# SOURCES — public gastric H&E sources assessed against 4096x2048 @ 0.25 um/px

Work package: REAL-SOURCES. Date: 2026-09-09. Companion to `INVENTORY.md`.

## 0. The arithmetic that decides everything

A 4096 x 2048 px image at 0.25 um/px is a **1024 um x 512 um** piece of tissue sampled at
8,388,608 pixels. A source can supply one only if it satisfies **both**:

- **Sampling test:** native pixel size <= ~0.25 um/px. Coarser means upscaling.
- **Field test:** the frame covers at least 1024 x 512 um. A smaller frame means the pixels
  exist but the *tissue* does not, and stitching is not an option for a single field.

Almost every published gastric dataset fails the field test, because almost every one of them is
distributed as small patches (80-2776 px). Almost every published *H. pylori* H&E image fails the
sampling test in the opposite direction — camera micrographs at 400x-1000x oversample the optics so
heavily that a 12 MP frame covers only 100-700 um of tissue.

**Only whole-slide images pass both tests.** That is the single most important finding in this package.

---

## 1. The survey

Verified 2026-09-09 by direct API calls (figshare, Zenodo, GDC, Commons) and page fetches.
Licence strings are quoted as published. "Organisms?" means: does the source assert or annotate the
presence of *H. pylori* **organisms** (not merely a gastritis diagnosis)?

| Source | URL / DOI | Licence (as published) | Format | Native px | um/px | Size | Organisms? | IM? | Access today | Passes both tests? |
|---|---|---|---|---|---|---|---|---|---|---|
| **HiESD** | `10.6084/m9.figshare.28919840` · paper `10.1038/s41597-025-05679-1` | **CC BY 4.0** on the figshare *data* record (see §4 caveat) | `.svs` pyramid | up to ~113,593 x 78,923 (level 0) | **0.2458** (verified in `openslide.mpp-x` on our own copies) | 58.65 GB, 104 slides | **No** — lymphoid follicles annotated as an HP *indicator* only | **Yes** — complete and incomplete IM, pixel-coded | direct HTTP, no login, no DUA | **YES** |
| TCGA-STAD diagnostic slides | GDC, `api.gdc.cancer.gov` | **no named licence**; NIH GDS Policy terms | `.svs` | varies | **mixed 20x/40x** — must read each SVS header | 442 open files | No | No | direct HTTP, open access | partly — only the 40x subset, and no redistribution grant |
| GTEx stomach histology | gtexportal.org / NCI BRD | **unverified** — custom NIH GDS terms, no named licence | `.svs` | varies | **unverified** | ~204 stomach samples (v8) | No | No | viewer; bulk needs a request | **unknown — do not assume** |
| **DeepHP** | `10.5281/zenodo.8117177` · paper `10.3390/ijms232314581` | Zenodo metadata `CC-BY-4.0` **but embargoed to 2030-12-31** | images | 2776 x 2080 | 0.16 (20x) | 394,926 images | image-level HP label | No | **blocked** — embargo + DUA + application | No — 2776 < 4096, and unavailable |
| NEU *Helicobacter pylori* dataset (Ibrahim 2024) | `github.com/ailhan-NEU/Helicobacter-Pylori-Dataset` · paper `10.1007/s10278-024-01021-0` | **CC BY-NC-ND 4.0** | JPEG | 1280 x 960 (3 files 2880 x 2048) | not stated; 400x total on Nikon Ni-U + DS-L3 | ~93 MB, 204 images / 50 cases | **image-level HP grade, incl. Sydney density in filenames** | No | direct HTTP | **No** — 3.2x short, and **ND forbids derivatives** |
| IEEE DataPort "The Helicobacter pylori dataset" | `10.21227/ryks-j091` | not stated on the page | zip | 1916 x 1010 | unknown | 1.35 GB | **Yes — point coordinates**, but on **IHC, not H&E** | No | subscription wall | No |
| GasHisSDB | `10.6084/m9.figshare.15066147` | **CC BY 4.0** on figshare; paper says "non-commercial"; Gitee declares none | PNG | 160/120/80 px | not stated (20x) | 4.38 GB, 245,196 patches | No | No | direct HTTP | No — 25.6x short |
| GAGL | `10.5281/zenodo.7032067` | **CC BY 4.0** | PNG | 2400x2880 to 4800x2016 | not stated; **"downsampled WSI"** | ~92 MB, **9 images only** | No | **Yes** (3 IM, 3 atrophy, 3 normal) | direct HTTP | No — downsampled, and no file reaches 4096 x 2048 |
| "Data for HCRF" gastric | `10.17632/thgf23xgy7.2` (Mendeley) | **CC BY 4.0** | `.tiff`/`.png` | 2048 x 2048 | 20x | 700 images | No | No | direct HTTP | No — 2048 < 4096, and 20x |
| GCHTID / HMU-GC-HE-30K | `10.6084/m9.figshare.25954813` | **CC BY 4.0** | zip | 224 x 224 | 20x | 3.25 GB | No | No | direct HTTP | No |
| PatchGastricADC22 | `10.5281/zenodo.6550925` | **CC BY 4.0** | zip | 300 x 300 | 20x | 7.2 GB | No | No | direct HTTP | No |
| GastritisMIL (ScienceDB) | `10.57760/sciencedb.19700` | see §5 | WSI | WSI | 40x | ~218 GB, 800 WSIs | not established | **Yes — IM grade per slide** | see §5 | see §5 |
| HISTAI gastrointestinal | `huggingface.co/datasets/histai/HISTAI-gastrointestinal` | **CC BY-NC 4.0** | WSI | WSI | per-slide | 202 slides / 120 cases | No | No | HF download | passes geometry, but **NC** |
| DigestPath 2019 | `digestpath2019.grand-challenge.org` | "DATABASE USE AGREEMENT" | WSI + regions | ~3000 x 3000 regions | not verified | — | No | No | **DUA wall** | No |
| Human Protein Atlas stomach | `proteinatlas.org` | **CC BY 4.0** | JPEG | not stated | not stated | — | No | No | direct HTTP | No — **IHC/DAB, not H&E**; 1 mm TMA cores |
| Wikimedia Commons gastric set (Nephron, CoRus13, Uthman, Alex_brollo) | `commons.wikimedia.org` | **CC BY-SA 3.0 / 4.0 / 2.0** — ShareAlike | JPEG | 848x640 to 6000x4000 | **not stated by any file** | per-file | **Yes — several files show organisms** | Yes | direct HTTP | **almost never** — see §3 |

### Searches that returned nothing usable

- **Zenodo**, `"Helicobacter pylori" AND (histopatholog* OR histolog* OR "whole slide" OR biopsy)`,
  type=dataset: **8 records total**. Seven are clinical/epidemiological tables with no images; the
  eighth is DeepHP (embargoed). **There is no Zenodo record of H. pylori H&E whole slides.**
- **Kaggle**: no gastritis or *H. pylori* H&E histology set. What exists is gastric *cancer* mirrors
  of GCHTID/TCGA, plus endoscopy sets (Kvasir, GastroVision) which are not histology.
- **"HP-Neu"**: no dataset of this name exists. The `NEU` in the GitHub handle above is *Near East
  University, Nicosia*. The separate Girona/Barcelona HP work (`arXiv:2412.13857`) is
  **immunohistochemistry, not H&E**, and has no public release.
- **Gastric CAMELYON/PANDA equivalent**: none exists. PAIP has never run a gastric task.
- **Paper supplementary figures**: not viable. Published H&E figures showing organisms are typically
  1000-2000 px on the long edge and JPEG-compressed. Upscaling 2-4x would destroy exactly the
  2.5-5 um curved-rod morphology that makes the category what it is.

---

## 2. The H. pylori organism problem — the honest assessment

**No public dataset provides H. pylori organism-level annotation on H&E at a resolution and field of
view that can produce a 4096 x 2048 image at 0.25 um/px.** That is the finding. It is not a gap in
the search; it is the state of the public record on 9 September 2026.

The three ways this fails, one per candidate class:

**(a) Sources that assert organisms but are too small and/or licence-blocked.**
The NEU dataset is the best of these: a pathologist graded HP density case by case, and the grades
survive in the filenames (`B129-19 HP (++)`, `B198-19 HP (+++)`). It is 1280 x 960 — a 3.2x linear
upscale short — and it is **CC BY-NC-ND 4.0**, where the **ND** term forbids distributing the
adapted material that any crop or resample would produce. Either failure alone is disqualifying.
DeepHP would be the strongest source of all (394,926 labelled images, 111k HP-positive) and is
**embargoed until 31 December 2030**.

**(b) Sources that annotate organisms precisely, but not on H&E.**
The IEEE DataPort HP set has actual point coordinates for organisms — on **immunohistochemistry**.
IHC is a different stain with a different appearance; it cannot be submitted as gastric H&E.

**(c) The one source with the right geometry does not assert organisms.**
HiESD is 40x, 0.2458 um/px, CC BY 4.0, whole slides. Its ten annotation classes include
`Chronic Gastritis` and `Chronic atrophic gastritis` but **no H. pylori class**. The dataset
annotates `lymphoid follicles`, which its authors flag as a morphological *indicator* of HP
infection — an indicator is not the organism.

We also tested this empirically rather than assuming it. Package `B1-ORGANISMS`
(`operations/research/sgh-program-20260908/B1_RESULT.md`) scanned **753 mucosal-surface windows**
across 13 HiESD slides with a classical curved-rod detector and graded the top 60 candidates by eye:

- **0 of 60 graded "likely organisms"**; 5 ambiguous, 55 not organisms.
- The top calls were collagen and muscle fibres, capillary walls, basement membrane, elongated
  endothelial/fibroblast/lymphocyte nuclei, surface fibrin and stain precipitate.
- **The negative arm scored like the positive arm** — normal-gland surface windows reached p90 20.8
  against 24.0 for gastritis surfaces, and the single highest-scoring window in the first sweep was
  a negative.
- Tightening the search region from "dim pale" to genuinely bright luminal space — the compartment
  where the organism actually lives — **collapsed the signal to zero on both arms**.

That is an engineering result about a classical detector at 0.2458 um/px, **not** a pathologist's
verification and **not** a claim that the HiESD slides are HP-negative. But it means we have no
positive evidence of organisms on any HiESD field, and we must not label one as HP-positive.

**Consequence for the submission.** We can source, at native resolution and under a clean CC BY 4.0
licence, real gastric H&E fields for:

- intestinal metaplasia,
- chronic / chronic atrophic gastritis,
- co-occurring intestinal metaplasia and gastritis in one field,
- normal gastric mucosa with neither.

We **cannot** source a real field that is 4096 x 2048 at 0.25 um/px and in which a *H. pylori*
organism is verified present. Any real image we label "H. pylori organism with gastritis" would be
labelled on a diagnosis of gastritis, not on an observed organism.

---

## 3. The Wikimedia Commons route, measured rather than assumed

Commons holds the best publicly licensed gastric H&E images that visibly contain organisms —
Nephron (Michael Bonert), CoRus13, Ed Uthman and others, at up to 6000 x 4000 px. **Not one of them
states a pixel size.** So we measured it.

**Method.** Nuclear texture has a near-constant physical size, so the scale-normalised
Laplacian-of-Gaussian response peaks at a sigma proportional to 1/(um per px). We calibrated the
constant on HiESD, whose `openslide.mpp-x` is known to be 0.2458: `K = sigma_peak * mpp = 1.61`,
stable to +/-1.5% across the native image and its 2x and 3x downsamples. Applying `mpp = K / sigma_peak`
to each Commons file gives the table in `candidates/wikimedia-commons/provenance.csv`. The estimator
is honest about its limits: where the peak pinned at the edge of the scale-space range, the row says
"not measurable" rather than guessing.

| File | Native px | Measured um/px | Physical field | Verdict |
|---|---|---|---|---|
| `Normal_gastric_mucosa_low_mag.jpg` | 4272 x 2848 | **0.252** | 1077 x 718 um | **marginal yes** — needs <=5% resample |
| `Gastritis_helicobacter_intermed_mag.jpg` | 2707 x 2442 | 0.287 | 778 x 701 um | no — 1.15x upscale needed |
| `Gastric_intestinal_metaplasia_low_mag.jpg` | 4272 x 2848 | 0.638 | 2727 x 1818 um | no — 2.55x upscale needed |
| `gastritis_hp_high.jpg` | 4272 x 2848 | 0.160 | 684 x 456 um | no — field too small |
| `gastritis_hp_very_high.jpg` | 2848 x 4272 | 0.095 | 271 x 406 um | no — field far too small |
| `normal_gastric_intermed.jpg` | 2848 x 4272 | 0.124 | 353 x 530 um | no — field too small |
| `Chronic_gastritis_intermed_mag.jpg` | 2848 x 4272 | 0.114 | 325 x 487 um | no — field too small |
| `Stomach_with_intestinal_metaplasia_intermed_mag.jpg` | 2848 x 4272 | 0.058 | 165 x 248 um | no — field far too small |
| `Helicobacter_gastritis_crop_extremely_high_mag.jpg` | 2848 x 1899 | 0.039 | 111 x 74 um | no — one cell cluster |
| `hp_gastritis_very_high.jpg`, `hp_gastritis_extremely_high.jpg`, `chronic_gastritis_very_high.jpg`, `gastric_im_very_high.jpg`, `stomach_im_high.jpg` | 4272 x 2848 / 2848 x 4272 | not measurable | — | no |

Visual inspection confirms the numbers. The "very high" and "extremely high" HP frames are oil-immersion
shots so heavily magnified that a 700 px crop contains a handful of cells and visible optical blur —
empty magnification, not extra information. This is the trap in the whole Google-Images approach: the
images that *show* the organisms best are precisely the ones whose field of view is smallest.

**The structural reason.** These are 12 MP DSLR frames (Canon EOS 450D / Rebel XSi / 1100D,
4272 x 2848, 5.196 um sensor pitch). A 12 MP frame at exactly 0.25 um/px would cover 1068 x 712 um —
just enough for one 4096 x 2048 field, and only if the shot happened to be taken at ~0.25 um/px.
Nephron's magnification tiers straddle that value rather than sitting on it, so the coincidence is rare.
Exactly **one of fourteen** files landed in the usable band, and it is a *normal mucosa* image —
the one category HiESD already supplies in quantity.

**And the licence carries an obligation.** Every Commons candidate is **CC BY-SA** (2.0 / 3.0 / 4.0).
ShareAlike is not NoDerivatives — cropping is permitted — but the crop is adapted material that must
itself be released under the same or a compatible licence, with attribution to the named author. If
the challenge asserts any rights over submitted images, a CC BY-SA image creates a conflict that
CC BY 4.0 material does not. **Recommendation: do not put CC BY-SA material in the 70.**

---

## 4. HiESD in detail — capacity, and one licence caveat

**Licence caveat, to be resolved before submission.** The figshare **data record** returns
`{"name": "CC BY", "url": "https://creativecommons.org/licenses/by/4.0/"}` via the API. The
*Scientific Data* **article** page renders a CC BY-NC-ND notice. These are different objects — the
article text and the deposited data — and the data record governs the `.svs` files. Because a
reviewer will ask, **screenshot the figshare licence field on the day of download and keep it with
the submission.**

### Capacity survey across the whole 104-slide collection

The `ESD_40X_annotation_downsample64` masks are present locally **for all 104 slides**, including
slides whose `.svs` we have never downloaded. That let us survey the entire collection's capacity
without any further downloads. For every slide we slid a 4165 x 2083 level-0 window (which
downsamples to exactly 4096 x 2048 at 0.25 um/px) over the mask and counted **disjoint,
non-overlapping** windows meeting each category rule, with any carcinoma label excluding the window
outright.

Script: `hiesd_field_yield.py`. Full per-slide table: `hiesd_field_yield.csv`.

| Category rule | Whole collection (104 slides) | Available on the workbench now (22 slides) | Needed |
|---|---:|---:|---:|
| IM-dominant, gastritis < 10%, no carcinoma | 539 windows / 65 slides | **316 windows / 18 slides** | 14 |
| IM >= 15% **and** gastritis >= 15% in the same window | 569 windows / 78 slides | **275 windows / 20 slides** | 14 |
| Gastritis-dominant, IM < 2%, no carcinoma | 3,428 windows / 104 slides | **951 windows / 22 slides** | 21 |
| Normal gland, IM < 1%, gastritis < 2%, no carcinoma | 731 windows / 57 slides | **403 windows / 15 slides** | 21 |

**Supply is not the constraint.** Even before this package's downloads, the material on hand
exceeded every requirement by more than an order of magnitude. What we bought with the six extra
slides is *specimen diversity* — the IM and IM+gastritis pools roughly doubled, so a 14-image
category can now be drawn from 18-20 distinct slides instead of 12-14.

**A limitation to state plainly:** HiESD is 104 slides from 44 patients, but the release publishes
**no slide-to-patient map**. We can guarantee 70 images from distinct slides; we cannot guarantee
70 images from 70 distinct patients, and must not claim it.

### The resampling is a downsample, and that matters

HiESD level 0 is 0.2458 um/px, and the target is 0.25 um/px. A 4096 x 2048 output at 0.25 um/px
covers 1024 x 512 um, which is **4165 x 2083 level-0 pixels**. So each field is read at 4165 x 2083
and **downsampled by a factor of 1.017** to 4096 x 2048. This is a reduction in pixel count, never an
increase. Nothing is interpolated into existence. Verification fields are in
`candidates/hiesd/` with the read coordinates, the source mpp and the resample factor recorded per file.

---

## 5. Sources still worth pursuing, and what blocks each

| Source | What it would add | Blocker | Effort |
|---|---|---|---|
| **GastritisMIL** (ScienceDB `10.57760/sciencedb.19700`) | 800 gastric WSIs at 40x with slide-level inflammation, activity, atrophy **and IM** grades — the largest gastritis WSI release in existence | data-specific terms and per-file anonymous download not yet confirmed; ~218 GB total | one slide is ~1-2 GB; a single-slide pull would settle it |
| **DeepHP authors** (`lghm.com.br/datasets`) | the only large HP-labelled H&E corpus | embargo to 2030 + DUA + application | write now; will not land before 11 Sep |
| **GAGL authors** (Barmpoutis, Jansen et al., UCLH) | 85 WSIs / 20 patients with 45 IM cases at full resolution | only a 9-image downsampled subset is published | email request; will not land before 11 Sep |
| **GasHisSDB authors** (Chen Li, `[contact omitted]`) | the 600 original larger fields behind the 245k patches, plus calibrated um/px and a slide map | not published | email request; will not land before 11 Sep |
| **TCGA-STAD** | large volume of gastric H&E, some at 40x | no named licence; mixed 20x/40x; NIH GDS terms silent on redistribution | usable as *normal/background* material if the challenge accepts GDS terms |
| **GTEx stomach** | genuinely normal (non-neoplastic) stomach | licence text and um/px both unverified — the licence page is a JS app that returns nothing to a fetch | someone must open it in a browser and read one SVS header |

---

## 6. Recommended composition of the 70

Planned split 21 / 14 / 14 / 21. **Every block below comes from HiESD** — figshare DOI
`10.6084/m9.figshare.28919840`, **CC BY 4.0**, 40x scan, `openslide.mpp-x` 0.2458, cut at
4165 x 2083 level-0 pixels and downsampled 1.017x to 4096 x 2048 at exactly 0.25 um/px.

| # | Challenge category | n | Source and selection rule | Available pool | Honest label status |
|---|---|---:|---|---:|---|
| (i) | **H. pylori organism with gastritis** | 21 | HiESD `Chronic Gastritis` (0,0,255) and/or `Chronic atrophic gastritis` (70,130,180) >= 35% of the field, IM < 2%, no carcinoma label, window centred on the mucosal surface band where organisms would reside; prefer slides that also carry `lymphoid follicles` (0,255,0) nearby, which HiESD's authors flag as an HP indicator | 951 disjoint windows / 22 slides | **CAVEAT REQUIRED.** HiESD asserts gastritis, not organisms. Our own classical detector found no organisms on these slides (0/60 candidates). These must be described as *chronic gastritis of a pattern consistent with H. pylori infection*, with organism status not established by the source. |
| (ii) | **Intestinal metaplasia only** | 14 | HiESD `Complete IM` (0,128,0) + `Incomplete IM` (255,255,0) >= 35%, gastritis < 10%, no carcinoma | 316 / 18 slides | Clean — IM is a pathologist-annotated pixel class in the source. |
| (iii) | **Co-occurring HP gastritis and IM** | 14 | HiESD, IM >= 15% **and** gastritis >= 15% in the same window, no carcinoma | 275 / 20 slides | Same caveat as (i) for the HP half; the IM half and the co-occurrence in one specimen are genuinely annotated. |
| (iv) | **No H. pylori and no IM** | 21 | HiESD `Normal Gland` (138,43,226) >= 35%, IM < 1%, gastritis < 2%, no carcinoma | 403 / 15 slides | Clean as *normal gastric mucosa without IM*. HP-negativity is inferred from the absence of gastritis, not tested. |

**Selection discipline for whoever cuts the final 70:** draw round-robin across slides so no single
slide contributes more than ~4 images to a category; enforce a 300 um minimum centre-to-centre
spacing so fields never overlap; exclude any window within 600 um of a carcinoma label; and apply the
GrandQC quality mask (`ESD_40X_Quality_Assessment_Mask`, 0.9995 um/px) to keep out folds, pen marks,
bubbles and out-of-focus tissue.

### What NOT to include, and why

- **Nothing from GasHisSDB.** 160 px patches; 25.6x upscale.
- **Nothing from the NEU HP dataset.** CC BY-NC-**ND** forbids derivatives, and 1280 x 960 is 3.2x short.
- **Nothing from Wikimedia Commons.** ShareAlike obligation, no stated pixel size, and thirteen of
  fourteen candidates fail the geometry outright.
- **Nothing from GAGL.** Downsampled, and no file even reaches 4096 x 2048 px.
- **No Google Images material.** No provenance, no licence, no calibrated scale.
- **No upscaled image of any kind**, from any source.

### Source statement to submit

> The 70 real images are 4096 x 2048 px fields cut from HiESD — *A Fully Annotated Pathology Slide
> Dataset for Early Gastric Cancer and Precancerous Lesions* (Ge et al., **Scientific Data**, 2025;
> DOI 10.1038/s41597-025-05679-1), deposited on figshare under **CC BY 4.0** at DOI
> 10.6084/m9.figshare.28919840. The slides are H&E-stained gastric endoscopic submucosal dissection
> specimens scanned at 40x objective. Each source slide reports `openslide.mpp-x` = 0.2458 um/px.
> Each field was read at 4165 x 2083 level-0 pixels — the exact extent of 1024 x 512 um — and
> resampled down by a factor of 1.017 to 4096 x 2048 px at 0.25 um/px. No image was upscaled.
> Category assignment follows the pathologist region annotations distributed with the dataset
> (`Annotation_Category_Label_Description.txt`): complete and incomplete intestinal metaplasia,
> chronic gastritis, chronic atrophic gastritis and normal glands. Fields containing any carcinoma
> annotation were excluded. **HiESD does not record Helicobacter pylori organism status; images
> assigned to the H. pylori categories are so assigned on the basis of the annotated chronic
> gastritis pattern, not on observed organisms.**

That last sentence is the one that must not be dropped.