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

Table 2 Unmixing results using a single image of smFISH data from mouse brain tissue sections

Method	DAPI	Opal 520	Opal 570	Opal 620	Opal 690	Mean
Root mean squared error (RMSE)
FCLSU	0.0052	0.0023	0.0225	0.0079	0.0021	0.0080
ELMM	0.0035	0.0012	0.0248	0.0083	0.0023	0.0080
GELMM	0.0052	0.0021	0.0252	0.0080	0.0021	0.0085
SØrensen–Dice similarity coefficient
FCLSU	0.6866	0.7574	0.8547	0.4545	0.4752	0.6457
ELMM	0.8559	0.8179	0.7799	0.3918	0.5196	0.6730
GELMM	0.7493	0.7927	0.7795	0.4394	0.5001	0.6522
Structural similarity index (SSIM)
FCLSU	0.8510	0.9737	0.9682	0.9754	0.9668	0.9470
ELMM	0.9306	0.9963	0.9713	0.9660	0.9580	0.9644
GELMM	0.8466	0.9736	0.9631	0.9748	0.9677	0.9452

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

ISSN: 1471-2202