数字病理切片处理与研究

references/data_management.md

This reference targets PathML 3.0.8 stable and a local, de-identified research workflow.

Data boundaries

Separate four classes of data:

1. Source slides — immutable, access-controlled originals.
2. Linkage data — direct identifiers and the pseudonym mapping, held outside the analysis workspace by an authorized custodian.
3. Analysis data — pseudonymous manifests, tiles, masks, counts, graphs, and features.
4. Reports/models — potentially identifying derived artifacts that still require governance.

Do not assume derived images or embeddings are anonymous. Rare morphology, scanner metadata, dates, or cohort combinations can re-identify a participant. Apply the minimum-necessary principle and institutional retention policy.

Manifest first

Use one row per slide. Recommended columns:

slide_id,patient_id,specimen_id,path,split,stain,backend,site,scanner

Rules:

• IDs are pseudonyms, not MRNs, accessions, initials, dates, or names.
• slide_id is unique.
• one patient_id maps to exactly one split;
• one slide path maps to one slide ID;
• paths are local, relative to a declared root where possible;
• URLs and symlinks are rejected;
• split values are a fixed allowlist such as train, validation, test;
• serial sections, rescans, and multiple blocks from one patient remain together.

Validate before PathML:

python scripts/slide_manifest.py validate \
  --manifest metadata/manifest.csv \
  --root .

The validator checks strict CSV structure, duplicate IDs/paths, missing local files, unsafe paths, supported suffixes, and patient/slide leakage. It does not upload data or inspect arbitrary clinical fields. If the split column is present, every row must supply it. Without that column, the report explicitly leaves patient split isolation unchecked.

h5path format

PathML processes slides into an HDF5-based .h5path file. Stable documentation describes:

root/
├── fields/
│   ├── labels/          # slide-level attributes
│   └── slide_type/      # stain/platform flags
├── masks/               # slide-level masks
├── counts/              # AnnData-like counts storage
└── tiles/
    ├── attributes       # tile_shape, tile_stride
    └── "(i, j)"/
        ├── array
        ├── masks/
        ├── labels/
        └── attributes   # coords, name

Write and reopen through public APIs:

from pathml.core import SlideData

slide.write("derived/slide-001.h5path")
reopened = SlideData("derived/slide-001.h5path")

There is no stable to_hdf5(), from_hdf5(), or load_tiles_from_hdf5() API. SlideDataset.write(directory, filenames=None) calls each slide's write().

Released h5pathManager.add_tile() explicitly casts both tile images and tile masks to float16. Integers through 2048 are exact, but 2049 rounds to 2048 and large values can overflow beyond 65504. Thus distinct instance IDs can merge; intensity values can also lose precision. This is a storage limitation, even when Bio-Formats read with normalize=False.

Preserve authoritative quantitative images and integer instance maps in a separate lossless, typed format. Check actual label identity and dtype after a small h5path round trip before downstream counting. Binary masks are not subject to the same instance-ID loss. Record tolerances for float measurements, but require exact equality for categorical labels.

h5path trust boundary

Treat .h5path as a structured binary input, not harmless data:

• HDF5 parsers have a large attack surface; open third-party files in isolation.
• PathML 3.0.8 TileDataset and core h5path/tile code dynamically interpret the stored tile_shape attribute as a Python expression. Never open an untrusted .h5path, including through SlideData.
• Labels can contain sensitive values. Do not copy direct identifiers into HDF5.
• A malformed file can request large allocations. Check file size and schema before loading.
• Do not edit HDF5 concurrently from multiple processes unless the access pattern is explicitly designed and tested.

Use a sidecar JSON manifest for provenance rather than relying on arbitrary HDF5 labels. Keep the JSON strict, bounded, pseudonymous, and versioned.

PyTorch tile dataset

The canonical stable import is:

from pathml.datasets import TileDataset
from torch.utils.data import DataLoader

tiles = TileDataset("derived/slide-001.h5path")
loader = DataLoader(
    tiles,
    batch_size=8,
    shuffle=False,
    num_workers=0,
)

Each item is:

(tile_image, tile_masks, tile_labels, slide_labels)

Shapes:

• RGB/multichannel 3-D input becomes (C, H, W).
• 5-D PathML input (i, j, z, c, t) becomes (T, C, Z, W, H) in stable source; verify axis semantics before use.
• masks are stacked as (n_masks, tile_height, tile_width) when present.
• values returned by the dataset are NumPy arrays; the PyTorch loader collates them to tensors. Missing masks are None, which default collation cannot handle; use a custom collate_fn for missing masks or nonstandard labels.

TileDataset does not return coordinates explicitly. Keep a parallel mapping from sample index to its stored tile key/coords (HDF5 iteration is not a numeric row/column sort), or include validated coordinate metadata in a custom dataset.

Do not assume mask dictionary order carries semantics. Persist ordered mask names in a separate schema and assert them when loading.

pathml.ml.TileDataset is also exported in 3.0.8, but pathml.datasets.TileDataset is the documented dataset API.

SlideDataset

SlideDataset(slides) accepts a list of already constructed SlideData objects:

from pathml.core import HESlide, SlideDataset

slides = [
    HESlide("data/slide-001.svs", backend="openslide"),
    HESlide("data/slide-002.svs", backend="openslide"),
]
cohort = SlideDataset(slides)
cohort.run(pipeline, distributed=False, tile_size=512, level=0)
cohort.write("derived")

It does not accept a glob/path list plus tiling arguments as a constructor. Preserve a deterministic manifest order and map output filenames explicitly.

Public data modules

Stable pathml.datasets exports:

from pathml.datasets import DeepFocusDataModule, PanNukeDataModule

PanNuke

pannuke = PanNukeDataModule(
    data_dir="approved_data/pannuke",
    download=False,
    shuffle=True,
    nucleus_type_labels=True,
    split=1,
    batch_size=8,
    hovernet_preprocess=True,
)

• 7,901 256-pixel patches, 19 tissue types, five nucleus categories plus background.
• download=False is the safe default.
• download=True downloads three ZIPs from https://warwick.ac.uk/fac/cross_fac/tia/data/pannuke/fold_{1,2,3}.zip and extracts them. These are binary GETs, without authentication or pagination.
• split must be 1, 2, 3, or None; each integer rotates the three published folds across train/validation/test.
• split=None exposes the whole dataset; do not use it for performance estimation.

Published folds are not a substitute for verifying patient/source-slide independence for the intended claim.

DeepFocus

deepfocus = DeepFocusDataModule(
    data_dir="approved_data/deepfocus",
    download=False,
    shuffle=True,
    batch_size=8,
)

• focus classification patches derived from four slides/patients and four stains;
• download=True GETs https://zenodo.org/record/1134848/files/outoffocus2017_patches5Classification.h5 without authentication/pagination;
• the integrity method checks a fixed MD5 before reusing a local file, but the download method does not perform a second post-download check. Verify the completed file independently before use.

MD5 here is an upstream integrity check, not a modern provenance guarantee. Record a SHA-256 and dataset license/source separately.

PathML 3.0.8 does not export TCGADataModule. Use a separately governed data acquisition process for TCGA/GDC and document its API/version/consent terms.

Download consent

Before changing any download flag to True, tell the user:

• exact host and expected dataset;
• approximate transfer and expanded sizes (review-time HEAD responses for the three PanNuke ZIPs total 2,077,087,715 bytes; processing/expansion needs much more disk, with upstream docs estimating ~37.33 GB; the DeepFocus HDF5 is 10,027,826,144 bytes);
• destination and available disk;
• dataset license/terms and citation;
• whether the environment logs outbound IP/account metadata; and
• that no local slide or clinical data will be uploaded.

Require explicit opt-in. Never place downloaded archives inside the repository.

Review-time HEAD checks reached all four source URLs; DeepFocus redirected from /record/ to /records/. No bodies were downloaded, so this establishes endpoint availability, not content integrity, license acceptance, or working extraction. The common download helper skips any existing file by name, including a partial one; independently verify size/checksum before treating such a file as complete.

Graph datasets and unsafe .pt files

pathml.datasets.EntityDataset assembles cell graphs, tissue graphs, and assignment matrices. Stable source opens .pt files using PyTorch object deserialization with unrestricted object loading.

Consequences:

• only load artifacts created by the trusted project;
• never load an emailed/downloaded .pt file merely to inspect it;
• verify SHA-256, producer, code revision, PyTorch/PyG versions, and schema;
• prefer non-executable interchange formats for exchange;
• run legacy artifacts in a disposable, network-disabled environment if review is unavoidable.

The bundled inference planner and graph validator never load .pt, .pth, .ckpt, pickle, ONNX, or other model/graph binaries.

Split design and leakage

Create the split column once, before tiling:

patient → specimen/block → slide/rescan/serial section → region → tile

Everything below a patient follows the patient's split unless the scientific design explicitly requires a stricter grouping.

Common leakage paths:

• overlapping tiles from one slide in different splits;
• serial sections or rescans assigned separately;
• stain reference fitted on all slides;
• QC threshold chosen after viewing test failures;
• normalization/scaling fit before split;
• graph neighborhoods crossing a split boundary;
• duplicated public patches;
• institution/scanner confounding;
• selecting a checkpoint on the test metric.

The manifest validator reports patient and slide leakage, but it cannot discover unknown biological relatedness. Document grouping assumptions.

Provenance sidecar

Recommended strict JSON fields:

{
  "schema_version": "1.0",
  "pathml_version": "3.0.8",
  "source_sha256": "hex-digest",
  "slide_id": "slide-001",
  "patient_id": "patient-001",
  "split": "train",
  "backend": "openslide",
  "level": 0,
  "downsample": 1.0,
  "mpp_x": null,
  "mpp_y": null,
  "tile_size_ij": [512, 512],
  "tile_stride_ij": [512, 512],
  "tile_pad": false,
  "pipeline_id": "he-v1",
  "code_revision": "project-commit",
  "created_utc": "RFC3339 timestamp"
}

Do not put a direct identifier in these fields. Add:

• ordered transform parameters and fitted stain arrays;
• mask/label schema;
• QC counts and exclusion reasons;
• dependency lock hash;
• model artifact SHA-256 and license;
• random seed manifest;
• coordinate units and conversion;
• output hashes and software/hardware details.

Use SHA-256 for provenance:

import hashlib
from pathlib import Path

def sha256_file(path: Path, chunk_bytes: int = 1024 * 1024) -> str:
    digest = hashlib.sha256()
    with path.open("rb") as handle:
        for chunk in iter(lambda: handle.read(chunk_bytes), b""):
            digest.update(chunk)
    return digest.hexdigest()

Hash only authorized local files and expect full-slide hashing to be I/O-heavy. Do not print paths containing identifiers.

Storage and lifecycle checklist

• Estimate raw, temporary, .h5path, mask, count, graph, and model storage.
• Write to a same-filesystem temporary destination, validate, then atomically rename where possible.
• Do not overwrite source slides.
• Use private permissions and encrypted storage/backups.
• Verify output counts, shapes, dtypes, coordinates, and hashes.
• Record partial failures and retry policy.
• Test disaster recovery and retention/deletion.
• Do not commit slide data, model binaries, linkage files, or manifests with PHI.

Sources and further reading

API baseline reviewed 2026-10-01 using the released wheel/tag; hosted docs may lag.

• Stable h5path guide: https://pathml.readthedocs.io/en/stable/h5path.html
• Stable datasets guide: https://pathml.readthedocs.io/en/stable/datasets.html
• Stable datasets API: https://pathml.readthedocs.io/en/stable/api_datasets_reference.html
• Stable TileDataset/EntityDataset source: https://github.com/Dana-Farber-AIOS/pathml/blob/v3.0.8/pathml/datasets/datasets.py
• Stable PanNuke source: https://github.com/Dana-Farber-AIOS/pathml/blob/v3.0.8/pathml/datasets/pannuke.py
• Stable DeepFocus source: https://github.com/Dana-Farber-AIOS/pathml/blob/v3.0.8/pathml/datasets/deepfocus.py
• PanNuke extension paper: https://arxiv.org/abs/2003.10778
• DeepFocus paper: https://doi.org/10.1371/journal.pone.0205387