- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
This paper studies the problem of predicting pixel-level multiplex spatial protein expression from routine H&E histology images and proposes a conditional diffusion framework that models the full multiplex protein panel jointly. The method performs diffusion directly in the normalized protein-pixel space on 384×384 tiles. At inference, full regions are reconstructed from overlapping tiles combined by Hann-weighted overlap-add. Experiments are conducted on a colorectal cancer CODEX cohort and a multi-tumor CODEX cohort against four baselines, with an additional exploratory downstream evaluation on TCGA-CRC whole-slide images.
Critical strengths
- The studied problem is meaningful. Multiplex spatial proteomics is costly and difficult to scale, whereas H&E is routinely available.
- Applying generative modeling to this domain is interesting. Treating the multiplex protein panel as a single structured generative target is a different perspective from existing deterministic methods such as ROSIE and HEX.
Critical weaknesses
- My main concern is the scalability of both the method and its experimental validation. In practice, H&E whole-slide images are gigapixel-scale, while the model operates on 384×384 tiles (~145 µm), and the reconstruction benchmark is limited to 720×960 regions. Inference is also expensive since each tile requires 50 denoising steps, and each region is covered by heavily overlapping tiles (stride 96, 75% overlap). No runtime or memory cost is reported. Moreover, the TCGA analysis does not demonstrate pixel-level whole-slide prediction.
- The choice of image encoders is not well justified. I wonder if we can directly apply a pathology foundation model (e.g., UNI, CONCH, Virchow). Given that the training cohorts are small (e.g., 24 training subjects for CRC), an encoder trained from scratch is unlikely to generalize well across tissues, stains, and institutions. At a minimum, I suggest an ablation study that replaces the encoders with a pathology foundation model, or a simple pathology-FM-based baseline.
- The ablation shows that removing the global ResNet-50 branch changes raw Pearson by only 0.001, which suggests the current global features contribute little.
- The intuition behind Hann-weighted overlap-add is not introduced. The paper does not explain why it chose this blending scheme over alternatives (e.g., uniform averaging or non-overlapping tiling), and it provides no ablation on stride or window size.
- It seems that the baselines are too weak. ROSIE achieves a variance ratio of 0.001 on CRC (Table 1), which is nearly equivalent to outputting a constant map. HEX obtains a negative Pearson correlation on Multi-Tumor.
Questions for authors
See Weaknesses.
3 comments
We also evaluated frozen pathology foundation model embeddings with PCA/Ridge regression on Multi-Tumor.
Table R9. Frozen pathology-FM linear probes on Multi-Tumor (66 cores, five outer folds).
| Model | Global-raw Pearson |
|---|---|
| CONCH v1 + PCA/Ridge | 0.1507 |
| UNI v1 + PCA/Ridge | 0.1676 |
| Virchow v1 + PCA/Ridge | 0.1676 |
| H&E Feature Ridge | 0.1680 |
| ProDICE | 0.2255 |
These frozen pathology foundation model features perform similarly to H&E Feature Ridge under a linear readout and remain below ProDICE. Together, these experiments show that ProDICE can benefit from stronger pretrained image representations, while its performance is not explained solely by the choice of image encoder.
W3: Removing the global ResNet-50 branch changes raw Pearson by only 0.001.
Response: We agree that the global ResNet-50 branch has little effect on raw Pearson correlation. However, it provides broader morphological context that complements the spatially aligned features from the local encoder. As shown in Table R10, removing this branch reduces region-centered Pearson by and SSIM by , and worsens Wasserstein distance across all three training seeds. These results indicate that the global branch contributes primarily to within-region spatial prediction rather than overall Pearson correlation.
Table R10. Conditioning ablations on CRC validation. Deltas are ablation minus the mixing-enabled reference (App. C) across three training seeds; negative Pearson/SSIM and positive Wasserstein deltas indicate degradation.
| Removed component | raw Pearson | centered Pearson | SSIM | |
|---|---|---|---|---|
| Global morphology tokens | +0.0013 | |||
| Morphology-conditioned protein queries | +0.0015 | |||
| Dense cross-marker mixing | +0.0000 |
W4: Hann-weighted overlap-add, alternatives, and stride or window size.
Response: Thank you for identifying this missing explanation. We use Hann-weighted overlap-add to reduce the contribution of tile boundaries, where convolutional context and padding effects may differ from central regions.
Each tile's prediction is least reliable near its borders, where the denoiser has the least context, and neighboring tiles are sampled independently, so switching abruptly from one tile to the next would leave seams. A Hann window gives each tile's center the most weight and tapers toward its edges (Eq. 7, App. F.4), so each output pixel is dominated by tiles in which it lies near the center and transitions between tiles are smooth.
The alternatives trade off differently. Uniform averaging gives tile-border pixels the same weight as tile centers and averages more tiles per pixel, which suppresses more of the sampled variation. Non-overlapping tiling keeps one sample per pixel but places every tile border, where predictions are weakest, along fixed seams. Stride sets how much blending occurs:
Table R11. Reconstruction weighting and stride on CRC validation (16 regions from 4 subjects; seed 42; 50 DDIM steps; tissue-supported evaluation). All rows reuse the same 35 sampled tiles (the stride-192 rows use their 12-tile subset), so differences reflect reconstruction alone. is the Kish effective number of contributing tiles per pixel, , a property of the weights rather than a count of independent observations. Matrix and Frobenius error are donor-pooled and not comparable with the region-wise values in Tables R1, R3, R5, and R12; ProDICE's Val-16 values in Tables R1 and R8 come from separate sampling and evaluation runs.
| Reconstruction | Stride | Pearson | RMSE | Native 2D SSIM | Variance ratio | Matrix | Matrix Frobenius error | ||
|---|---|---|---|---|---|---|---|---|---|
| Disjoint assignment (one tile per pixel) | 96 | 1.00 | 0.1797 | 0.1819 | 0.2204 | 0.6086 | 0.0299 | 0.8815 | 0.0099 |
| Hann overlap-add (default) | 96 | 4.18 | 0.2534 | 0.1583 | 0.2313 | 0.2807 | 0.0477 | 0.8750 | 0.0136 |
| Uniform averaging | 96 | 8.02 | 0.2827 | 0.1534 | 0.2307 | 0.2169 | 0.0532 | 0.8634 | 0.0179 |
| Hann overlap-add | 192 | 1.63 | 0.2027 | 0.1715 | 0.2240 | 0.4562 | 0.0366 | 0.8821 | 0.0100 |
| Uniform averaging | 192 | 2.69 | 0.2272 | 0.1639 | 0.2231 | 0.3526 | 0.0422 | 0.8802 | 0.0111 |
At stride 96, uniform averaging gives the best pointwise scores (Pearson 0.2827, RMSE 0.1534) but the lowest variance ratio (0.2169) and the worst and correlation-matrix agreement. Disjoint assignment retains the most variance (0.6086) and has the best at the lowest Pearson (0.1797). Hann weighting lies between them, with SSIM nearly tied with uniform averaging and better , matrix , and Frobenius error (0.0477 vs. 0.0532; 0.8750 vs. 0.8634; 0.0136 vs. 0.0179). Moving from stride 96 to 192 reduces blending, raises the Hann variance ratio to 0.456, lowers Pearson to 0.203, and cuts sampling time (Table R7). Independently resampled stride-192 tiles give nearly identical values (Hann: Pearson 0.202, variance ratio 0.458; uniform: 0.227 and 0.354). Stride therefore trades pointwise accuracy against variance and cost. Disjoint assignment still samples all 35 tiles, so it is not the cheaper non-overlapping tiling (six tiles at stride 384, about 23.6 s per region), whose output quality we did not evaluate. In the tested Hann configurations, we observed no MAE increase in an eight-pixel band around tile seams. The window size is tied to the training tile, so we varied weighting, stride, and steps rather than retraining at other tile sizes. We have added this explanation and the stride study to App. F.4 (Table 8).
W5: Baselines appear too weak.
Response: We agree that the weak results warrant careful examination and that performance under one adaptation protocol should not be interpreted as each method's best attainable performance. ROSIE and HEX were both fine-tuned on our training splits and kept their native footprints and reconstruction (Sec. 4.3; App. G.2), and every method's reconstructed full-resolution maps were scored by the same evaluator. Checking these baselines, we found that the full-canvas ROSIE and HEX inference behind Table 1 used inputs that differ from their training preprocessing (response to Reviewer 1qce, Q4), which accounts for ROSIE's variance ratio of 0.001 and HEX's negative Multi-Tumor Pearson. With training-matched inference on a common tissue support, ROSIE is essentially tied with ProDICE on CRC Pearson (0.204 vs. 0.209) and Spearman, while ProDICE retains lower (0.045 vs. 0.083) and a higher variance ratio (0.276 vs. at most 0.082). The revised Table 1 reports both protocols (Table R6), and we also corrected an export inconsistency in the supplementary per-marker tables.
Because performance under one adaptation protocol should not be read as a method's best attainable performance, we also added three stronger comparisons, all on the same 28 held-out CRC test regions:
- Matched deterministic U-Net (Table R1, response to Reviewer Qq5x), matched to ProDICE in conditioning, output channels, tile reconstruction, and evaluation pixels. It beats ProDICE on Pearson (0.3077 vs. 0.2262) and RMSE (0.1441 vs. 0.1558), while ProDICE retains more variance (0.3216 vs. 0.1833), has lower (0.0390 vs. 0.0539), and better reproduces the inter-marker correlation matrix (MAE 0.1200 vs. 0.2976, lower in all 28 regions). The same pattern holds across three training seeds and on Multi-Tumor (Table R12), and adding variance-matched noise to the U-Net does not close the gap without losing most of its spatial fidelity (Table R5, response to Reviewer 1qce). The two models also differ in training objective and number of updates (the U-Net received about half as many), but averaging eight ProDICE draws reaches the U-Net's pointwise scores on a six-region subset (Table R4, response to Reviewer 1qce), so the pointwise gap is consistent with sampling versus point estimation rather than a weaker network.
- STFlow, a recent flow-matching method, has higher Pearson (0.2493 vs. 0.2262) and lower RMSE (0.1459 vs. 0.1558) than ProDICE, but compresses variance much more (variance ratio 0.0842 vs. 0.3216) and has higher (0.0699 vs. 0.0390; Table R3, response to Reviewer 1qce).
- DiffVS (Oh et al., AAAI 2026), a marker-wise conditioned diffusion method for virtual multiplex staining and a more directly relevant generative comparator than ROSIE or HEX. ProDICE has higher Pearson (0.2262 vs. 0.1699), lower (0.0390 vs. 0.1100), a higher variance ratio (0.3216 vs. 0.1640), and closer agreement with the measured inter-marker correlation matrices ( = 0.7476 vs. 0.4565; Table R3).
We have added these comparisons, the training-matched ROSIE and HEX results, and the matched U-Net results to the revision (Table 1; Apps. G.3, H, and I) and updated our claims accordingly.
We hope these additions address the reviewer's concerns and are happy to provide further details.
Scalability, Conditioning, Reconstruction, and Stronger Baselines
Response to Reviewer a64o
We thank the reviewer for the constructive feedback and for recognizing the importance of virtual spatial proteomics and joint generative modeling. To directly address the concerns, we conducted additional experiments on (W1) computational cost and scalability, (W2) pathology foundation model encoders, (W3) global morphology conditioning, (W4) reconstruction strategies and stride selection, and (W5) stronger deterministic and generative baselines. We present the results below. Tables are numbered consistently across our responses to all reviewers.
Response: We agree that the submission should have reported cost.
Table R7. Sampling time per region under the CRC validation reconstruction protocol (one NVIDIA L4, batch size 1; ± SD where available).
Runtime is proportional to the number of denoiser evaluations (tiles × steps), so ProDICE's extra cost over a deterministic network comes from its sampling steps rather than from a heavier network. A separate repeated pure-inference timing of the default setting gave 137.51±0.04 s per region, consistent with Table R7, and peak allocated GPU memory was about 779 MiB at 50 steps and batch size 1. The cheaper settings change the output: halving the steps lowers the variance ratio slightly (0.2807→0.2622) with no detected change in pointwise metrics or SSIM. Doubling the stride reduces tile blending, which raises the variance ratio to about 0.46 and lowers Pearson to about 0.20 (W4). Dense pixel-level whole-slide inference therefore remains expensive with the current sampler, although tiles are independent and parallelize across GPUs, and step distillation could reduce the number of sampling steps further.
The TCGA analyses ran tiled inference at cohort scale, with the earlier mixing-enabled checkpoint in the paper (710,812 tiles from 388 slides; Sec. 4.5) and since with the final model (49,205 tiles), but in both each tile was reduced to its mean predicted intensities. They show that cohort-scale tiled inference and feature extraction are feasible. We have added the cost analysis to App. H ("Runtime, stride, and sampling steps") and the Limitations.
Response: Thank you for the suggestion. We agree that pathology-specific pretrained representations may generalize better than generic ImageNet features, particularly with limited paired H&E–CODEX data.
One clarification: the global encoder is not trained from scratch. It is a frozen ImageNet-pretrained ResNet-50 with a trainable projection; the only image encoder trained from scratch is the small local convolutional encoder.
We tested the suggested substitution, replacing the frozen ResNet-50 with frozen UNI, CONCH, or Virchow while keeping the rest of the model fixed:
Table R8. Global-encoder substitutions on CRC validation (16 regions, 4 subjects; seed 42; same training schedule and epoch-69 EMA checkpoint for every row).
All three pathology foundation models improve validation Pearson over ResNet-50, with CONCH achieving the largest gain (+0.0305) and UNI the highest SSIM. We originally selected ResNet-50 as a simple, established pretrained encoder to focus on evaluating the proposed joint diffusion framework rather than optimizing the image backbone. The additional results show that pathology-specific encoders can further improve prediction accuracy without changing ProDICE's core architecture.