Back to Search Start Over

Three-Dimensional Pathology Specimen Modeling Using "Structure-From-Motion" Photogrammetry: A Powerful New Tool for Surgical Pathology.

Authors :
Turchini J
Buckland ME
Gill AJ
Battye S
Source :
Archives of pathology & laboratory medicine [Arch Pathol Lab Med] 2018 Nov; Vol. 142 (11), pp. 1415-1420. Date of Electronic Publication: 2018 May 30.
Publication Year :
2018

Abstract

Context.—: Three-dimensional (3D) photogrammetry is a method of image-based modeling in which data points in digital images, taken from offset viewpoints, are analyzed to generate a 3D model. This modeling technique has been widely used in the context of geomorphology and artificial imagery, but has yet to be used within the realm of anatomic pathology.<br />Objective.—: To describe the application of a 3D photogrammetry system capable of producing high-quality 3D digital models and its uses in routine surgical pathology practice as well as medical education.<br />Design.—: We modeled specimens received in the 2 participating laboratories. The capture and photogrammetry process was automated using user control software, a digital single-lens reflex camera, and digital turntable, to generate a 3D model with the output in a PDF file.<br />Results.—: The entity demonstrated in each specimen was well demarcated and easily identified. Adjacent normal tissue could also be easily distinguished. Colors were preserved. The concave shapes of any cystic structures or normal convex rounded structures were discernable. Surgically important regions were identifiable.<br />Conclusions.—: Macroscopic 3D modeling of specimens can be achieved through Structure-From-Motion photogrammetry technology and can be applied quickly and easily in routine laboratory practice. There are numerous advantages to the use of 3D photogrammetry in pathology, including improved clinicopathologic correlation for the surgeon and enhanced medical education, revolutionizing the digital pathology museum with virtual reality environments and 3D-printing specimen models.

Details

Language :
English
ISSN :
1543-2165
Volume :
142
Issue :
11
Database :
MEDLINE
Journal :
Archives of pathology & laboratory medicine
Publication Type :
Academic Journal
Accession number :
29846102
Full Text :
https://doi.org/10.5858/arpa.2017-0145-OA