

[35]  H J Gundersen,et al.  Estimation of structural anisotropy based on volume orientation. A new concept , 1990, Journal of microscopy.

[36]  T. Mattfeldt,et al.  Dipyridamole-induced capillary endothelial cell proliferation in the rat heart--a morphometric investigation. , 1983, Cardiovascular research.

[37]  L. M. Orive,et al.  On the estimation of particle number , 1980 .

[38]  Alan J. Lee,et al.  A COMPARISON OF TESTS OF UNIFORMITY FOR SPHERICAL DATA , 1985 .

[39]  Gary Ulrich,et al.  Computer Generation of Distributions on the M‐Sphere , 1984 .

[40]  N. C. Davidson,et al.  Measurements of fibre direction in reinforced polymer composites , 1993 .

[41]  T. Mattfeldt Stochastic Geometry and Its Applications , 1996 .

[42]  E. Weibel,et al.  Estimating length density and quantifying anisotropy in skeletal muscle capillaries , 1983, Journal of microscopy.

[43]  E. Philofsky,et al.  On the measurement of the orientation distribution of lineal and areal arrays , 1969 .

[44]  T. Mattfeldt,et al.  Estimation of length and surface of anisotropic capillaries , 1984, Journal of microscopy.