Update README: literature comparison + 3D versions (ACDC 3D, BraTS 2021 3D)

README.md

@@ -18,15 +18,19 @@ tags:
 - mri
 ---
 
-# MedSeg-7D: Seven Public 2D Medical Segmentation Benchmarks (with Canonical Splits)
+# MedSeg-7D: Seven Public Medical Segmentation Benchmarks (2D + 3D)
 
-A curated bundle of seven public 2D medical segmentation datasets, packaged
+A curated bundle of seven public medical segmentation datasets, packaged
 with **canonical leakage-free splits** for the four datasets where one is
 needed (ACDC patient-level, BraTS2020 volume-level, CVC-ClinicDB
 video-level, plus seed-fixed image-level for the rest). All raw images and
 masks are retained at their original resolution; no resizing, no
 preprocessing baked in.
 
+For the two volumetric MRI datasets (ACDC, BraTS), this release ships
+**both 2D slice extracts and the original 3D NIfTI volumes**, so users can
+choose 2D or 3D pipelines without re-downloading.
+
 This is the **dataset-only release** that accompanied an evaluation-protocol
 audit of pixel- vs.\ latent-space diffusion augmentation for medical image
 segmentation. The bundle is reusable for any 2D medical-segmentation
@@ -46,10 +50,18 @@ research, not just the original study.
 ```
 MedSeg-7D/
 ├── README.md
-├── ACDC/                         (cardiac MRI, 100 patients → 1841 slices)
-│   ├── images/                   patient<id>_frame<f>_slice_<s>.png
-│   ├── masks/                    matching mask filenames
+├── ACDC/                         (cardiac MRI, 100 patients)
+│   ├── images/                   2D slices: patient<id>_frame<f>_slice_<s>.png
+│   ├── masks/                    matching 2D-slice mask filenames (any-structure binary)
+│   ├── 3D/                       ORIGINAL 3D NIfTI volumes (challenge layout)
+│   │   ├── training/
+│   │   │   ├── patient001/       Info.cfg + patient001_4d.nii.gz + frame01.nii.gz + frame01_gt.nii.gz + frame12.nii.gz + frame12_gt.nii.gz
+│   │   │   └── ... (100 patients)
+│   │   └── testing/              50 held-out patients (challenge test set)
 │   └── split_info.json           CANONICAL patient-level split (seed=42, 80/20)
+│   # 2D and 3D share the same 100-patient training cohort. The 3D side
+│   # additionally ships the official 50 challenge test patients, which the
+│   # 2D side does NOT include (we re-split the 100 train patients 80/20).
 │
 ├── BraTS2020/                    (brain MRI FLAIR, 369 volumes → 22677 slices)
 │   ├── images/
@@ -64,6 +76,16 @@ MedSeg-7D/
 │   # HuggingFace's 10000 files-per-directory limit. Filenames preserve the
 │   # original volume_X_slice_Y.png convention.
 │
+├── BraTS2021_3D/                 (brain MRI 3D NIfTI, 1251 patients — superset of BraTS2020)
+│   ├── BraTS2021_00000/          5 NIfTI files: t1, t1ce, t2, flair, seg (4 modalities + GT)
+│   ├── BraTS2021_00002/
+│   └── ... (1251 patient dirs)
+│   # IMPORTANT: This is BraTS *2021*, a SUPERSET of BraTS 2020. The 369
+│   # volumes in our 2D `BraTS2020/` are a subset of the 1251 here. Patient
+│   # IDs differ between the 2020 and 2021 releases, so split_info.json
+│   # (volume-level for 2020) does NOT apply to BraTS2021_3D directly. Do
+│   # not mix 2D and 3D Dice numbers.
+│
 ├── BUSI/                         (breast ultrasound, 780 images)
 │   ├── images/
 │   └── masks/                    masks suffixed _mask.png
@@ -83,10 +105,10 @@ MedSeg-7D/
 │   ├── bbox/                     bounding boxes (auxiliary)
 │   └── kavsir_seg_README.md      original release notes
 │
-├── REFUGE2/                      (fundus optic disc, 400 images)
-│   ├── train/                    {images/, mask/}
-│   ├── val/                      {images/, mask/}
-│   └── test/                     {images/, mask/}
+├── REFUGE2/                      (fundus optic disc, 1200 images = 400 train + 400 val + 400 test)
+│   ├── train/                    {images/, mask/}   400 images
+│   ├── val/                      {images/, mask/}   400 images
+│   └── test/                     {images/, mask/}   400 images
 │
 └── ISIC2018/                     (dermoscopy lesions, 2594 train + 100 val + 1000 test)
     ├── train/                    {images/, masks/}
@@ -182,9 +204,9 @@ Approximate total size: ~18 GB.
 |---|---|
 | **Modality** | Fundus photography |
 | **Original task** | Optic disc and cup segmentation |
-| **Images** | 400 (across train/val/test as released) |
+| **Images** | 1200 = 400 train + 400 validation + 400 test (full official challenge release) |
 | **Mask convention** | Multi-class (BG / disc / cup) preserved; for binary disc segmentation, treat any non-background pixel as foreground |
-| **Split type** | Pre-released train/val/test split is preserved |
+| **Split type** | Pre-released **400/400/400 train/val/test** split is preserved |
 | **Leakage risk** | None — each image is from a different patient by protocol. |
 | **Caveat** | Modern segmenters reach ≥99.9 Dice on optic-disc segmentation; this dataset is **saturated** for that task. Use only when you specifically need fundus / glaucoma data. |
 | **Source** | [REFUGE2 Challenge](https://refuge.grand-challenge.org/) |
@@ -207,6 +229,53 @@ Approximate total size: ~18 GB.
 
 ---
 
+## Comparison to literature and existing HuggingFace cards
+
+We audited the most common split conventions in published segmentation papers
+(MICCAI / IEEE TMI / MIA / CVPR / NeurIPS) and the two existing HuggingFace
+community cards for the same datasets, then aligned our defaults where
+sensible. Summary:
+
+| Dataset | Mainstream literature default | HuggingFace community card | **Our default** | Verdict |
+|---|---|---|---|---|
+| CVC-ClinicDB | Image-level via PraNet's release files (550 train + 62 test pooled with Kvasir into 1450/100) | `Angelou0516/CVC-ClinicDB`: 80/10/10 image-level, ESFPNet split | Image-level 80/20 seed=42; **also a video-level 23/6-sequence split** is shipped | Matches PraNet practice; we additionally fix the same-video leakage that PraNet's image split exposes |
+| Kvasir-SEG | Image-level via PraNet's 900/100 release files | `kowndinya23/Kvasir-SEG`: 880/120 (no test) | Image-level 80/20 seed=42 | Close to mainstream; if you need PraNet-comparable numbers, use 900/100 instead |
+| BUSI | Image-level random; growing minority does 5-fold + de-duplication (BUS-Set, Med Phys 2023, documents duplicate leakage) | n/a | Image-level 80/20 seed=42 | Matches majority; **flag**: BUSI release has documented duplicates, and patient IDs are not public, so true patient-level splits are not possible |
+| ISIC 2018 | Official 2594/100/1000 OR pooled 80/20 | varies | Official 2594/100/1000 preserved | Matches official challenge split |
+| REFUGE2 | Official 400/400/400 (train/val/test domain-shift design) | varies | Official train/val/test preserved | Matches official |
+| ACDC | Patient-level; TransUNet 70/10/20 of the 100 train OR nnU-Net 5-fold patient CV | rarely correct on HF | Patient-level 80/20 seed=42 (in `split_info.json`); 3D side **also** ships official 100-train + 50-test challenge layout | Stricter than the careless cards; consistent with TransUNet/nnU-Net practice |
+| BraTS 2020 | Volume-level; nnU-Net 5-fold patient CV is the most-cited recipe | rarely correct on HF | Volume-level 80/20 seed=42 (295/74) | Matches the careful camp; nnU-Net's 5-fold is a reasonable alternative on the same volumes |
+
+**Mainstream papers we cross-checked**: PraNet (Fan et al., MICCAI 2020),
+Polyp-PVT (Dong et al., 2021), ESFPNet (Chang et al., 2024), BUS-Set
+(Thomas et al., Med Phys 2023), TransUNet (Chen et al., 2021), SwinUNet
+(Cao et al., 2022), nnU-Net (Isensee et al., Nat. Methods 2021).
+
+### Notable disagreements with HuggingFace community cards
+
+- `kowndinya23/Kvasir-SEG` (880/120) merges the test fold into validation,
+  making it **non-comparable to PraNet's 900/100**. Ours preserves
+  test/val separation.
+- `Angelou0516/CVC-ClinicDB` does image-level 80/10/10 without flagging
+  the same-video frame leakage that affects all 3 splits. We add an
+  explicit video-level split for leakage-free evaluation.
+- Neither HuggingFace card we found exposes patient-level splits for
+  ACDC or BraTS — we provide them via `split_info.json`.
+
+### When to *not* use our defaults
+
+- If you must **directly compare to PraNet/Polyp-PVT** numbers, use their
+  released 1450/test files (not in this bundle, but reproducible from the
+  raw images here).
+- If you need **nnU-Net 5-fold CV** on ACDC or BraTS, regenerate folds
+  with the standard nnU-Net recipe — our 80/20 split is a single-fold
+  approximation.
+- If you need **BraTS 2021 (1251 volumes)** instead of 2020 (369), the
+  3D version is shipped under `BraTS2021_3D/` (subset of 2020 patients
+  is included; new 2021-specific patients are added).
+
+---
+
 ## Recommended use
 
 **For paired-comparison evaluation across methods**, lock to the canonical
@@ -236,6 +305,25 @@ info = json.load(open(os.path.join(ROOT, "CVC-ClinicDB", "video_split_seed42.jso
 train_seqs = set(info["train_sequences"])
 ```
 
+For the **3D NIfTI versions**:
+
+```python
+import nibabel as nib
+import os
+
+# ACDC 3D (original challenge layout, 100 train + 50 test patients)
+patient_dir = os.path.join(ROOT, "ACDC", "3D", "training", "patient001")
+img = nib.load(os.path.join(patient_dir, "patient001_frame01.nii.gz")).get_fdata()
+gt  = nib.load(os.path.join(patient_dir, "patient001_frame01_gt.nii.gz")).get_fdata()
+# img shape: (H, W, num_short_axis_slices); gt has 4 classes (0=BG, 1=RV, 2=Myo, 3=LV)
+
+# BraTS 2021 3D (1251 patients, 4 modalities + GT each)
+pat = os.path.join(ROOT, "BraTS2021_3D", "BraTS2021_00000")
+flair = nib.load(os.path.join(pat, "BraTS2021_00000_flair.nii.gz")).get_fdata()
+seg   = nib.load(os.path.join(pat, "BraTS2021_00000_seg.nii.gz")).get_fdata()
+# seg has 4 classes (0=BG, 1=necrotic, 2=edema, 4=enhancing); whole-tumor = (seg > 0)
+```
+
 For datasets without a `split_info.json`, the canonical image-level split
 is reproducible from `seed=42`: