Pith. sign in

REVIEW 4 major objections 7 minor 73 references

NACNet: A Histology Context-aware Transformer Graph Convolution Network for Predicting Treatment Response to Neoadjuvant Chemotherapy in Triple Negative Breast Cancer

T0 review · 4 major / 7 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 NACNet predicts breast cancer chemo response from spatial tissue graphs, 90% accurate on 105 patients.

desk verdict A plausible graph-based WSI pipeline for NAC response, but the headline 90% accuracy is compromised by a non-patient-stratified tile classifier split that leaks test-patient information into the histology maps. read the letter →

arxiv 2411.09766 v2 pith:VNX6NYR4 submitted 2024-11-14 cs.CV q-bio.QM

Qiang Li , George Teodoro , Yi Jiang , Jun Kong This is my paper · ORCID
classification cs.CVq-bio.QM
keywords wholeslideimageneoadjuvantchemotherapytriplenegativebreastcancerpathologicalcompleteresponsegraphconvolutionnetworktransformertumormicroenvironmenthistologylabelmap
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

The paper claims that NACNet, a pipeline that turns whole-slide H&E images into spatial graphs of tissue regions, can predict whether a triple-negative breast cancer patient will respond to neoadjuvant chemotherapy, achieving 90.0% accuracy, 96.0% sensitivity, 88.0% specificity, and an AUC of 0.82 in eight-fold cross-validation on 105 patients. The point of caring: only about a third of TNBC patients benefit from NAC, and the rest suffer toxicity and delay without benefit; a slide-based predictor at diagnosis could spare non-responders. The paper further claims that representing the tumor microenvironment as a graph with node features from histology labels, texture, and social-network measures outperforms both graph and non-graph baselines, and that specific spatial interactions (e.g., immune-tumor and necrosis-tumor edges) are enriched in responders.

What carries the argument

The load-bearing object is the spatial TME graph built from a WSI-derived histology label map. Each node is a spatially contiguous cluster of tiles sharing the same histology label, with a clinically weighted cluster-size threshold; each edge connects nodes within a spatial distance cutoff. Node features combine the local histology label and count, 12-dimensional texture embeddings from an autoencoder with a VGG flatten layer, and SNA centrality measures. The predictor is a transformer GCN with GIN layers that performs message passing with self-attention and graph-isomorphism updates to produce a WSI-level representation for binary classification.

What would settle it

Retrain the VGG16 histology classifier using patient-stratified splits so that all tiles of a patient are either in the tile-classifier training set or the test set, then rerun the eight-fold patient-level cross-validation of NACNet; if accuracy falls well below 90%, the original result depended on tile-level leakage. An independent external TNBC cohort run through the same fixed pipeline would serve the same check.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes that adding spatial TME context to tile-level histology analysis improves NAC response prediction. NACNet first labels 150×150-pixel tiles with a VGG16 classifier into 12 histology classes, then clusters same-label neighboring tiles into nodes of a spatial graph, connects nodes within 1,500 pixels, and represents each node by its label, label count, autoencoder-derived texture features, and social network analysis measures (degree, betweenness, PageRank, closeness). A transformer graph convolution network enhanced with graph isomorphism network layers reads this graph and predicts pathological complete response versus residual disease. Reported performance on the 105-patient cohort is 90.0% accuracy and AUC 0.82, exceeding the compared state-of-the-art graph pooling, MIL deep learning, and traditional machine learning baselines under the same eight-fold cross-validation scheme.

Load-bearing premise

The evaluation assumes that training the tile classifier on a random 80/20 split of labeled image tiles, without separating patients, does not let tiles from patients who appear in the test folds leak into the training of the histology map generator, and therefore does not inflate the patient-level cross-validation accuracy.

Editorial extensions

If this is right

  • If the result holds, spatial interactions among tumor, immune, stromal, and necrotic regions carry independent predictive signal beyond what isolated tiles provide, so tissue-arrangement-aware models should be preferred over pure MIL for heterogeneous tumors.
  • The 90% accuracy implies that, on this cohort, most patients could be stratified before treatment: responders identified at high sensitivity and non-responders at high specificity, reducing unnecessary NAC toxicity for the latter group.
  • The clinically-weighted node clustering and the reported edge/subgraph enrichments (immune-tumor, necrosis-tumor, MVD-stroma) suggest concrete, interpretable TME patterns that can be tested as biomarkers in prospective studies.
  • Because the graph construction is modality-agnostic, the pipeline can be transferred to other cancer types and to other spatially resolved tissue measurements beyond H&E.

Reading between the lines

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

  • A patient-stratified retraining of the tile classifier would test how much of the reported accuracy depends on avoiding tile-level information leakage across the cross-validation folds.
  • The SNA feature recipe (degree, betweenness, PageRank, closeness) is transferable to other spatial tissue-architecture tasks, such as survival prediction or immunotherapy response, on the same whole-slide inputs.
  • Combining this graph-of-tissue-regions representation with cell-level graphs or raw tile attention could preserve TME context while adding finer spatial resolution, a natural next architectural step.
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 / 7 minor

Summary. This manuscript presents NACNet, a multi-step deep learning pipeline for predicting pathological complete response (pCR) versus residual disease (RD) to neoadjuvant chemotherapy in triple-negative breast cancer from H&E whole-slide images. The pipeline consists of (i) a VGG16-based tile classifier that assigns one of 12 histology labels to 150x150 image tiles, (ii) construction of a spatial tumor-microenvironment graph by clustering same-label tiles in a sliding window and connecting clusters within a distance threshold, (iii) node features formed by histology label/count, autoencoder texture embeddings, and social-network-analysis measures, and (iv) a transformer graph convolutional network with graph-isomorphism-network layers. The authors evaluate the method on 105 TNBC patients with 8-fold patient-level cross-validation and report 90.0% accuracy, 96.0% sensitivity, 88.0% specificity, and AUC 0.82, together with ablations and comparisons to graph, non-graph deep learning, and traditional machine learning baselines. They also report statistical analyses of edge and 3-node subgraph distributions between pCR and RD patients.

Significance. If the reported results are valid, the paper would provide a clinically relevant demonstration that spatial TME graph context, rather than isolated-tile MIL, can improve NAC response prediction. The ablation analysis gives useful evidence about the contribution of label/count, texture, and SNA features, and the spatial interpretability analyses (edge/subgraph enrichment, attention maps) are notable strengths. However, the current evaluation protocol contains a patient-level leakage risk in the tile-classification step, and several F1 values in Table 1 are internally inconsistent. These issues must be resolved before the headline performance can be taken as evidence of generalization to new patients. No code or data are released, which further limits independent verification.

major comments (4)
  1. [§3.5, §3.1, §3.2] The VGG16 tile classifier is trained and tested on a random 80/20 split of labeled image tiles, while the NAC response model is evaluated with patient-level 8-fold cross-validation. Because the tile pool is not stratified by patient, tiles from patients in a held-out response fold can appear in the VGG16 training set. The VGG16 output defines the histology label map, and every graph node, node count, SNA feature, and associated texture embedding is derived from that map; in the ablation study, the label/count features are the most important predictors (Table 1). Patient-specific memorization by VGG16 would therefore leak information from test patients into the graph features of the response model, inflating the reported 90.0% accuracy. The authors should re-run the entire pipeline with a patient-disjoint split for the tile classifier (and for the autoencoder if it is fit on all WSIs), or provide a quantitative comparison showing that the results are unchanged under such a split.
  2. [Table 1] Several F1 values are inconsistent with the reported precision and sensitivity under the standard definition F1 = 2*precision*recall/(precision+recall). For example, NACNet-IS* reports Sensitivity 0.10±0.14 and Precision 0.14±0.19 but F1 0.79±0.16; NACNet-I reports Sensitivity 0.47, Precision 0.45, F1 0.65; NACNet-S reports 0.48/0.49/0.66; and NACNet-IS reports 0.33/0.35/0.64. These discrepancies suggest either a different averaging procedure or reporting errors. Please clarify how F1 is computed and correct the affected entries, since the ablation conclusions rely on these metric values.
  3. [§3.5] The model-selection protocol is not fully specified. The text says different learning rates were tested and a fixed hidden size, two attention heads, and two MLP layers were used (Supplement Table S1), but it does not state whether these choices were made on held-out validation folds or using the same test folds that produce the reported metrics. If the final model is selected using the test folds, the reported 8-fold accuracy is optimistically biased. Please describe a nested or otherwise leakage-free selection procedure, or report the performance of a model obtained with fixed a-priori hyperparameters.
  4. [§3.2] The graph construction depends on several hand-chosen thresholds: node cluster count thresholds η=5 and η=10, edge distance threshold ε=1500 pixels, the 10x10 sliding window, and the autoencoder latent dimension. No sensitivity analysis is provided. Because the central claim is that the spatial TME graph representation is responsible for the improvement over baselines, the authors should show that the reported performance is stable over reasonable variations of these parameters, or justify them with a validation-based selection procedure.
minor comments (7)
  1. [§3.5] The notation 'e−4, 5e−4, 1e−3, 5e−3' should be written in standard scientific notation (e.g., 10^{-4}) for clarity.
  2. [§2, §3.2] The scanner resolution is given as 0.23 µm/pixel in Section 2, but Section 3.2 describes the edge threshold with '(0.25 µm/pixel)'. Please clarify the correct conversion from pixels to physical distance.
  3. [Figure 2(a) caption] The caption says 'predict the histology labels for the rest of the titles'; 'titles' should be 'tiles'.
  4. [Algorithm 1] The allcycles function is invoked as 'allcycles(G, MCL, λ = 3)' and 'allcycles(G, MCL, λ = 2)', but MCL is described as the maximum cycle length and λ is undefined. Please clarify the roles of MCL and λ.
  5. [§3.3, Eq. (1)] The neighborhood set is written as 'µ ∈ N(v)' in the text but 'u ∈ N(v)' in the displayed equation; use a single consistent symbol.
  6. [§4.4] The paper reports many p-values for edge types (66 comparisons) and 3-node subgraph types (220 comparisons) without multiple-testing correction. Please either adjust the p-values or explicitly state that these analyses are exploratory.
  7. [General] No code or data availability statement is provided. For a methods paper of this kind, a statement on data access and implementation availability would substantially improve reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: NACNet's response prediction is not definitionally tied to its fitted inputs, and the response label is not used to construct the graph features.

full rationale

The paper's claimed derivation chain is a supervised pipeline: pathologist-annotated image tiles train a VGG16 histology classifier; predicted histology maps are clustered into spatial TME graph nodes; node features combine histology label, label count, autoencoder texture embeddings, and SNA features; a transformer-GCN with GIN layers maps these graphs to pCR/RD. At no point is the NAC response label used to define the graph nodes, edges, or node features, and no equation defines the prediction from a fitted parameter that was itself fit to the response. The evaluation uses patient-level 8-fold cross-validation, and the response model is trained and tested on held-out patients, so the headline accuracy is not the same quantity as a training fit. The only self-citation by an author (ref. [20]) is introductory support for prior deep-learning prediction work and is not load-bearing for the method or the main claim. The more substantive concern is evaluation leakage: Section 3.5 states that the VGG16 tile classifier is trained and tested on 'randomly select[ed] 80% and 20% of the labeled image tiles' without patient-level stratification, so tiles from patients in held-out response folds can enter tile-classifier training. This is an optimism/validity concern about the reported 90.0% generalization estimate, not a circularity: the tile classifier learns histology appearance from pathologist labels, not from the pCR/RD outcome, and the response label is not used to define any graph feature. Therefore no definitional reduction, fitted-input-as-prediction, or load-bearing self-citation chain is present.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The central prediction claim rests on the hand-chosen graph construction (eta, epsilon), the 12-class histology model, the assumption that the graph captures TME biology, and the standard GNN machinery. No new physical entities are introduced. The main unstated cost is the tile classifier's training split, which can bias the response evaluation.

free parameters (5)
  • Node cluster count thresholds eta (eta_critical=5, eta_other=10) = eta=5 for tumor/necrosis/immune/MVD/stroma/PGCC; eta=10 for others
    Hand-chosen thresholds in Section 3.2 to decide which tile clusters become graph nodes; no data-driven selection or sensitivity analysis reported.
  • Edge distance threshold epsilon = 1,500 pixels (paper states 0.25 um/pixel; scanner resolution is 0.23 um/pixel)
    Section 3.2: nodes within epsilon are connected; value appears arbitrary and the physical unit is inconsistent with dataset description.
  • Autoencoder latent dimension = 12
    Section 3.2: 'resulting in a feature vector of length 12'; no justification for this dimensionality.
  • VGG16 background cutoff and tile retention rule = intensity >220 in all RGB channels; discard tiles >90% background
    Section 3.1: hand-chosen thresholds for background detection.
  • Model hyperparameters (hidden size, heads, layers, dropout, learning rate) = hidden size 256, 2 attention heads, 2 MLP layers, dropout 0.5; LR searched {1e-4..5e-3}
    Section 3.5 and Supplement Table S1; hyperparameters selected without a nested cross-validation, so the reported CV performance can be optimistically biased.
assumptions (6)
  • domain assumption The 12 annotated histology classes (adipose, PGCC, MVD, tumor, stroma, CIS, hemorrhage, mucinous, apocrine, immune, necrosis, muscle) are sufficient to represent TME components relevant to NAC response.
    Section 2 defines the annotation set; graph nodes are formed only from these labels, so any TME signal outside these classes is invisible to the model.
  • domain assumption The VGG16 tile classifier's labels are accurate enough that classification errors do not materially change the TME graph or response predictions.
    Section 4.1 reports 90% tile accuracy; no error propagation analysis is given.
  • ad hoc to paper The graph constructed with hand-chosen parameters (eta=5/10, epsilon=1500 pixels) preserves clinically meaningful spatial TME interactions.
    Section 3.2 describes thresholds chosen for clinical relevance; no sensitivity analysis is provided.
  • standard math GIN, transformer self-attention, and GCN update rules in Equations (1)-(3) behave as standard graph learning layers.
    Based on references [40,50,51]; standard machinery.
  • domain assumption The single-center cohort of 105 TNBC patients is representative enough to support the stated generalizable claims.
    Section 2; no external cohort, no patient demographic details, so generalizability is assumed rather than shown.
  • domain assumption Training the tile classifier on a non-patient-stratified tile split does not introduce leakage into the patient-level response evaluation.
    Section 3.5: tile classifier uses a random 80/20 split; the response model's CV is patient-level; the paper implicitly assumes no leakage.

how reviews work

0 comments
Cite this review

Pith. "Pith review of NACNet: A Histology Context-aware Transformer Graph Convolution Network for Predicting Treatment Response to Neoadjuvant Chemotherapy in Triple Negative Breast Cancer." pith.science (2026). https://pith.science/paper/VNX6NYR4

@misc{pith2026241109766,
  author       = {Pith},
  title        = {Pith review of: NACNet: A Histology Context-aware Transformer Graph Convolution Network for Predicting Treatment Response to Neoadjuvant Chemotherapy in Triple Negative Breast Cancer},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VNX6NYR4}},
  note         = {Machine review of arXiv:2411.09766}
}
read the original abstract

Neoadjuvant chemotherapy (NAC) response prediction for triple negative breast cancer (TNBC) patients is a challenging task clinically as it requires understanding complex histology interactions within the tumor microenvironment (TME). Digital whole slide images (WSIs) capture detailed tissue information, but their giga-pixel size necessitates computational methods based on multiple instance learning, which typically analyze small, isolated image tiles without the spatial context of the TME. To address this limitation and incorporate TME spatial histology interactions in predicting NAC response for TNBC patients, we developed a histology context-aware transformer graph convolution network (NACNet). Our deep learning method identifies the histopathological labels on individual image tiles from WSIs, constructs a spatial TME graph, and represents each node with features derived from tissue texture and social network analysis. It predicts NAC response using a transformer graph convolution network model enhanced with graph isomorphism network layers. We evaluate our method with WSIs of a cohort of TNBC patient (N=105) and compared its performance with multiple state-of-the-art machine learning and deep learning models, including both graph and non-graph approaches. Our NACNet achieves 90.0% accuracy, 96.0% sensitivity, 88.0% specificity, and an AUC of 0.82, through eight-fold cross-validation, outperforming baseline models. These comprehensive experimental results suggest that NACNet holds strong potential for stratifying TNBC patients by NAC response, thereby helping to prevent overtreatment, improve patient quality of life, reduce treatment cost, and enhance clinical outcomes, marking an important advancement toward personalized breast cancer treatment.

Figures

Figures reproduced from arXiv: 2411.09766 by the authors.

Figure 1
Figure 1. Overview of the NACNet architecture. Each WSI is partitioned into non-overlapping image tiles of size 150×150 pixels.A pre-trained convolutional model for image tile classification identifies all image tile classes, resulting in a tile-level histology label map where each pixel represents the histology label of an image tile. A sliding window moves over each histology label map and defines neighbors that share the s… view at source ↗
Figure 2
Figure 2. Feature extration from WSI spatial TME graph. [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. WSI-derived spatial TME graph. We present representative WSIs (top), the corresponding histology maps (middle), and the resulting WSI TME graph node distributions (bottom). The histology labels of image tiles of size 150 × 150 are classified and assembled to construct the histology map. In total, 12 histology labels are color coded, including hemorrhage, immune cells, carcinoma in situ (CIS), MVD, mucinous changes, … view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: ROC curves of NACNet models in the ablation study. [PITH_FULL_IMAGE:figures/full_fig_p013_4.png]
Figure 5
Figure 5. Figure 5: The ROC curves of methods for comparison. [PITH_FULL_IMAGE:figures/full_fig_p015_5.png]
Figure 6
Figure 6. Figure 6: Comparison of histology profiles between pCR and RD patients. [PITH_FULL_IMAGE:figures/full_fig_p017_6.png]
Figure 7
Figure 7. Figure 7: Comparison of WSI-derived graph edge profiles between pCR and RD patients. [PITH_FULL_IMAGE:figures/full_fig_p018_7.png]
Figure 8
Figure 8. Figure 8: Comparison of WSI-derived subgraph profiles between pCR and RD patients. [PITH_FULL_IMAGE:figures/full_fig_p019_8.png]
Figure 9
Figure 9. Figure 9: Prediction model interpretations with correlation and relevancy importance maps. [PITH_FULL_IMAGE:figures/full_fig_p020_9.png]
Figure 10
Figure 10. Figure 10: Attention visualization for typical tissue regions. [PITH_FULL_IMAGE:figures/full_fig_p021_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

73 extracted references · 58 canonical work pages

  1. [1]

    Triple-negative breast cancer: epidemiological considerations and recommendations

    P Boyle. Triple-negative breast cancer: epidemiological considerations and recommendations. Annals of oncology, 23:vi7–vi12, 2012

    work page 2012

  2. [2]

    Tissue-resident macrophages are major tumor- associated macrophage resources, contributing to early tnbc development, recurrence, and metastases

    Ryuichiro Hirano, Koki Okamoto, Miyu Shinke, Marika Sato, Shigeaki Watanabe, Hitomi Watanabe, Gen Kondoh, Tetsuya Kadonosono, and Shinae Kizaka-Kondoh. Tissue-resident macrophages are major tumor- associated macrophage resources, contributing to early tnbc development, recurrence, and metastases. Communications Biology, 6(1):144, 2023

    work page 2023

  3. [3]

    Molecular features and clinical implications of the heterogeneity in chinese patients with her2-low breast cancer

    Lei-Jie Dai, Ding Ma, Yu-Zheng Xu, Ming Li, Yu-Wei Li, Yi Xiao, Xi Jin, Song-Yang Wu, Ya-Xin Zhao, Han Wang, et al. Molecular features and clinical implications of the heterogeneity in chinese patients with her2-low breast cancer. Nature Communications, 14(1):5112, 2023

    work page 2023

  4. [4]

    Triple- negative breast cancer: current perspective on the evolving therapeutic landscape

    Joe Mehanna, Fady GH Haddad, Roland Eid, Matteo Lambertini, and Hampig Raphael Kourie. Triple- negative breast cancer: current perspective on the evolving therapeutic landscape. International journal of women’s health, 11:431, 2019

    work page 2019

  5. [5]

    Immunotherapy in triple-negative breast cancer: A literature review and new advances

    Guillermo Arturo Valencia, Patricia Rioja, Zaida Morante, Rossana Ruiz, Hugo Fuentes, Carlos A Castaneda, Tatiana Vidaurre, Silvia Neciosup, and Henry L Gomez. Immunotherapy in triple-negative breast cancer: A literature review and new advances. World Journal of Clinical Oncology, 13(3):219, 2022

    work page 2022

  6. [6]

    Triple- negative breast cancer: challenges and opportunities of a heterogeneous disease

    Giampaolo Bianchini, Justin M Balko, Ingrid A Mayer, Melinda E Sanders, and Luca Gianni. Triple- negative breast cancer: challenges and opportunities of a heterogeneous disease. Nature reviews Clinical oncology, 13(11):674–690, 2016

    work page 2016

  7. [7]

    Pathological complete response and long-term clinical benefit in breast cancer: the ctneobc pooled analysis

    Patricia Cortazar, Lijun Zhang, Michael Untch, Keyur Mehta, Joseph P Costantino, Norman Wolmark, Herv´ e Bonnefoi, David Cameron, Luca Gianni, Pinuccia Valagussa, et al. Pathological complete response and long-term clinical benefit in breast cancer: the ctneobc pooled analysis. The Lancet, 384(9938):164– 172, 2014

    work page 2014

  8. [8]

    Relevance of pathological complete response after neoadjuvant therapy for breast cancer

    Angela Pennisi, Thomas Kieber-Emmons, Issam Makhoul, and Laura Hutchins. Relevance of pathological complete response after neoadjuvant therapy for breast cancer. Breast cancer: basic and clinical research, 10:BCBCR–S33163, 2016. 23

    work page 2016

Show all 73 references
  1. [9]

    Conservative surgery after neoadjuvant chemotherapy in patients with operable breast cancer

    Gianluca Franceschini, Alba Di Leone, Maria Natale, Martin Aleandro Sanchez, and Riccardo Masett. Conservative surgery after neoadjuvant chemotherapy in patients with operable breast cancer. Annali Italiani di Chirurgia, 89:290–290, 2018

    2018

  2. [10]

    Pathological tumor response to neoadjuvant chemotherapy using anthracycline and taxanes in patients with triple-negative breast cancer

    Kaori Sakuma, Masafumi Kurosumi, Hanako Oba, Yasuhito Kobayashi, Hiroyuki Takei, Kenichi Inoue, Toshio Tabei, and Tetsunari Oyama. Pathological tumor response to neoadjuvant chemotherapy using anthracycline and taxanes in patients with triple-negative breast cancer. Experiment...

    2011

  3. [11]

    A pathology foundation model for cancer diagnosis and prognosis prediction

    Xiyue Wang, Junhan Zhao, Eliana Marostica, Wei Yuan, Jietian Jin, Jiayu Zhang, Ruijiang Li, Hongping Tang, Kanran Wang, Yu Li, et al. A pathology foundation model for cancer diagnosis and prognosis prediction. Nature, pages 1–9, 2024

    2024

  4. [12]

    Histopathologic image–based deep learning classifier for predicting platinum-based treatment responses in high-grade serous ovarian cancer

    Byungsoo Ahn, Damin Moon, Hyun-Soo Kim, Chung Lee, Nam Hoon Cho, Heung-Kook Choi, Dong- min Kim, Jung-Yun Lee, Eun Ji Nam, Dongju Won, et al. Histopathologic image–based deep learning classifier for predicting platinum-based treatment responses in high-grade serous ovarian can...

    2024

  5. [13]

    Deep learning for the prediction of early on-treatment response in metastatic colorectal cancer from serial medical imaging.Nature communications, 12(1):6654, 2021

    Lin Lu, Laurent Dercle, Binsheng Zhao, and Lawrence H Schwartz. Deep learning for the prediction of early on-treatment response in metastatic colorectal cancer from serial medical imaging.Nature communications, 12(1):6654, 2021

    2021

  6. [14]

    A deep-learning framework to predict cancer treatment response from histopathology images through imputed transcriptomics

    Danh-Tai Hoang, Gal Dinstag, Eldad D Shulman, Leandro C Hermida, Doreen S Ben-Zvi, Efrat Elis, Katherine Caley, Stephen-John Sammut, Sanju Sinha, Neelam Sinha, et al. A deep-learning framework to predict cancer treatment response from histopathology images through imputed tran...

    2024

  7. [15]

    Prediction of early clinical response to neoadjuvant chemotherapy in triple-negative breast cancer: Incorporating radiomics through breast mri

    Hyo-jae Lee, Jeong Hoon Lee, Jong Eun Lee, Yong Min Na, Min Ho Park, Ji Shin Lee, and Hyo Soon Lim. Prediction of early clinical response to neoadjuvant chemotherapy in triple-negative breast cancer: Incorporating radiomics through breast mri. Scientific Reports, 14(1):21691, 2024

    2024

  8. [16]

    Machine learning for genetics-based classification and treatment response prediction in cancer of unknown primary

    Intae Moon, Jaclyn LoPiccolo, Sylvan C Baca, Lynette M Sholl, Kenneth L Kehl, Michael J Hassett, David Liu, Deborah Schrag, and Alexander Gusev. Machine learning for genetics-based classification and treatment response prediction in cancer of unknown primary. Nature Medicine, ...

    2023

  9. [17]

    Whole slide imaging (wsi) in pathology: current perspectives and future directions

    Neeta Kumar, Ruchika Gupta, and Sanjay Gupta. Whole slide imaging (wsi) in pathology: current perspectives and future directions. Journal of digital imaging, 33(4):1034–1040, 2020

    2020

  10. [18]

    Deep learning for whole slide image analysis: an overview

    Neofytos Dimitriou, Ognjen Arandjelovi´ c, and Peter D Caie. Deep learning for whole slide image analysis: an overview. Frontiers in medicine, 6:264, 2019

    2019

  11. [19]

    Orbit image analysis: an open-source whole slide image analysis tool

    Manuel Stritt, Anna K Stalder, and Enrico Vezzali. Orbit image analysis: an open-source whole slide image analysis tool. PLoS computational biology, 16(2):e1007313, 2020. 24

    2020

  12. [20]

    Prediction of pathological complete response to neoadjuvant chemo- therapy in breast cancer using deep learning with integrative imaging, molecular and demographic data

    Hongyi Duanmu, Pauline Boning Huang, Srinidhi Brahmavar, Stephanie Lin, Thomas Ren, Jun Kong, Fusheng Wang, and Tim Q Duong. Prediction of pathological complete response to neoadjuvant chemo- therapy in breast cancer using deep learning with integrative imaging, molecular and ...

    2020

  13. [21]

    Han-Byoel Lee, Wonshik Han, Soo-Yeon Kim, Nariya Cho, Kyoung-Eun Kim, Jung Hyun Park, Young Wook Ju, Eun-Shin Lee, Sung-Joon Lim, Jung Ho Kim, et al. Prediction of pathologic complete re- sponse using image-guided biopsy after neoadjuvant chemotherapy in breast cancer patients...

    2020

  14. [22]

    Prediction of pathologic response to neoadjuvant chemotherapy in patients with breast cancer using diffusion-weighted imaging and mrs

    Hee Jung Shin, Hyeon-Man Baek, Jin-Hee Ahn, Seunghee Baek, Hyunji Kim, Joo Hee Cha, and Hak Hee Kim. Prediction of pathologic response to neoadjuvant chemotherapy in patients with breast cancer using diffusion-weighted imaging and mrs. NMR in Biomedicine, 25(12):1349–1359, 2012

    2012

  15. [23]

    Prediction of pathological complete response to neoadjuvant chemotherapy in breast cancer using a deep learning (dl) method

    Yu-Hong Qu, Hai-Tao Zhu, Kun Cao, Xiao-Ting Li, Meng Ye, and Ying-Shi Sun. Prediction of pathological complete response to neoadjuvant chemotherapy in breast cancer using a deep learning (dl) method. Thoracic Cancer, 11(3):651–658, 2020

    2020

  16. [24]

    A machine learning model to predict the triple negative breast cancer immune subtype

    Zihao Chen, Maoli Wang, Rudy Leon De Wilde, Ruifa Feng, Mingqiang Su, Luz Angela Torres-de la Roche, and Wenjie Shi. A machine learning model to predict the triple negative breast cancer immune subtype. Frontiers in immunology, 12, 2021

    2021

  17. [25]

    A graph-transformer for whole slide image classification

    Yi Zheng, Rushin H Gindra, Emily J Green, Eric J Burks, Margrit Betke, Jennifer E Beane, and Vijaya B Kolachalama. A graph-transformer for whole slide image classification. IEEE transactions on medical imaging, 41(11):3003–3015, 2022

    2022

  18. [26]

    Multiple instance learning with general- ized support vector machines

    Stuart Andrews, Thomas Hofmann, and Ioannis Tsochantaridis. Multiple instance learning with general- ized support vector machines. In AAAI/IAAI, pages 943–944, 2002

    2002

  19. [27]

    Multimodal co-attention transformer for survival prediction in gigapixel whole slide images

    Richard J Chen, Ming Y Lu, Wei-Hung Weng, Tiffany Y Chen, Drew FK Williamson, Trevor Manz, Maha Shady, and Faisal Mahmood. Multimodal co-attention transformer for survival prediction in gigapixel whole slide images. In Proceedings of the IEEE/CVF International Conference on Co...

    2021

  20. [28]

    Automatic detection of invasive ductal carcinoma in whole slide images with convolutional neural networks

    Angel Cruz-Roa, Ajay Basavanhally, Fabio Gonz´ alez, Hannah Gilmore, Michael Feldman, Shridar Ganesan, Natalie Shih, John Tomaszewski, and Anant Madabhushi. Automatic detection of invasive ductal carcinoma in whole slide images with convolutional neural networks. In Medical Im...

    2014

  21. [29]

    Gradient-based learning applied to document recognition

    Yann LeCun, L´ eon Bottou, Yoshua Bengio, and Patrick Haffner. Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11):2278–2324, 1998. 25

    1998

  22. [30]

    Weakly supervised discriminative localization and classification: a joint learning process

    Minh Hoai Nguyen, Lorenzo Torresani, Fernando De La Torre, and Carsten Rother. Weakly supervised discriminative localization and classification: a joint learning process. In 2009 IEEE 12th International Conference on Computer Vision, pages 1925–1932. IEEE, 2009

    2009

  23. [31]

    Treatment landscape of triple- negative breast cancer—expanded options, evolving needs

    Giampaolo Bianchini, Carmine De Angelis, Luca Licata, and Luca Gianni. Treatment landscape of triple- negative breast cancer—expanded options, evolving needs. Nature reviews Clinical oncology, 19(2):91–113, 2022

    2022

  24. [32]

    A structured tumor-immune microenvironment in triple negative breast cancer revealed by multiplexed ion beam imaging

    Leeat Keren, Marc Bosse, Diana Marquez, Roshan Angoshtari, Samir Jain, Sushama Varma, Soo-Ryum Yang, Allison Kurian, David Van Valen, Robert West, et al. A structured tumor-immune microenvironment in triple negative breast cancer revealed by multiplexed ion beam imaging. Cell,...

    2018

  25. [33]

    Mucosal b cell differentiation and regulation

    Nils Lycke, Mats Bemark, and Jo Spencer. Mucosal b cell differentiation and regulation. pages 701–719, 2015

    2015

  26. [34]

    An overview of lasers in dentistry

    Donald J Coluzzi. An overview of lasers in dentistry. The Alpha omegan, 101(3):125–126, 2008

    2008

  27. [35]

    Graph cnn for survival analysis on whole slide pathological images

    Ruoyu Li, Jiawen Yao, Xinliang Zhu, Yeqing Li, and Junzhou Huang. Graph cnn for survival analysis on whole slide pathological images. pages 174–182, 2018

    2018

  28. [36]

    Topological feature extraction and visualization of whole slide images using graph neural networks

    Joshua Levy, Christian Haudenschild, Clark Barwick, Brock Christensen, and Louis Vaickus. Topological feature extraction and visualization of whole slide images using graph neural networks. pages 285–296, 2020

    2020

  29. [37]

    Whole slide images are 2d point clouds: Context-aware survival prediction using patch-based graph convolutional networks

    Richard J Chen, Ming Y Lu, Muhammad Shaban, Chengkuan Chen, Tiffany Y Chen, Drew FK Willi- amson, and Faisal Mahmood. Whole slide images are 2d point clouds: Context-aware survival prediction using patch-based graph convolutional networks. pages 339–349, 2021

    2021

  30. [38]

    Representation learning on graphs: Methods and applications

    William L Hamilton, Rex Ying, and Jure Leskovec. Representation learning on graphs: Methods and applications. arXiv preprint arXiv:1709.05584, 2017

    2017 arXiv

  31. [39]

    Slide- graph+: Whole slide image level graphs to predict her2 status in breast cancer

    Wenqi Lu, Michael Toss, Muhammad Dawood, Emad Rakha, Nasir Rajpoot, and Fayyaz Minhas. Slide- graph+: Whole slide image level graphs to predict her2 status in breast cancer. Medical Image Analysis, 80:102486, 2022

    2022

  32. [40]

    How powerful are graph neural networks? arXiv preprint arXiv:1810.00826, 2018

    Keyulu Xu, Weihua Hu, Jure Leskovec, and Stefanie Jegelka. How powerful are graph neural networks? arXiv preprint arXiv:1810.00826, 2018

    2018 arXiv

  33. [41]

    Very deep convolutional networks for large-scale image recogni- tion

    Karen Simonyan and Andrew Zisserman. Very deep convolutional networks for large-scale image recogni- tion. arXiv preprint arXiv:1409.1556, 2014

    2014 arXiv

  34. [42]

    Guided soft attention network for classification of breast cancer histopathology images

    Heechan Yang, Ji-Ye Kim, Hyongsuk Kim, and Shyam P Adhikari. Guided soft attention network for classification of breast cancer histopathology images. IEEE transactions on medical imaging, 39(5):1306– 1315, 2019. 26

    2019

  35. [43]

    Hallmarks of cancer: the next generation

    Douglas Hanahan and Robert A Weinberg. Hallmarks of cancer: the next generation. cell, 144(5):646–674, 2011

    2011

  36. [44]

    Tumor-associated lymphocytes as an independent predictor of response to neoadjuvant chemotherapy in breast cancer

    Carsten Denkert, Sibylle Loibl, Aurelia Noske, Marc Roller, Berit Maria M¨ uller, Martina Komor, Jan Budczies, Silvia Darb-Esfahani, Ralf Kronenwett, Claus Hanusch, et al. Tumor-associated lymphocytes as an independent predictor of response to neoadjuvant chemotherapy in breas...

    2010

  37. [45]

    Role of angiogenesis in tumor growth and metastasis

    Judah Folkman. Role of angiogenesis in tumor growth and metastasis. In Seminars in oncology, volume 29, pages 15–18. Elsevier, 2002

    2002

  38. [46]

    Neural graph modelling of whole slide images for survival ranking

    Callum Christopher Mackenzie, Muhammad Dawood, Simon Graham, Mark Eastwood, et al. Neural graph modelling of whole slide images for survival ranking. In Learning on Graphs Conference, pages 48–1. PMLR, 2022

    2022

  39. [47]

    Social network analysis: a powerful strategy, also for the information sciences

    Evelien Otte and Ronald Rousseau. Social network analysis: a powerful strategy, also for the information sciences. Journal of information Science, 28(6):441–453, 2002

    2002

  40. [48]

    The google pagerank algorithm and how it works

    Ian Rogers. The google pagerank algorithm and how it works. 2002

    2002

  41. [49]

    Autoencoder for words

    Cheng-Yuan Liou, Wei-Chen Cheng, Jiun-Wei Liou, and Daw-Ran Liou. Autoencoder for words. Neurocomputing, 139:84–96, 2014

    2014

  42. [50]

    Universal graph transformer self-attention networks

    Dai Quoc Nguyen, Tu Dinh Nguyen, and Dinh Phung. Universal graph transformer self-attention networks. In Companion Proceedings of the Web Conference 2022, pages 193–196, 2022

    2022

  43. [51]

    Semi-supervised classification with graph convolutional networks

    Thomas N Kipf and Max Welling. Semi-supervised classification with graph convolutional networks. arXiv preprint arXiv:1609.02907, 2016

    2016 arXiv

  44. [52]

    Deep graph kernels

    Pinar Yanardag and SVN Vishwanathan. Deep graph kernels. pages 1365–1374, 2015

    2015

  45. [53]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. pages 770–778, 2016

    2016

  46. [54]

    Layer normalization

    Jimmy Lei Ba, Jamie Ryan Kiros, and Geoffrey E Hinton. Layer normalization. arXiv preprint arXiv:1607.06450, 2016

    2016 arXiv

  47. [55]

    Minimum redundancy feature selection from microarray gene expression data

    Chris Ding and Hanchuan Peng. Minimum redundancy feature selection from microarray gene expression data. Journal of bioinformatics and computational biology, 3(02):185–205, 2005

    2005

  48. [56]

    Adam: A method for stochastic optimization

    Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014

    2014 arXiv

  49. [57]

    Maximum entropy weighted independent set pooling for graph neural networks

    Amirhossein Nouranizadeh, Mohammadjavad Matinkia, Mohammad Rahmati, and Reza Safabakhsh. Maximum entropy weighted independent set pooling for graph neural networks. arXiv preprint arXiv:2107.01410, 2021. 27

    2021 arXiv

  50. [58]

    An end-to-end deep learning architecture for graph classification

    Muhan Zhang, Zhicheng Cui, Marion Neumann, and Yixin Chen. An end-to-end deep learning architecture for graph classification. 32(1), 2018

    2018

  51. [59]

    Self-attention graph pooling

    Junhyun Lee, Inyeop Lee, and Jaewoo Kang. Self-attention graph pooling. In International conference on machine learning, pages 3734–3743. PMLR, 2019

    2019

  52. [60]

    Hierarchical graph pooling with structure learning

    Zhen Zhang, Jiajun Bu, Martin Ester, Jianfeng Zhang, Chengwei Yao, Zhi Yu, and Can Wang. Hierarchical graph pooling with structure learning. arXiv preprint arXiv:1911.05954, 2019

    1911 arXiv

  53. [61]

    Imagenet classification with deep convolutional neural networks

    Alex Krizhevsky, Ilya Sutskever, and Geoffrey E Hinton. Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems, 25, 2012

    2012

  54. [62]

    Applied logistic regression, 2

    David W Hosmer. Applied logistic regression, 2. 2000

    2000

  55. [63]

    Nearest neighbor pattern classification

    Thomas Cover and Peter Hart. Nearest neighbor pattern classification. IEEE transactions on information theory, 13(1):21–27, 1967

    1967

  56. [64]

    Support-vector networks

    Corinna Cortes and Vladimir Vapnik. Support-vector networks. Machine learning, 20:273–297, 1995

    1995

  57. [65]

    The optimality of naive bayes

    Harry Zhang. The optimality of naive bayes. Aa, 1(2):3, 2004

    2004

  58. [66]

    Ross Quinlan

    J. Ross Quinlan. Induction of decision trees. Machine learning, 1:81–106, 1986

    1986

  59. [67]

    Leo Breiman. Rd. Machine learning, 45:5–32, 2001

    2001

  60. [68]

    Extremely randomized trees

    Pierre Geurts, Damien Ernst, and Louis Wehenkel. Extremely randomized trees. Machine learning, 63:3– 42, 2006

    2006

  61. [69]

    Linear discriminant analysis-a brief tutorial

    Suresh Balakrishnama and Aravind Ganapathiraju. Linear discriminant analysis-a brief tutorial. Institute for Signal and information Processing, 18(1998):1–8, 1998

    1998

  62. [70]

    Scikit-learn: Machine learning in python

    Fabian Pedregosa, Ga¨ el Varoquaux, Alexandre Gramfort, Vincent Michel, Bertrand Thirion, Olivier Grisel, Mathieu Blondel, Peter Prettenhofer, Ron Weiss, Vincent Dubourg, et al. Scikit-learn: Machine learning in python. the Journal of machine Learning research, 12:2825–2830, 2011

    2011

  63. [71]

    Inflammation and cancer: triggers, mechanisms, and con- sequences

    Florian R Greten and Sergei I Grivennikov. Inflammation and cancer: triggers, mechanisms, and con- sequences. Immunity, 51(1):27–41, 2019

    2019

  64. [72]

    Deep learning-based tumor microenvironment analysis in colon adenocarcinoma histopathological whole-slide images

    Yiping Jiao, Junhong Li, Chenqi Qian, and Shumin Fei. Deep learning-based tumor microenvironment analysis in colon adenocarcinoma histopathological whole-slide images. Computer Methods and Programs in Biomedicine, 204:106047, 2021

    2021

  65. [73]

    A population-level compu- tational histologic signature for invasive breast cancer prognosis

    Mohamed Amgad, James Hodge, Maha Elsebaie, Clara Bodelon, Samantha Puvanesarajah, David Gut- man, Kalliopi Siziopikou, Jeffery Goldstein, Mia Gaudet, Lauren Teras, et al. A population-level compu- tational histologic signature for invasive breast cancer prognosis. 2023. 28

    2023

Pith tools

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