Table 2.
The definition of cell populations used to classify cells in ACCEPT after deep learning segmentation in 300 non-small cell lung cancer (NSCLC) patient samples and 127 samples from healthy volunteers.
Cell Population | Mean Intensity Value | Overlay Nucleus | Size (µm2) | |||||||
---|---|---|---|---|---|---|---|---|---|---|
DAPI | CD45 | CK | CD16 1 | wga 1 | with CK | with CD45 | with CD16 | CK | ||
CTC | CK+/CD45- | >0 | 0 | ≥50 | 0 | >0 | >0.4 | ND | ND | >9 |
Leukocytes | CD45+/CD16- | >0 | >0 | 0 | 0 | >0 | ND | >0 | ND | ND |
CD45+/CD16+ | >0 | >0 | 0 | >0 | >0 | ND | >0 | >0 | ND | |
CD45-/CD16+ | >0 | 0 | 0 | >0 | >0 | ND | ND | >0 | ND | |
Nucleus | Bare nucleus | >0 | 0 | 0 | 0 | 0 | ND | ND | ND | ND |
Unstained cell | >0 | 0 | 0 | 0 | >0 | ND | ND | ND | ND |
1 Only for those samples in which CD16 or wga was added to the immunostaining cocktail. Abbreviations used: DAPI: 4′6-diamidino-2-phenylindole; CD: cluster of differentiation; CK: cytokeratins; wga: wheat germ agglutinin.