Pith. sign in

REVIEW 4 major objections 5 minor 40 references

Multimodal Outer Arithmetic Block Dual Fusion of Whole Slide Images and Omics Data for Precision Oncology

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Reintroducing the same DNA methylation vector at both patch-level and slide-level fusion stages improves CNS tumor subtyping and survival prediction over either stage alone.

desk verdict Plausible dual-fusion architecture; evaluation leak and weak stats mean the headline gains are not yet credible. read the letter →

arxiv 2411.17418 v2 pith:4MXSROLP submitted 2024-11-26 cs.CV

Omnia Alwazzan , Amaya Gallagher-Syed , Thomas O. Millner , Sebastian Brandner , Ioannis Patras , Silvia Marino , Gregory Slabaugh This is my paper · ORCID
classification cs.CV
keywords dualfusionwholeslideimagesDNAmethylationCNStumorclassificationmultipleinstancelearningsurvivalpredictionmultimodaldeepouterarithmeticblock
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper proposes MOAD-FNet, a dual-fusion network that feeds one DNA methylation embedding into a whole-slide image pipeline twice: first concatenated with each patch embedding and mapped by an MLP before gated-attention pooling, and second reintroduced alongside the pooled slide embedding through an outer-arithmetic block. The central claim is that this two-stage reintroduction captures complementary local and global cross-modal interactions, and that single-stage early or late fusion leaves performance on the table. On a 1,504-patient CNS tumor cohort spanning 20 subtypes, ABMIL-MOAD-FNet reaches F1-Macro 0.745, compared with 0.644 for early fusion and 0.690 for late fusion; on the two public cancer cohorts it reaches c-index 0.691 for bladder cancer and 0.726 for breast cancer. If the claim holds, dual fusion is a straightforward design choice for integrating methylation with histology in precision oncology, with attention maps that make the molecular-morphological evidence visible.

What carries the argument

The load-bearing machinery is the MOAD-FNet architecture and its MOAB fusion block. In the early fusion stage, the omic embedding $o_i \in \mathbb{R}^{256}$ is concatenated with each UNI patch embedding $e_{ij} \in \mathbb{R}^{1024}$ to form $z_{ij} = [e_{ij}, o_i]$; an MLP $f_E$ maps each concatenation to a joint per-patch embedding $p_{ij}$, and ABMIL gated attention pools these into a slide embedding $v_i \in \mathbb{R}^{256}$. In the late fusion stage, MOAB takes $v_i$ and $o_i$, appends constants, and computes outer product, outer division, outer subtraction, and outer addition, producing four $257 \times 257$ interaction matrices that are concatenated along the channel dimension and condensed by a 2D convolution before the classifier. The appended constants preserve the original unimodal features inside the interaction matrices, which is why the block can richly intermingle modalities without losing either one.

What would settle it

Run the entire MOAD-FNet pipeline under nested cross-validation, selecting the CpG feature count (4K, 8K, or 10K) inside each training fold only, and compare ABMIL-MOAD-FNet against early and late fusion on the brain-tumor subtyping dataset; if the F1-Macro gap (0.745 versus 0.644 and 0.690) shrinks or reverses, the dual-fusion benefit is partly a feature-selection artifact.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that the omic vector should not be used only once. The authors show that when the same methylation embedding is projected onto every whole-slide patch in latent space and then reintroduced at the slide level through the Multimodal Outer Arithmetic Block, the resulting dual-fusion model outperforms both early-only and late-only variants on fine-grained CNS tumor subtyping, and improves or matches the strongest compared methods on survival prediction. They interpret this as evidence that local patch-level interactions and global slide-level interactions carry complementary information, and that an outer-product family of arithmetic operations captures cross-modal correlations more richly than concatenation or Kronecker-product fusion.

Load-bearing premise

The reported gains assume the 8K CpG methylation feature set was chosen without using the held-out test folds; the paper compares 4K, 8K, and 10K sites before describing the split and does not mention nested cross-validation, so if test data influenced the feature count, the F1 and c-index improvements are optimistic.

Editorial extensions

If this is right

  • ABMIL-MOAD-FNet reports the best subtyping result on the 20-subtype CNS tumor dataset, with F1-Macro 0.745, beating early fusion by 0.101 and late fusion by 0.055 in the ablation.
  • Dual fusion also improves survival prediction, reaching c-index 0.691 on the bladder cancer cohort, the best among compared methods, and 0.726 on the breast cancer cohort, on par with the strongest baseline.
  • The MOAB aggregation block consistently outperforms concatenation and Kronecker-product fusion in the survival ablation across ABMIL and TransMIL backbones, under both late and dual fusion settings.
  • The improvement is attributed to the fusion design rather than the WSI encoder: swapping the UNI encoder for ConvNeXt only lowers F1-Macro from 0.745 to 0.732.
  • Attention scores from the early fusion stage can be visualized as heatmaps that highlight diagnostically relevant patches, linking molecular signals to morphology.
  • The same dual-fusion scheme also improves survival prediction on the two public cohorts, with c-index 0.691 on bladder cancer and 0.726 on breast cancer.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: if the dual-fusion advantage is robust, single-stage fusion is systematically underusing the omic modality; a direct test is to repeat the comparison with other omic types such as RNA expression or copy-number data, and on other tumor sites.
  • Editorial inference: the four outer arithmetic operations form a parameter-light bilinear interaction layer; one can probe whether the gain comes from true cross-modal correlation by replacing the methylation embedding with a permuted or zeroed vector and checking that the dual-fusion gap disappears.
  • Editorial inference: the paper's Limitations section notes that the identity of decisive CpG sites is hard to recover after early fusion; an attribution analysis on the early-fusion MLP could recover per-CpG importance and make the dual-fusion approach more actionable clinically.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes MOAD-FNet, a dual-fusion architecture that integrates whole-slide images (WSIs) with DNA methylation or pathway-level omics data. It performs early fusion by concatenating an omic embedding to each WSI patch embedding and processing the result with an MLP, followed by ABMIL attention pooling; it then reintroduces the omic embedding at a late stage through a Multimodal Outer Arithmetic Block (MOAB) that combines outer product, division, addition, and subtraction. The method is evaluated on a private NHNN BRAIN UK CNS tumor dataset with 20 subtypes (reporting F1-Macro 0.745 for the full model vs 0.644 for early-only and 0.690 for late-only fusion) and on TCGA-BLCA and TCGA-BRCA for survival prediction (reporting c-index 0.691 and 0.726, respectively), with ablations comparing fusion stages, backbones, and aggregation methods.

Significance. If the reported gains are leakage-free, the paper makes a useful contribution: it demonstrates that reintroducing omic data at both patch and slide levels can improve multimodal WSI-omics classification and survival prediction, and it provides a concrete fusion block (MOAB) with ablations against concatenation and Kronecker-product alternatives. The study is also strengthened by comparing against several recent baselines (SurvPath, MCAT, TransMIL, PIBD, MMP) on public TCGA cohorts and by including attention heatmap-based interpretability. However, the central comparison between dual and single-stage fusion on the CNS dataset rests on a feature-selection protocol that appears to use the full dataset before the evaluation split, which could inflate the reported advantage. The survival results also have overlapping standard deviations, so the empirical support for the headline claims is currently incomplete.

major comments (4)
  1. [Section IV, subtyping results] The CpG feature selection is described before the 2-fold split and appears to be performed on the full 1,504-patient dataset: the text states that variance, coefficient of variation, median absolute deviation, and inter-quartile range were intersected to select the 8K most variable CpG sites, and that 4K and 10K alternatives were compared on performance before selecting 8K. No nested cross-validation or outer-loop feature selection is described. If the test fold informed the choice of feature subset or its size, the reported F1-Macro values in Tables I and II are optimistically biased. Because all compared fusion variants share the same CpG features, this bias could either inflate or mask the dual-fusion advantage, and the main claim of the paper is therefore not yet supported by a leakage-free protocol. The authors should either describe a nested CV procedure, fix the feature set a priori based on prior work, or re-run the central ablations with feature selection confined to training folds.
  2. [Table III and Table IV] The Wilcoxon rank-sum test reported as p=0.043 is under-specified. It is not stated what the units of comparison are (number of folds, number of runs, per-class F1 values, or per-patient predictions), how many such units were used, or whether the comparison against 'all other multimodal models' involved multiple tests and any correction. With only two cross-validation folds, a rank-sum test would have very low resolution. As written, the sentence 'the test yielded a p-value of 0.043, indicating a statistically significant difference' does not establish significance in a defensible way. The authors should specify the test design, sample size, and correction, or refrain from claiming statistical significance.
  3. [Eq. (1)-(2)] The survival-prediction improvements over baselines are generally within one standard deviation of the baseline values. For example, ABMIL-MOAD-FNet on BLCA achieves 0.691±0.069, while SNN achieves 0.671±0.058 and MoME 0.686±0.041; on BRCA the proposed method's 0.726±0.049 is lower than PIBD-MOAB's 0.749±0.062. The text states 'improved survival prediction on TCGA-BLCA and competitive performance on TCGA-BRCA,' but without statistical tests or a clear demonstration that the differences are not noise, the survival claim is not yet supported. The authors should report confidence intervals or paired significance tests across their five runs, or soften the claim accordingly.
  4. [Eq. (1)-(2)] The early fusion definition has a likely error or redundancy: Eq. (1) defines z_ij = [e_ij, o_i], and Eq. (2) then applies f_E([z_ij, o_i]), which concatenates o_i a second time. If this is a typo, the intended input should be f_E(z_ij) or f_E([e_ij, o_i]); if o_i is intentionally duplicated, the text should explain why. As written, the formulation is ambiguous and affects reproducibility of the central early-fusion module.
minor comments (5)
  1. [Section II-B] The text says the method is aimed at 'survival prediction in both lung and breast cancer,' but the datasets are TCGA-BLCA (bladder urothelial carcinoma) and TCGA-BRCA (breast invasive carcinoma). This should be corrected to bladder and breast cancer.
  2. [Table III] There are two rows both labeled 'MMP [22]' with different c-index values (0.628 and 0.635 for BLCA). The authors should distinguish the re-implemented result from the originally reported value, for example by adding an asterisk or a footnote.
  3. [Table IV caption] The caption reads 'TGGA BRCA AND BLCA DATASETS'; this should be 'TCGA.'
  4. [Table II description] The ablation table lists 'ABMIL Early fusion with p_ij' and 'ABMIL - MOAB Late fusion with e_ij,' but the rows do not specify whether the late-fusion row uses MOAB on e_ij-derived slide embeddings or some other aggregation. A brief clarifying note would help readers map the ablations to the model description in Section II-B.
  5. [Section IV, subtyping results] The text says ConvNeXt achieves F1-Macro 0.732±0.012 while Table II reports 0.732±0.012 for 'MOAD-FNet ConvNeXt encoder'; the text says this is 'marginally lower than the pretrained UNI encoder,' but the difference is small relative to the reported standard deviations. This point is not a blocker, but the phrasing overstates the contrast.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the dual-fusion advantage is an empirical ablation result, not an identity or a fitted parameter renamed as prediction.

full rationale

MOAD-FNet's central claim is architectural: reintroducing omic embeddings at both early and late fusion stages improves CNS tumor subtyping and survival prediction. The paper's derivation chain is a constructive pipeline—Eqs. (1)-(7) define concatenation, MLP mapping, gated attention, and the MOAB outer-arithmetic fusion—and the reported F1-Macro and c-index values are empirical outcomes of the ablations in Tables I-IV, not quantities defined to equal the inputs. No equation in the paper makes the dual-fusion result equivalent to the omic input or to the early-fusion embedding by construction. The only self-citation is to the authors' earlier ISBI paper [28] for the MOAB block; the block is re-evaluated here as an ablation component (Tables II-IV) and was externally published before this work, so the citation is not load-bearing in the sense of forcing the present result. The feature-selection passage in Section II-A (choosing 8K CpG sites after comparing 4K and 10K) is a potential evaluation-leakage concern because the 2-fold split is described afterward, but it is not circular: all compared fusion variants share the same selected CpG features, and the dual-versus-single fusion gap is not forced by the feature count. The Limitations passage about difficulty attributing outcomes to specific CpG sites is an interpretability caveat, not a circularity. Overall, the paper's predictions are empirical comparisons over held-out folds, and the central claim does not reduce to its own inputs.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central claim rests on modality complementarity, frozen foundation-model features, and a feature-selection protocol whose leakage status is unclear. No new physical entities are introduced; the main hidden cost is dataset-dependent selection of CpG features and embedding dimensions.

free parameters (4)
  • CpG feature subset size (K=8,000) = 8,000
    Chosen by comparing 4K, 8K, and 10K variants on the NHNN dataset; 8K selected because 4K reduced performance and 10K showed no improvement. This is a dataset-dependent model selection and may leak if not nested in the 2-fold CV.
  • Omic embedding dimension (d_o) = 256
    Set for the SNN encoder; follows prior work [13] rather than derived from a principle.
  • Slide-level embedding dimension (v_i) = 256
    Aligned to d_o to enable MOAB late fusion; no analysis justifying this specific dimension.
  • MOAB numerical epsilon = 1e-10
    Small constant in outer division Eq. (6) to avoid division by zero; not a substantive free parameter.
assumptions (4)
  • domain assumption DNA methylation and WSI morphology provide complementary, non-redundant information for CNS tumor classification.
    The dual fusion premise; supported broadly by cited clinical literature but not independently established in this paper.
  • domain assumption The UNI foundation model's patch embeddings are suitable for NHNN BRAIN UK subtypes without fine-tuning.
    Section II-B2 uses UNI as a fixed feature extractor; no validation that its pretraining covers these rare subtypes.
  • domain assumption Attention weights from ABMIL on early-fused embeddings mark diagnostically relevant tissue regions.
    Interpretability claims in Fig. 3 assume a correspondence between attention and pathology; no pathologist-reader study is provided.
  • standard math The discrete hazard survival formulation and NLL loss (Eq. 8) are correctly implemented.
    Rests on prior survival analysis literature; the printed formula has a suspicious term log(fsurv(yi) - 1[yi]) that is not defended.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Multimodal Outer Arithmetic Block Dual Fusion of Whole Slide Images and Omics Data for Precision Oncology." pith.science (2026). https://pith.science/paper/4MXSROLP

@misc{pith2026241117418,
  author       = {Pith},
  title        = {Pith review of: Multimodal Outer Arithmetic Block Dual Fusion of Whole Slide Images and Omics Data for Precision Oncology},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4MXSROLP}},
  note         = {Machine review of arXiv:2411.17418}
}
read the original abstract

The integration of DNA methylation data with a Whole Slide Image (WSI) offers significant potential for enhancing the diagnostic precision of central nervous system (CNS) tumor classification in neuropathology. While existing approaches typically integrate encoded omic data with histology at either an early or late fusion stage, the potential of reintroducing omic data through dual fusion remains unexplored. In this paper, we propose the use of omic embeddings during early and late fusion to capture complementary information from local (patch-level) to global (slide-level) interactions, boosting performance through multimodal integration. In the early fusion stage, omic embeddings are projected onto WSI patches in latent-space, which generates embeddings that encapsulate per-patch molecular and morphological insights. This effectively incorporates omic information into the spatial representation of the WSI. These embeddings are then refined with a Multiple Instance Learning gated attention mechanism which attends to diagnostic patches. In the late fusion stage, we reintroduce the omic data by fusing it with slide-level omic-WSI embeddings using a Multimodal Outer Arithmetic Block (MOAB), which richly intermingles features from both modalities, capturing their correlations and complementarity. We demonstrate accurate CNS tumor subtyping across 20 fine-grained subtypes and validate our approach on benchmark datasets, achieving improved survival prediction on TCGA-BLCA and competitive performance on TCGA-BRCA compared to state-of-the-art methods. This dual fusion strategy enhances interpretability and classification performance, highlighting its potential for clinical diagnostics.

Figures

Figures reproduced from arXiv: 2411.17418 by the authors.

Figure 2
Figure 2. Distribution of training and testing data points across 20 classes/subtypes. The bar chart illustrates the number of patients allo￾cated to training and testing sets for each class, highlighting the balance of data used for model development and evaluation. 1) Slide Subtyping - NHNN BRAIN UK Brain Tumor Dataset: We obtained WSIs and matched DNA methylation data from the NHNN, University College London Hospital (UCLH… view at source ↗
Figure 1
Figure 1. Overview of the proposed MOAD-FNet framework. Data engineering and encoding for each modality are performed in the preprocessing block. The early fusion block (top) receives encoded inputs from both modalities, where omic data is concatenated to form a matrix zi,j which is processed by an MLP encoder that learns a joint mapping, resulting in output pi,j . A gated attention via multiple instance learning (MIL) scores… view at source ↗
Figure 3
Figure 3. Visual representation of attention heatmap generated by MOAD-FNet for a diffuse glioma, IDH-mutant and 1p19q-retained (astroglial type) tumor (Class 0). (A) The original histology slide is displayed. (B) The heatmap shows areas of high attention (red) and low attention (blue), with regions of diagnostic relevance highlighted. (C) Representative patches with high attention are bordered in red, potentially indicating … view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Comparison of F1-scores across different fusion methods for glioma classification. The radar chart illustrates the F1-score performance across all 20 classes, highlighting distinct patterns for Early Fusion, MOAD-FNet, and Late Fusion. The bar plots zoom in on specific…
Figure 5
Figure 5. Figure 5: Comparison of confusion matrices and t-SNE visualizations for three fusion strategies: (A) MOAD-FNet, (B) Late Fusion, and (C) Early Fusion for brain tumor subtyping. The corresponding t-SNE plots are labeled as (A.1), (B.1), and (C.1), respectively. strengths of each,…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

40 extracted references · 40 canonical work pages

  1. [1]

    Epigenetic signatures in the diagnosis and prognosis of cancer,

    A. Lopomo and F. Copped `e, “Epigenetic signatures in the diagnosis and prognosis of cancer,” in Epigenetic Mechanisms in Cancer . Elsevier, 2018, pp. 313–343

    work page 2018

  2. [2]

    Integrative multi-omic cancer profiling reveals dna methylation patterns associated with therapeutic vulnerability and cell-of-origin,

    W.-W. Liang, R. J.-H. Lu, R. G. Jayasinghe, S. M. Foltz, E. Porta-Pardo, Y . Geffen, M. C. Wendl, R. Lazcano, I. Kolodziejczak, Y . Song et al., “Integrative multi-omic cancer profiling reveals dna methylation patterns associated with therapeutic vulnerability and cell-of-origin,” Cancer cell, vol. 41, no. 9, pp. 1567–1585, 2023

    work page 2023

  3. [3]

    Un- classifiable cns tumors in dna methylation-based classification: clinical challenges and prognostic impact,

    R. Drexler, F. Brembach, J. Sauvigny, F. L. Ricklefs, A. Eckhardt, H. Bode, J. Gempt, K. Lamszus, M. Westphal, U. Sch ¨uller et al., “Un- classifiable cns tumors in dna methylation-based classification: clinical challenges and prognostic impact,” Acta Neuropathologica Communica- tions, vol. 12, no. 1, p. 9, 2024

    work page 2024

  4. [4]

    Intraoperative dna methylation classification of brain tumors impacts neurosurgical strategy,

    L. Djirackor, S. Halldorsson, P. Niehusmann, H. Leske, D. Capper, L. P. Kuschel, J. Pahnke, B. J. Due-Tønnessen, I. A. Langmoen, C. J. Sandberg et al. , “Intraoperative dna methylation classification of brain tumors impacts neurosurgical strategy,”Neuro-Oncology Advances, vol. 3, no. 1, p. vdab149, 2021

    work page 2021

  5. [5]

    Dna methylation-based profiling for paediatric cns tumour diagnosis and treatment: a population-based study,

    J. C. Pickles, A. R. Fairchild, T. J. Stone, L. Brownlee, A. Merve, S. A. Yasin, A. Avery, S. W. Ahmed, O. Ogunbiyi, J. G. Zapata et al., “Dna methylation-based profiling for paediatric cns tumour diagnosis and treatment: a population-based study,” The lancet child & adolescent health, vol. 4, no. 2, pp. 121–130, 2020

    work page 2020

  6. [6]

    Methy- lation array profiling of adult brain tumours: diagnostic outcomes in a large, single centre,

    Z. Jaunmuktane, D. Capper, D. T. Jones, D. Schrimpf, M. Sill, M. Dutt, N. Suraweera, S. M. Pfister, A. von Deimling, and S. Brandner, “Methy- lation array profiling of adult brain tumours: diagnostic outcomes in a large, single centre,” Acta neuropathologica communications, vol. 7, pp. 1–18, 2019

    work page 2019

  7. [7]

    Major features of the 2021 who classification of cns tumors,

    H. L. Smith, N. Wadhwani, and C. Horbinski, “Major features of the 2021 who classification of cns tumors,”Neurotherapeutics, vol. 19, no. 6, pp. 1691–1704, 2022

    work page 2021

  8. [8]

    Pan-cancer integrative histology-genomic analysis via multimodal deep learning,

    R. J. Chen, M. Y . Lu, D. F. Williamson, T. Y . Chen, J. Lipkova, Z. Noor, M. Shaban, M. Shady, M. Williams, B. Jooet al., “Pan-cancer integrative histology-genomic analysis via multimodal deep learning,” Cancer Cell, vol. 40, no. 8, pp. 865–878, 2022

    work page 2022

Show all 40 references
  1. [9]

    Dna methylation- based classification of central nervous system tumours,

    D. Capper, D. T. Jones, M. Sill, V . Hovestadt, D. Schrimpf, D. Sturm, C. Koelsche, F. Sahm, L. Chavez, D. E. Reuss et al., “Dna methylation- based classification of central nervous system tumours,” Nature, vol. 555, no. 7697, pp. 469–474, 2018

    2018

  2. [10]

    Multimodal co-attention trans- former for survival prediction in gigapixel whole slide images,

    R. J. Chen, M. Y . Lu, W.-H. Weng, T. Y . Chen, D. F. Williamson, T. Manz, M. Shady, and F. Mahmood, “Multimodal co-attention trans- former for survival prediction in gigapixel whole slide images,” in Proceedings of the IEEE/CVF ICCV , 2021, pp. 4015–4025

    2021

  3. [11]

    Towards a general-purpose foundation model for computational pathology,

    R. J. Chen, T. Ding, M. Y . Lu, D. F. Williamson, G. Jaume, A. H. Song, B. Chen, A. Zhang, D. Shao, M. Shaban et al. , “Towards a general-purpose foundation model for computational pathology,” Nature Medicine, vol. 30, no. 3, pp. 850–862, 2024

    2024

  4. [12]

    Prediction of dna methylation-based tumor types from histopathology in central nervous system tumors with deep learning,

    D.-T. Hoang, E. D. Shulman, R. Turakulov, Z. Abdullaev, O. Singh, E. M. Campagnolo, H. Lalchungnunga, E. A. Stone, M. P. Nasrallah, E. Ruppin et al., “Prediction of dna methylation-based tumor types from histopathology in central nervous system tumors with deep learning,” Natu...

    2024

  5. [13]

    Modeling dense multimodal interactions between biolog- ical pathways and histology for survival prediction,

    G. Jaume, A. Vaidya, R. J. Chen, D. F. Williamson, P. P. Liang, and F. Mahmood, “Modeling dense multimodal interactions between biolog- ical pathways and histology for survival prediction,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (C...

    2024

  6. [14]

    Practical implementation of dna methylation and copy-number-based cns tumor diagnostics: the heidelberg experience,

    D. Capper, D. Stichel, F. Sahm, D. T. Jones, D. Schrimpf, M. Sill, S. Schmid, V . Hovestadt, D. E. Reuss, C. Koelsche et al. , “Practical implementation of dna methylation and copy-number-based cns tumor diagnostics: the heidelberg experience,”Acta neuropathologica, vol. 136, ...

    2018

  7. [15]

    Image-based deep learning model using dna methylation data predicts the origin of cancer of unknown primary,

    J. Hwang, Y . Lee, S.-K. Yoo, and J.-I. Kim, “Image-based deep learning model using dna methylation data predicts the origin of cancer of unknown primary,” Neoplasia, vol. 55, p. 101021, 2024

    2024

  8. [16]

    Whole slide images reflect dna methylation patterns of human tumors,

    H. Zheng, A. Momeni, P.-L. Cedoz, H. V ogel, and O. Gevaert, “Whole slide images reflect dna methylation patterns of human tumors,” NPJ genomic medicine, vol. 5, no. 1, p. 11, 2020

    2020

  9. [17]

    Multiomic neuropathology improves diagnostic accuracy in pediatric neuro-oncology,

    D. Sturm, D. Capper, F. Andreiuolo, M. Gessi, C. K ¨olsche, A. Rein- hardt, P. Sievers, A. K. Wefers, A. Ebrahimi, A. K. Suwala et al. , “Multiomic neuropathology improves diagnostic accuracy in pediatric neuro-oncology,” Nature medicine, vol. 29, no. 4, pp. 917–926, 2023

    2023

  10. [18]

    Predicting cancer outcomes from histology and genomics using convolutional networks,

    P. Mobadersany, S. Yousefi, M. Amgad, D. A. Gutman, J. S. Barnholtz- Sloan, J. E. Vel ´azquez Vega, D. J. Brat, and L. A. Cooper, “Predicting cancer outcomes from histology and genomics using convolutional networks,” Proceedings of the National Academy of Sciences , vol. 115, ...

    2018

  11. [19]

    Mbfusion: Multi- modal balanced fusion and multi-task learning for cancer diagnosis and prognosis,

    Z. Zhang, W. Yin, S. Wang, X. Zheng, and S. Dong, “Mbfusion: Multi- modal balanced fusion and multi-task learning for cancer diagnosis and prognosis,” Computers in Biology and Medicine , vol. 181, p. 109042, 2024

    2024

  12. [20]

    Prototypical infor- mation bottlenecking and disentangling for multimodal cancer survival prediction,

    Y . Zhang, Y . Xu, J. Chen, F. Xie, and H. Chen, “Prototypical infor- mation bottlenecking and disentangling for multimodal cancer survival prediction,” ICLR, 2024

    2024

  13. [21]

    Deep neural networks integrating genomics and histopathological images for predicting stages and survival time-to-event in colon cancer,

    O. Ogundipe, Z. Kurt, and W. L. Woo, “Deep neural networks integrating genomics and histopathological images for predicting stages and survival time-to-event in colon cancer,” Plos one , vol. 19, no. 9, p. e0305268, 2024

    2024

  14. [22]

    Multimodal prototyping for cancer survival prediction,

    A. H. Song, R. J. Chen, G. Jaume, A. J. Vaidya, A. S. Baras, and F. Mahmood, “Multimodal prototyping for cancer survival prediction,” ICML, 2024

    2024

  15. [23]

    Attention- challenging multiple instance learning for whole slide image classifica- tion,

    Y . Zhang, H. Li, Y . Sun, S. Zheng, C. Zhu, and L. Yang, “Attention- challenging multiple instance learning for whole slide image classifica- tion,” ECCV, 2024

    2024

  16. [24]

    Ensemble of prior-guided expert graph models for survival prediction in digital pathology,

    V . Ramanathan, P. Pati, M. McNeil, and A. L. Martel, “Ensemble of prior-guided expert graph models for survival prediction in digital pathology,” in International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) . Springer, 2024, pp. 262–272

    2024

  17. [25]

    Agnostic-specific modality learning for cancer survival prediction from multiple data,

    H. Liu, Y . Shi, Y . Xu, A. Li, and M. Wang, “Agnostic-specific modality learning for cancer survival prediction from multiple data,”IEEE Journal of Biomedical and Health Informatics , 2024

    2024

  18. [26]

    Multimodal optimal transport-based co-attention transformer with global structure consistency for survival prediction,

    Y . Xu and H. Chen, “Multimodal optimal transport-based co-attention transformer with global structure consistency for survival prediction,” in Proceedings of the IEEE/CVF ICCV , 2023, pp. 21 241–21 251

    2023

  19. [27]

    Morphological prototyping for unsupervised slide representation learning in computational pathology,

    A. H. Song, R. J. Chen, T. Ding, D. F. Williamson, G. Jaume, and F. Mahmood, “Morphological prototyping for unsupervised slide representation learning in computational pathology,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024...

    2024

  20. [28]

    Moab: Multi-modal outer arithmetic block for fusion of histopathological images and genetic data for brain tumor grading,

    O. Alwazzan, A. Khan, I. Patras, and G. Slabaugh, “Moab: Multi-modal outer arithmetic block for fusion of histopathological images and genetic data for brain tumor grading,” inInternational Symposium on Biomedical Imaging (ISBI). IEEE, 2023

    2023

  21. [29]

    Brain uk: Accessing nhs tissue archives for neuroscience research,

    J. A. Nicoll, T. Bloom, A. Clarke, D. Boche, and D. Hilton, “Brain uk: Accessing nhs tissue archives for neuroscience research,” Neuropathol- ogy and Applied Neurobiology , vol. 48, no. 2, p. e12766, 2022

    2022

  22. [30]

    Qupath: Open source software for digital pathology image analysis,

    P. Bankhead, M. B. Loughrey, J. A. Fern ´andez, Y . Dombrowski, D. G. McArt, P. D. Dunne, S. McQuaid, R. T. Gray, L. J. Murray, H. G. Coleman et al. , “Qupath: Open source software for digital pathology image analysis,” Scientific reports, vol. 7, no. 1, pp. 1–7, 2017

    2017

  23. [31]

    Epigenetic profiling for the molecular classification of metastatic brain tumors,

    J. I. Orozco, T. A. Knijnenburg, A. O. Manughian-Peter, M. P. Salomon, G. Barkhoudarian, J. R. Jalas, J. S. Wilmott, P. Hothi, X. Wang, Y . Takasumiet al., “Epigenetic profiling for the molecular classification of metastatic brain tumors,” Nature communications , vol. 9, no. 1...

    2018

  24. [32]

    Application of feature selection and deep learning for cancer prediction using dna methylation markers,

    R. Gomes, N. Paul, N. He, A. F. Huber, and R. J. Jansen, “Application of feature selection and deep learning for cancer prediction using dna methylation markers,” Genes, vol. 13, no. 9, p. 1557, 2022

    2022

  25. [33]

    Transmil: Transformer based correlated multiple instance learning for whole slide image classification,

    Z. Shao, H. Bian, Y . Chen, Y . Wang, J. Zhang, X. Ji et al., “Transmil: Transformer based correlated multiple instance learning for whole slide image classification,” Advances in Neural Information Processing systems, vol. 34, pp. 2136–2147, 2021

    2021

  26. [34]

    Self- normalizing neural networks,

    G. Klambauer, T. Unterthiner, A. Mayr, and S. Hochreiter, “Self- normalizing neural networks,” Advances in NeurIPS , vol. 30, 2017

    2017

  27. [35]

    Attention-based deep multiple instance learning,

    M. Ilse, J. Tomczak, and M. Welling, “Attention-based deep multiple instance learning,” in ICML. PMLR, 2018, pp. 2127–2136

    2018

  28. [36]

    Transcriptomics-guided slide representation learning in computational pathology,

    G. Jaume, L. Oldenburg, A. Vaidya, R. J. Chen, D. F. Williamson, T. Peeters, A. H. Song, and F. Mahmood, “Transcriptomics-guided slide representation learning in computational pathology,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVP...

    2024

  29. [37]

    A convnet for the 2020s,

    Z. Liu, H. Mao, C.-Y . Wu, C. Feichtenhofer, T. Darrell, and S. Xie, “A convnet for the 2020s,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , 2022, pp. 11 976– 11 986

    2022

  30. [38]

    Evaluation metrics and statistical tests for machine learning,

    O. Rainio, J. Teuho, and R. Kl ´en, “Evaluation metrics and statistical tests for machine learning,” Scientific Reports , vol. 14, no. 1, p. 6086, 2024

    2024

  31. [39]

    Mome: Mixture of multimodal experts for cancer survival prediction,

    C. Xiong, H. Chen, H. Zheng, D. Wei, Y . Zheng, J. J. Sung, and I. King, “Mome: Mixture of multimodal experts for cancer survival prediction,” in International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) . Springer, 2024, pp. 318–328

    2024

  32. [40]

    Mugi: Multi- granularity interactions of heterogeneous biomedical data for survival prediction,

    L. Long, J. Cui, P. Zeng, Y . Li, Y . Liu, and Y . Wang, “Mugi: Multi- granularity interactions of heterogeneous biomedical data for survival prediction,” in International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) . Springer, 2024, pp. 490–500

    2024

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.