Hugging Face's logo

Datasets:
ChipYTY
/
segmamba

Dataset card Files
xet
Community
segmamba
File size: 2,230 Bytes
fe8202e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
 
# Visualization config for gliomasam3_moe
# Output: vis_res/ (each visualization type in a separate subfolder)

data:
  format: "segmamba_npz"
  root_dir: "/data/yty/brats23_segmamba_processed"
  npz_dir: "/data/yty/brats23_segmamba_processed"
  modalities: ["t1n", "t1c", "t2f", "t2w"]
  seg_name: "seg"

predictions:
  ours:
    name: "GliomaSAM3-MoE"
    dir: "/root/githubs/gliomasam3_moe/vis_res/predictions_ours"
    type: "label"
  baselines:
    - name: "SegMamba"
      dir: "/root/githubs/SegMamba/prediction_results/segmamba_brats23_ep799"
      type: "segmamba_3c"
  aux_dir: "/root/githubs/gliomasam3_moe/vis_res/aux_cache"

visualization:
  output_dir: "/root/githubs/gliomasam3_moe/vis_res"
  overlay_modality: "t1c"
  alpha: 0.45
  colors:
    WT: [1.0, 0.85, 0.0]
    TC: [0.0, 1.0, 0.25]
    ET: [1.0, 0.0, 0.0]
  boundary_region: "ET"
  frag_bins: [1, 3, 5]
  scale_bins: [50, 200, 500]

# Case selection
cases:
  # 1) Main qualitative comparison (Ours vs SegMamba)
  qualitative:
    - "BraTS-GLI-00005-000"
    - "BraTS-GLI-00006-000"
    - "BraTS-GLI-00017-000"

  # 2) ET-absent case study (best candidate with highest pre/post diff)
  et_absent:
    - "BraTS-GLI-00012-000"

  # 3) Boundary-focused visualization
  boundary:
    - "BraTS-GLI-00005-000"
    - "BraTS-GLI-00017-000"

  # 4) Tiny/fragmented ET cases
  tiny_et:
    - "BraTS-GLI-00005-000"
    - "BraTS-GLI-00006-000"

  # 5) Cross-year robustness (placeholder - needs cross-year data)
  cross_year: {}

  # 6) MoE routing interpretability
  moe:
    - "BraTS-GLI-00018-000"
    - "BraTS-GLI-00012-000"

  # 7) Concept tokens interpretability
  concept_tokens:
    - "BraTS-GLI-00005-000"
    - "BraTS-GLI-00006-000"

  # 8) Dual-domain enhancement effect
  dual_domain:
    - "BraTS-GLI-00005-000"
    - "BraTS-GLI-00017-000"

  # 9) Augmentation robustness (AmpMix)
  ampmix:
    - base: "BraTS-GLI-00005-000"
      mix: "BraTS-GLI-00006-000"
      lam: 0.5

  # 10) Failure cases / limitations
  failure:
    - "BraTS-GLI-00020-000"
  failure_notes:
    "BraTS-GLI-00020-000": "potential boundary ambiguity"

# 11) Efficiency / inference visualization
efficiency:
  case_id: "BraTS-GLI-00005-000"
  roi_size: [128, 128, 128]
  overlap: 0.5