Table 1.
Quantitative comparison of virtual H&E staining on two TMA sections of lung cancerous tissues from intensity, α-FLIM, and IW-FLIM using the normalised root-mean-squared error (NRMSE), the normalised mutual information (MNI), the peak signal-to-noise ratio (PSNR), and the mean structural similarity index metric (MSSIM).
| NRMSE | NMI | PSNR | MSSIM | ||
|---|---|---|---|---|---|
| Lung sample 1 (Fig. 1a) | Intensity | 0.127 | 1.093 | 20.085 | 0.529 |
| a-FLIM | 0.127 | 1.091 | 20.121 | 0.53 | |
| IW-FLIM | 0.126 | 1.096 | 20.202 | 0.533 | |
| Lung sample 2 (Fig. 1c) | Intensity | 0.095 | 1.154 | 22.767 | 0.689 |
| a-FLIM | 0.093 | 1.161 | 22.962 | 0.695 | |
| IW-FLIM | 0.092 | 1.164 | 23.026 | 0.697 |
Note that smaller NRMSE and NMI indicate more similar images, whereas larger PSNR and MSSIM represent better similarity between images.