AUTOMATIC EXTRACTION OF 2D AND 3D MORPHOLOGICAL DATA FROM HIGH-THROUGHPUT DARK-FIELD MICROSCOPY IMAGES OF FORAMINIFERA AND TEETH

Advances in computational techniques and storage technologies have ushered in the era of big data across the sciences. The ability to generate large, digitized datasets necessitates the concurrent development of methods for extracting relevant information from this preponderance of data in order to address scientific queries. Here, we present the run2dmorph and run3dmorph software packages (part of the Automorph software suite: http://people.earth.yale.edu/automorph/pincelli-hull), which automatically extract measures of both 2D and 3D morphology from focus-stacked images of individual foraminifera. Measures of 2D morphology include area, perimeter, rugosity, aspect ratio, etc.; the output for 3D morphology includes volume (with minimum and maximum bounds calculated based on assumptions of a pyramidal, cubic, or ellipsoidal base) and 3D meshes in Wavefront OBJ format and Open file format (OFF) for easy downstream morphometric analysis. To demonstrate the software, examples utilizing a publicly available database of high-throughput dark-field microscopy images of foraminifera (via the Yale Peabody Museum of Natural History; Elder et al. 2015) will be presented. The run2dmorph and run3dmorph software can be used with images of any type taken at known focal depths, and has additionally been applied to images of limpets and fish teeth imaged at known focal planes to date.