Skip to main content
. 2025 Jun 14;20(7):1551–1560. doi: 10.1007/s11548-025-03405-1

Table 2.

Comparison of Dice scores for all categories after incremental learning of experiment 7–4 (2 steps) using the DeepLabV3+ framework

Step 0 Step 1 Mean
AWL COL LIV PAN SIN SPL STO URE VGL IMA INV
DeepLabV3+ Framework
Fine tuning 0.00 0.00 0.00 0.00 0.00 0.00 0.00 9.01 23.54 29.14 20.51 7.47
MiB [9] 68.60 40.68 52.86 33.69 43.61 35.67 48.62 10.18 25.76 26.79 45.76 39.29
PLOP [13] 68.46 37.28 45.73 21.95 43.61 26.51 38.75 8.98 23.05 24.05 41.58 34.54
SSUL [15] 79.98 58.26 71.49 40.84 74.76 52.47 4.99 24.62 40.95 44.43 46.43 49.02
InSeg [14] 78.42 63.07 47.10 44.63 72.66 55.89 12.01 20.36 39.64 44.07 46.98 47.71
NeST [16] 75.68 50.75 73.85 45.61 60.85 47.93 56.38 21.70 33.77 25.63 33.27 47.77
IDEC [17] 83.14 67.14 76.50 29.27 77.59 58.88 61.49 21.61 46.65 51.01 43.10 56.03
Ours 86.87 74.54 74.93 31.82 80.49 71.33 61.59 30.85 48.80 53.05 56.66 60.99
Offline 78.40 73.00 70.60 46.30 77.00 71.50 67.90 18.80 39.10 55.60 60.90 59.92
ViT Encoder + Mask Decoder (MedSAM) Framework
MBS [18] 76.44 74.24 82.24 37.01 79.12 48.76 33.20 32.21 17.10 53.36 56.07 53.61
Ours 82.34 72.32 78.27 43.29 80.15 70.37 23.64 23.15 42.51 46.61 50.70 55.76
Offline 72.62 73.97 80.08 48.53 76.54 45.82 55.66 21.16 42.32 46.77 49.55 55.73

Highest results and second highest results are highlighted in bold and underlined, respectively