Back

IMCellXMBD: A statistical approach for robust cell identification and quantification from imaging mass cytometry images

Xiao, X.; Su, N.; Kong, Y.; Zhang, L.; Ding, X.; Yang, W.; Yu, R.

2021-09-27 bioinformatics
10.1101/2021.09.27.461899 bioRxiv
Show abstract

Imaging Mass Cytometry (IMC) has become a useful tool in biomedical research due to its capability to measure over 100 markers simultaneously. Unfortunately, some protein channels in IMC images can be very noisy, which may significantly affect the phenotyping results without proper data processing. We developed IMCellXMBD1, a highly effective and generalizable cell identification and quantification method for IMC images. IMCell performs denoising by subtracting an estimated background noise value from pixel values for each individual protein channel, identifies positive cells from negative cells by comparing the distribution between segmented cells and decoy cells, and normalize the protein expression levels of the identified positive cells for downstream data analysis. Experimental results demonstrate that our method significantly improves the reliability of cell phenotyping which is essential for using IMC in biomedical studies.

Matching journals

The top 5 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.