Back

CytoPb: Segmentation-free and threshold-free analysis of highly multiplexed images of biological tissue

Barber, P. R.; Dolcetti, L.; Yeh, C.-A.; Ng, T.

2023-07-29 immunology
10.1101/2023.07.26.550516 bioRxiv
Show abstract

We present a novel method and code, CytoPb, for the analysis of multiplexed images that is segmentation-free and threshold-free and produces results that correlate well with existing methods but is faster to run, needs minimal training and removes subjectivity associated with the validation of single-cell segmentation and thresholding. In three data sets cell abundance measures correlate with previous methods (Pearsons coef. >0.75). Neighbourhood analysis and cell-cell proximity are possible.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.