Modern medical imaging techniques, such as solid CT and laser light scanning, record and display surface form in remarkable detail. Yet the application of such images in scientific and clinical research lags badly behind the elegance of their visualization. Most quantitative studies rely on either the extraction of volumes or the location of discrete, named points, """"""""landmarks,"""""""" for subsequent statistical processing. However, most of the references investigators make to solid biological surfaces are anchored not at points but upon extended features. Geometrically, these may be curves, like the orbital rim, or whole surface patches, like the forehead. Except for the trivial integrals (distances and angles among points, and enclosed volumes of surfaces) there exist no morphometric methods for these extended records of biological form. This grant proposes a first step in this direction. A ridge curve, such as the orbital rim or the lower border of the mandible, is a surface feature characterized by the directional dominance of one principal curvature over the other at all its points. We have developed an algorithm for the automatic extraction of such curves from surfaces detected in solid CT scans and from laser light scans of surfaces. This grant proposes a biometric method for such curves.
Our Specific Aims i nclude the improvement of algorithms for their tracing, optimization of automatic methods for landmark point extraction upon them, and tests of their efficacy in typical biometric tasks such as Apert syndromology. Exploiting recent developments in the measurement of deformation and in the application of multivariate statistics in morphometrics, we will calibrate the reliability of many different aspects of the curves and the landmarks upon them, and we will carefully separate the information content of these curves into its components at different physical scales and over the different regions of the cranium. Ridge curves represent a rich concentration of crucial information about curving surfaces. Our techniques, applicable to a class of images already routinely clinically archived, should be of value in studies of many other organs: livers, pelves, the heart, and others. Future research will determine features of the curves most relevant for diagnosis and treatment of diverse morphological anomalies and will exploit the information content of these curves to pursue even more complete decompositions of volume-sampled anatomy by incorporating other tools of differential geometry.

Agency
National Institute of Health (NIH)
Institute
National Institute of Dental & Craniofacial Research (NIDCR)
Type
Research Project (R01)
Project #
1R01DE009560-01A1
Application #
3223330
Study Section
Oral Biology and Medicine Subcommittee 1 (OBM)
Project Start
1991-04-01
Project End
1994-03-31
Budget Start
1991-04-01
Budget End
1992-03-31
Support Year
1
Fiscal Year
1991
Total Cost
Indirect Cost
Name
New York University
Department
Type
Schools of Medicine
DUNS #
004514360
City
New York
State
NY
Country
United States
Zip Code
10012
Cutting, C; Grayson, B; McCarthy, J G et al. (1998) A virtual reality system for bone fragment positioning in multisegment craniofacial surgical procedures. Plast Reconstr Surg 102:2436-43
Dean, D; Hublin, J J; Holloway, R et al. (1998) On the phylogenetic position of the pre-Neandertal specimen from Reilingen, Germany. J Hum Evol 34:485-508
Ziegler, R; Dean, D (1998) Mammalian fauna and biostratigraphy of the pre-Neandertal site of Reilingen, Germany. J Hum Evol 34:469-84
Cutting, C; Dean, D; Bookstein, F L et al. (1995) A three-dimensional smooth surface analysis of untreated Crouzon's syndrome in the adult. J Craniofac Surg 6:444-53