Table 2 Performance of the proposed de-arraying method on three datasets under four setting options: (1) both the segmentation and the non-linear estimation (for the deformation) modules are deactivated, (2) the segmentation is activated but the non-linear estimation is deactivated, (3) the segmentation is deactivated and the non-linear estimation is activated, and (4) both of them are activated
From: ATMAD: robust image analysis for Automatic Tissue MicroArray De-arraying
| Â | Localization | Â | Estimation of array coordinates | Â | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Â | A (a) | P (b) | R (c) | F (d) | G (e) | JSC (f) | A (a) | P (b) | R (c) | F (d) | G (e) | JSC (f) |
SIMULATED IMAGES | Â | |||||||||||
Average on 31 images | Â | |||||||||||
Wang et al. [22] | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 |
ATMAD Opt. #1 | 1 | 1 | 1 | 1 | 1 | 1 | 0.93 | 1 | 0.93 | 0.96 | 0.96 | 0.93 |
Opt. #2 | 0.95 | 1 | 0.95 | 0.98 | 0.98 | 0.95 | 0.85 | 1 | 0.84 | 0.91 | 0.91 | 0.84 |
Opt. #3 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Opt. #4 | 0.95 | 1 | 0.95 | 0.98 | 0.98 | 0.95 | 0.95 | 1 | 0.95 | 0.98 | 0.98 | 0.95 |
Fig. 11a | Â | |||||||||||
ATMAD Opt. #1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Opt. #2 | 0.97 | 1 | 0.96 | 0.98 | 0.98 | 0.96 | 0.97 | 1 | 0.96 | 0.98 | 0.98 | 0.96 |
Opt. #3 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Opt. #4 | 0.97 | 1 | 0.96 | 0.98 | 0.98 | 0.96 | 0.97 | 1 | 0.96 | 0.98 | 0.98 | 0.96 |
Fig. 11b | Â | |||||||||||
ATMAD Opt. #1 | 1 | 1 | 1 | 1 | 1 | 1 | 0.93 | 1 | 0.92 | 0.96 | 0.96 | 0.92 |
Opt. #2 | 0.93 | 1 | 0.92 | 0.96 | 0.96 | 0.92 | 0.90 | 1 | 0.89 | 0.94 | 0.95 | 0.89 |
Opt. #3 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Opt. #4 | 0.93 | 1 | 0.92 | 0.96 | 0.96 | 0.92 | 0.93 | 1 | 0.92 | 0.96 | 0.96 | 0.92 |
BRIGHTFIELD IMAGES | Â | |||||||||||
Average on 8 images | Â | |||||||||||
ATMAD Opt. #1 | 0.94 | 1 | 0.93 | 0.96 | 0.96 | 0.93 | 0.91 | 1 | 0.88 | 0.94 | 0.94 | 0.88 |
Opt. #2 | 0.87 | 1 | 0.83 | 0.91 | 0.91 | 0.83 | 0.84 | 1 | 0.79 | 0.88 | 0.89 | 0.79 |
Opt. #3 | 0.94 | 1 | 0.93 | 0.96 | 0.96 | 0.93 | 0.94 | 1 | 0.93 | 0.96 | 0.96 | 0.93 |
Opt. #4 | 0.87 | 1 | 0.83 | 0.91 | 0.91 | 0.83 | 0.93 | 1 | 0.91 | 0.95 | 0.95 | 0.91 |
Fig. 12 | Â | |||||||||||
ATMAD Opt. #1 | 0.96 | 1 | 0.94 | 0.97 | 0.97 | 0.94 | 0.91 | 1 | 0.89 | 0.94 | 0.94 | 0.89 |
Opt. #2 | 0.87 | 1 | 0.83 | 0.91 | 0.91 | 0.83 | 0.84 | 1 | 0.80 | 0.89 | 0.89 | 0.80 |
Opt. #3 | 0.96 | 1 | 0.94 | 0.97 | 0.97 | 0.94 | 0.96 | 1 | 0.94 | 0.97 | 0.97 | 0.94 |
Opt. #4 | 0.87 | 1 | 0.83 | 0.91 | 0.91 | 0.83 | 0.93 | 1 | 0.91 | 0.95 | 0.95 | 0.91 |
FLUORESCENCE IMAGES | Â | |||||||||||
Average on 8 DNA microarray images | Â | |||||||||||
ATMAD Opt. #1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Opt. #2 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Opt. #3 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Opt. #4 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Fig. 13 | Â | |||||||||||
ATMAD Opt. #1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Opt. #2 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Opt. #3 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Opt. #4 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Average on 4 TMA images | Â | |||||||||||
ATMAD Opt. #1 | 0.79 | 0.77 | 1 | 0.87 | 0.88 | 0.77 | 0.92 | 0.98 | 0.93 | 0.95 | 0.95 | 0.91 |
Opt. #2 | 0.91 | 0.99 | 0.90 | 0.94 | 0.95 | 0.89 | 0.73 | 1 | 0.69 | 0.82 | 0.83 | 0.69 |
Opt. #3 | 0.79 | 0.77 | 1 | 0.87 | 0.88 | 0.77 | - | - | - | - | - | - |
Opt. #4 | 0.91 | 0.99 | 0.90 | 0.94 | 0.95 | 0.89 | 0.92 | 0.98 | 0.93 | 0.96 | 0.96 | 0.91 |
Fig. 14 | Â | |||||||||||
ATMAD Opt. #1 | 0.88 | 0.86 | 1 | 0.93 | 0.93 | 0.86 | 0.94 | 1 | 0.94 | 0.97 | 0.97 | 0.94 |
Opt. #2 | 0.86 | 1 | 0.84 | 0.91 | 0.92 | 0.84 | 0.80 | 1 | 0.77 | 0.87 | 0.88 | 0.77 |
Opt. #3 | 0.88 | 0.86 | 1 | 0.93 | 0.93 | 0.86 | 0.97 | 1 | 0.97 | 0.98 | 0.98 | 0.97 |
Opt. #4 | 0.86 | 1 | 0.84 | 0.91 | 0.92 | 0.84 | 1 | 1 | 1 | 1 | 1 | 1 |