Skip to main content
Fig. 5 | BMC Bioinformatics

Fig. 5

From: CellSeg: a robust, pre-trained nucleus segmentation and pixel quantification software for highly multiplexed fluorescence images

Fig. 5

Testing lateral bleed compensation on a CODEX dataset of colorectal cancer samples. A. Schematic demonstrating post-processing of CellSeg segmentation with following steps. (1) Grow cell boundaries by user defined number of pixels (growth of two pixels shown). (2) Compute inverse adjacency matrix from cell–cell. (3) Multiply inverted adjacency matrix by marker pixel intensity vector to obtain compensated single-cell pixel quantifications table. B. Effects of lateral bleed compensation on double-positive cell populations in the CRC dataset for three pairs of mutually exclusive markers (CD8 vs. CD4, Cytokeratin vs. CD45, CD20 vs. CD3). Data shown are from one of the two CRC TMAs (TMA A), with comparable bleed compensation performance for the other TMA (TMA B, data not shown)

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us