Skip to main content

Table 5 Unmixing results using a single image of immunofluorescence data from a post-mortem human brain tissue section from an AD donor

From: SUFI: an automated approach to spectral unmixing of fluorescent multiplex images captured in mouse and post-mortem human brain tissues

Method

DAPI

Abeta

pTau

MAP2

Lipofuscin

Mean

Root mean squared error (RMSE)

 FCLSU

0.0087

0.0039

0.0031

0.0031

0.0031

0.0059

 ELMM

0.0081

0.0034

0.0031

0.0025

0.0025

0.0055

 GELMM

0.0081

0.0034

0.0036

0.0025

0.0025

0.0055

SØrensen–Dice similarity coefficient

 FCLSU

0.4362

0.7778

0.7016

0.6938

0.6938

0.7065

 ELMM

0.4436

0.7265

0.5257

0.6988

0.6988

0.6660

 GELMM

0.4263

0.7118

0.7699

0.6022

0.6022

0.6871

Structural similarity index (SSIM)

 FCLSU

0.9511

0.9819

0.9627

0.9842

0.9842

0.9282

 ELMM

0.9566

0.9886

0.9565

0.9891

0.9891

0.9334

 GELMM

0.9568

0.9911

0.9443

0.9835

0.9835

0.9330

  1. Here we compare the performance of SUFI against the ZEN unmixed image using three metrics (RMSE, dice similarity, SSIM) for all three methods (FCLSU, ELMM, GELMM). For the three metrics, values range between 0 and 1 with better performance indicated by lower RMSE values and higher SSIM and dice similarity values. Mean over all channels is presented as last column