Back

CellDF: Quality-controlled cell matching for whole-slide HE-IHC label transfer

Jang, E.; Huh, Y.-M.

2026-06-24 pathology
10.64898/2026.06.18.733058 bioRxiv
Show abstract

Serial-section immunohistochemistry (IHC) is the largest available source of paired hematoxylin and eosin (HE) and IHC whole slide images, yet it remains underexploited for cell-level supervision: adjacent sections sample non-identical cells, and residual registration error prevents direct assignment of IHC labels to individual HE cells. We present CellDF (Cell Displacement Field), which turns registered serial-section data into pairs of HE cells and their IHC labels by solving cell matching at whole-slide scale and assessing its reliability without ground-truth correspondences. CellDF estimates a locally adaptive residual displacement field through iterated kernel regression over each HE cells K nearest IHC candidates; a sparse-kernel variant keeps it tractable at the cell counts of a whole slide, where pairwise matchers are not. The within-tile distribution of the estimated displacements yields two ground-truth-free statistics, the directional scatter{sigma}{theta} and the between-tile angular deviation |{Delta}{theta}|, that localize matching quality more finely than landmark-based target registration error and drive a two-stage outlier filter that withholds labels where matching is unreliable. On 54 same-section HyReCo pairs,{sigma}{theta} correlates only moderately with landmark error and flags localized restaining damage that global error misses; on 30 four-marker Acrobat serial-section cases, the same statistic flags which IHC marker, if any, lies physically close enough to HE to support cell-level transfer. As a proof of concept, IHC labels transferred through CellDF trained a cell classifier on HE embeddings that generalized to held-out cells within the sample (F1 0.85, AUROC 0.88), establishing serial-section IHC as a usable cell-level labeling resource. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=78 SRC="FIGDIR/small/733058v1_ufig1.gif" ALT="Figure 1"> View larger version (42K): org.highwire.dtl.DTLVardef@a9b3dcorg.highwire.dtl.DTLVardef@15f652corg.highwire.dtl.DTLVardef@1eb3396org.highwire.dtl.DTLVardef@87dda2_HPS_FORMAT_FIGEXP M_FIG C_FIG

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

1
Nature Methods
385 papers in training set
Top 0.3%
19.1%
2
Nature Communications
5641 papers in training set
Top 15%
12.3%
3
Nature Machine Intelligence
70 papers in training set
Top 0.2%
8.1%
4
Cell Reports Methods
165 papers in training set
Top 0.1%
6.9%
5
Nature
645 papers in training set
Top 2%
6.4%
50% of probability mass above
6
Advanced Intelligent Systems
11 papers in training set
Top 0.1%
5.3%
7
eLife
5828 papers in training set
Top 26%
4.5%
8
Modern Pathology
22 papers in training set
Top 0.1%
2.9%
9
Advanced Science
286 papers in training set
Top 3%
2.5%
10
Nature Aging
60 papers in training set
Top 0.9%
1.8%
11
npj Digital Medicine
118 papers in training set
Top 2%
1.5%
12
Cell Reports
1498 papers in training set
Top 22%
1.2%
13
Proceedings of the National Academy of Sciences
2444 papers in training set
Top 33%
1.2%
14
Cancer Discovery
66 papers in training set
Top 1%
1.2%
15
Nature Medicine
125 papers in training set
Top 3%
1.1%
16
iScience
1154 papers in training set
Top 32%
1.0%
17
Light: Science & Applications
16 papers in training set
Top 0.3%
1.0%
18
Scientific Reports
3612 papers in training set
Top 72%
0.9%
19
New Phytologist
346 papers in training set
Top 4%
0.9%
20
PLOS ONE
5266 papers in training set
Top 60%
0.9%
21
ACS Nano
113 papers in training set
Top 2%
0.9%
22
Science Advances
1243 papers in training set
Top 29%
0.9%
23
Cell Reports Medicine
153 papers in training set
Top 5%
0.6%
24
Communications Biology
993 papers in training set
Top 33%
0.6%