US 12,308,120 B2
Analysis and characterization of epithelial tissue structure
Xavier Descombes, Sophia Antipolis (FR); Imane Lboukili, Val de Reuil (FR); Thierry Oddos, Clamart (FR); and Georgios N. Stamatas, Val de Reuil (FR)
Assigned to Kenvue Brands LLC/Inria (Institut national de recherche en informatique et en automatique), Summit, NJ (US)
Filed by Kenvue Brands LLC, Summit, NJ (US); and Inria (Institut national de recherche en informatique et en automatique), Les Chesnay-Rocquencourt (FR)
Filed on Sep. 13, 2022, as Appl. No. 17/943,477.
Claims priority of provisional application 63/244,981, filed on Sep. 16, 2021.
Prior Publication US 2023/0084952 A1, Mar. 16, 2023
Int. Cl. G16H 50/20 (2018.01)
CPC G16H 50/20 (2018.01) 13 Claims
OG exemplary drawing
 
1. A non-invasive or minimally invasive method for characterizing epithelial tissue structure, comprising:
identifying cell coordinates information in an unknown image segment, wherein the unknown image segment is a Reflectance Confocal Microscopy image and comprises tissue area and dark background, and wherein identifying the cell coordinates information comprises:
performing, on the unknown image segment, image processing to differentiate between the tissue area and the dark ground; and
identifying individual cells in the tissue area by producing a further filtered unknown image segment;
determining a degree of match between the further filtered unknown image segment and a known epithelial tissue structure model image segment by comparing, by one or more computing devices, the segmented cellular areas in the further filtered unknown image segment to cellular areas in the known epithelial tissue structure model image segment, wherein determining the degree of match further comprises determining a difference value relative to a derived value from the known epithelial tissue structure model image segment;
comparing, by the one or more computing devices, the determined degree of match to a predetermined value; and
characterizing, by the one or more computing devices, the further filtered unknown image segment as an epithelial tissue structure based on the comparison of the determined degree of match to the predetermined value indicating the determined difference value is equal to or below a predetermined threshold.