Original article3D laser surface scanning and geometric morphometric analysis of craniofacial shape as an index of cerebro-craniofacial morphogenesis: initial application to sexual dimorphism
Introduction
Though numerous lines of evidence continue to indicate abnormalities of early brain development in schizophrenia Murray and Lewis 1987, Weinberger 1987, Ward and Jamison 1991, Jones et al 1994, Waddington et al 1999a, few constitute ‘hard’ biological findings which might inform specifically as to the nature and timing of the underlying developmental disturbance(s). Neuropathological examination of brain structure and cytoarchitecture has provided important information in this regard (Harrison and Roberts 2000), but such studies are predicated on the demise of often elderly patients. Thus, clarification of these issues would be aided considerably by an index of early developmental disturbance that could be accessed readily in living patients and related to clinical and other biological assessments.
Over early fetal life, cerebral morphogenesis proceeds in exquisite embryological intimacy with craniofacial morphogenesis, such that classical neurodevelopmental disorders such as Down’s syndrome and velo-cardiofacial syndrome are well recognized to be characterized also by dysmorphic features which can involve several body regions but effect the craniofacies in particular Smith 1988, Kjaer 1995, Waddington et al 1999a. Minor physical anomalies (MPAs) are slight anatomical malformations, usually of no medical or cosmetic significance, which constitute biological markers of first/early second trimester dysmorphogenesis; several studies have now indicated MPAs to occur to excess among patients with schizophrenia, with the most consistent findings being subtle dysmorphogenesis of the craniofacial region Lane et al 1996, Waddington et al 1998, McNeil et al 2000; however, the topography of essentially qualitative MPAs, and hence their full biological import, is poorly understood.
On applying an anthropometric approach, we have recently found patients with schizophrenia to show numerous quantitative as well as qualitative dysmorphic features, particularly of craniofacial structures, with a ‘core’ topography of dysmorphology characterized by an overall narrowing and elongation of the mid- and lower anterior [frontonasal] regions of the face (Lane et al 1997). On developmental considerations, this topography is most consistent with dysmorphogenesis over a time-frame of weeks 9/10–14/15 of gestation, during which fundamental relationships between frontal brain regions and the anterior midface are determined Cohen et al 1993, Diewert and Lozanoff 1993a, Diewert and Lozanoff 1993b, Diewert et al 1993, Kjaer 1995, Waddington et al 1999a, Waddington et al 1999b; however, this anthropometric approach, while considerably increasing our understanding of these processes, involves linear [i.e., 2-dimensional] measurements of inherently 3-dimensional morphology, and thus fails to capture those important, and potentially critical, geometric relationships that are essential elements of developmental processes. To realize the potential of such an approach would require recording of the craniofacial surface in 3D, obtained ideally using portable rather than static technology, and numerical output; furthermore, biological interpretability would be greatly facilitated if the statistical model of facial shape differences could also be directly visualized.
3D digitization technologies now allow facial surfaces to be recorded and 3D landmark coordinates obtained, while geometric morphometrics, which analyses landmark coordinates directly, provides the basis for modeling differences in craniofacial shape both statistically and visually. We describe here, for the first time, the use of a portable, hand-held 3D laser surface scanning system and geometric morphometrics to study craniofacial morphology, via a logistic regression-based method, and its application to sexual dimorphism as an exemplar. The developmental biology of sexual dimorphism is important in itself (Swaab and Hofman 1984), but has direct relevance also to schizophrenia in that gender differences in the phenomenology, biology and course of illness are well established Goldstein 1988, Goldstein 1996, Lewis 1992.
Section snippets
Subjects
Following approval by the Ethics [Research] Committee of the Royal College of Surgeons in Ireland, 131 otherwise unselected members of staff [82 females: age 32.2 (SD 9.2), range 20–59 years; 49 males, age 33.1 (9.7), range 22–65 years] gave written, informed consent to participation in the study; all subjects were of Irish, Scottish, Welsh or English origin.
3D laser surface scanning
Facial surfaces of the subjects were recorded in 3D using a portable, handheld laser scanner [Polhemus FastScan, Polhemus Inc, Vermont,
Reliability
Intra-class correlation coefficients were 0.99 for centroid size and 0.88–0.99 for interlandmark distances.
Centroid size
Centroid size was smaller in females [229.8 ± 5.9 mm] than in males [247.8 ± 8.0 mm]. Gender difference in centroid size on controlling for height and age was investigated by analysis of covariance. The centroid size differences, female minus male, were: no covariate −17.9 ± 1.2 mm [t = −14.7, p < .001]; height as covariate −11.2 ± 1.6 mm [t = −6.82, p < .001]; height and age as covariates
Discussion
This study describes, for the first time, the application of a portable, hand-held 3D laser surface scanning system, rigorous statistical analysis of 3D coordinate data using geometric morphometrics, and dynamic 3D graphic display. In applying it to the investigation of sexual dimorphism in craniofacial development, a primary purpose was to exemplify the utility of this approach in relation to an issue both of inherent interest to clinical neuroscience and of relevance to the developmental
Acknowledgements
The authors’ studies are supported by the Stanley Foundation.
References (47)
Shape and the information in medical imagesA decade of the morphometric synthesis
Comput Vis Image Understand
(1997)- et al.
A three-dimensional computerised mesh diagram analysis and its application in soft tissue facial morphometry
Am J Orthod Dentofacial Orthop
(1998) - et al.
Child developmental risk factors for adult schizophrenia in the British 1946 birth cohort
Lancet
(1994) - et al.
Obstetric complications and congenital malformations in schizophrenia
Brain Res Rev
(2000) - et al.
Sexual dimorphism in the human brainEvaluation of tissues volume, tissue composition and surface anatomy using magnetic resonance imaging
Psychiat ResNeuroimaging
(2000) - et al.
Comparaison tridimensionnelle des cranes de Pan et Pongo par superpositions procrusteennes
C R Acad Sci Paris Sciences de la Vie
(1999) - et al.
Structural differences in the cerebral cortex of healthy female and male subjectsA magnetic resonance imaging study
Psychiat Res Neuroimaging
(1995) - et al.
Sexual differentiation of the human brainA historical perspective
Prog Brain Res
(1984) - et al.
The neurodevelopmental basis of schizophreniaClinical clues from cerebro-craniofacial dysmorphogenesis, and the roots of a lifetime trajectory of disease
Biol Psychiatry
(1999) - et al.
Neurodevelopmental and neuroprogressive processes in schizophrenia
Psychiatric Clin North Am
(1998)