On the importance of local and global feature learning for automated measurable residual disease detection in flow cytometry data

Lisa Weijler; Margarita Maurer-Granofszky; Michael Dworzak; Michael Reiter; Pedro Hermosilla

REVIEW 4 major objections 5 minor 35 references

On the importance of local and global feature learning for automated measurable residual disease detection in flow cytometry data

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 Combining farthest-point-sampled attention anchors with a GNN layer improves automated MRD detection in flow cytometry data.

desk verdict A solid empirical paper on MRD detection with a genuinely sensible architectural tweak (FPS-selected inducing points plus a GNN layer), held back by missing significance tests and an unclarified patient/sample split. read the letter →

arxiv 2411.15621 v1 pith:FAUFCCZ7 submitted 2024-11-23 cs.CV

Lisa Weijler , Michael Reiter , Pedro Hermosilla , Margarita Maurer-Granofszky , Michael Dworzak This is my paper · ORCID

classification cs.CV

keywords flowcytometrymeasurableresidualdiseaseacutelymphoblasticleukemiadeeplearningsettransformergraphneuralnetworksfarthestpointsamplingsingle-cellclassification

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 asks whether automated detection of measurable residual disease (MRD) in flow cytometry benefits from modelling long-range dependencies, from how the global summary is obtained, and from explicit local feature learning. It answers yes to all three, with a caveat on 'how': replacing the learned query vectors of a Set Transformer with event feature vectors selected by farthest-point sampling improves results, and inserting one graph neural network layer for local feature aggregation helps further. The resulting model, GIN-ST-FPS, reaches an average $F_1$ of $0.8665 \pm 0.0083$ on the public Vie dataset, compared with $0.8284 \pm 0.0117$ for the previous transformer baseline, and it generalizes better to two other laboratories' datasets. The practical point is a concrete architectural recipe for applying deep learning to unstructured single-cell data where marker panels vary between samples.

What carries the argument

The load-bearing component is the induced set attention block (ISAB) of the Set Transformer, modified so that the inducing points are not learned query vectors but event feature vectors picked from the input by farthest-point sampling, combined with a single graph neural network layer (GIN or GAT) on a $k$-NN graph with $k=10$. The FPS-sampled inducing points give attention a sample-specific global summary while keeping complexity linear in the number of events, and the GNN layer injects explicit local structure that the otherwise permutation-equivariant transformer lacks. The paper's experiments isolate the contribution of each part: removing self-attention hurts, using FPS beats learned inducing points even when the transformer is given more learned points, and capping the GNN receptive field at $k=3$ hurts performance.

What would settle it

Train GIN-ST-FPS and the ST baseline on a patient-stratified split of the Vie dataset, placing every sample from a given patient in the same fold, and compare average $F_1$ on the held-out patients; if the advantage of GIN-ST-FPS shrinks or reverses, the reported gains are partly patient-specific memorization rather than generalization.

Watch

Extended reading notes

Core claim

The paper's central claim is that the current state of the art for automated MRD detection can be pushed further by two targeted changes to the Set Transformer. First, instead of using a fixed set of learned inducing points, the model should sample roughly 150 event feature vectors from the input sample using farthest-point sampling, so the self-attention mechanism is anchored in the sample's own cell-population structure. Second, one GNN layer (GIN or GAT) should be placed before the attention blocks to explicitly aggregate local neighborhoods via a $k$-NN graph with $k=10$, giving the model an inductive bias toward spatial locality while the attention layers still capture long-range dependencies. With both changes, GIN-ST-FPS reaches average $F_1 = 0.8665 \pm 0.0083$ on Vie, outperforming the ST baseline at $0.8284 \pm 0.0117$, and improves average $F_1$ on the cross-laboratory Bln and Bue sets from about $0.61$ to $0.70$ and from $0.73$ to $0.84$. The paper also reports that when three diagnostically important markers are removed from node features but kept for graph construction, the GNN-infused models degrade less, supporting the claim that graph structure itself carries usable sample-specific information.

Load-bearing premise

The load-bearing premise is that a random 50/25/25 split by sample keeps training and test sets independent, even though the same patient can contribute several follow-up samples in the Vie dataset.

Editorial extensions

If this is right

On the public Vie dataset, GIN-ST-FPS achieves average $F_1$ of $0.8665 \pm 0.0083$, a gain over the previous transformer baseline's $0.8284 \pm 0.0117$.
On the cross-laboratory Bln and Bue datasets, the same model improves average $F_1$ substantially over the ST baseline, indicating better generalization to unseen laboratories' data.
How the global summary is constructed matters: sample-derived FPS anchors outperform learned inducing points, and merely increasing the number of learned inducing points does not reproduce the gain.
Explicit local feature learning is beneficial on its own: GNN-only models match or slightly exceed the transformer baseline when the $k$-NN graph is connected enough, and a restricted $k=3$ graph degrades results.
Graph structure carries usable information beyond the marker values: when CD10, CD19, and CD45 are removed from node features but kept for graph construction, GNN-infused models retain noticeably more $F_1$ than transformer-only models.

Reading between the lines

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

An implication the paper leaves implicit is that a patient-level split of the Vie dataset would be a stronger generalization test, because repeated samples from the same patient could inflate the reported gains.
The FPS-and-GNN recipe could transfer to other rare-cell-detection tasks in flow or mass cytometry, where event counts are high and labeled data are scarce.
Making the local graph adaptive (e.g., choosing $k$ per cell-population density) is a natural extension that might sharpen detection of leukemic cells sitting on the edge of healthy clusters.

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.

Desk Editor's Note

read the letter

Read this paper. It is a straightforward, well-run empirical study that proposes two small architectural changes to the Set Transformer for flow cytometry MRD detection: replacing learned inducing points with farthest-point-sampled event vectors, and adding one GNN layer for local features. The result is a new SOTA on the Vie dataset (avg F1 0.8665 vs 0.8284 for ST) and better cross-laboratory numbers on Bln and Bue. The paper also has a genuinely useful experiment where CD10, CD19, CD45 are removed from node features but kept in the graph, showing GNN-infused models can partly recover from missing markers.

What is actually new: FPS-based inducing points are, as far as I know, not in the Set Transformer literature, and the FCM-specific combination with a GNN layer is novel. The ablation with ST-150I (more inducing points does not help) is a good control. The study answers its three questions cleanly. Code is promised.

Soft spots, in order. First, no significance tests anywhere. The ST vs ST-FPS gap (0.8369 vs 0.8284) is within one standard deviation; the GIN-ST-FPS gap looks real, but with five runs a paired t-test would settle it. Second, the dataset split is by sample, not by patient. The paper says all data were collected on day 15 after induction, which might mean one sample per patient, but it never says so and it does not report patient counts. If the same patient appears in train and test, the headline numbers could be inflated. This is a request for clarification, not a proven flaw; the day-15 detail suggests the stress-test's 'high probability' claim may be too strong. Third, the 50k-event subsampling could hurt rare-cell detection when MRD is 0.01%, and the paper does not report sensitivity to that. This matters for the clinical framing but not for the relative comparison.

Overall, the central claim holds up as far as I can tell. The paper is honest, the experiments are reproducible in structure, and the improvements are plausible. I would send it to review. A careful referee should ask for significance tests and a patient-stratified split, or an explicit statement that each patient contributes exactly one sample. If the authors add those, this becomes a solid contribution.

Referee Report

4 major / 5 minor

Summary. The paper investigates whether local and global feature learning improves automated measurable residual disease (MRD) detection in flow cytometry data. It compares no-context models (MLP), global-context models (MLP-mean, MLP-max, MLP-pma, PointNet, Set Transformer and variants including reluFormer and the proposed ST-FPS), local-context GNN models (GCN, GAT, GIN, and ASAP variants), and two proposed hybrid models (GAT-ST-FPS and GIN-ST-FPS) that combine one GNN layer with a Set Transformer using FPS-sampled inducing points. On the Vie dataset, GIN-ST-FPS achieves an average F1 of 0.8665 ± 0.0083 versus 0.8284 ± 0.0117 for the previous ST baseline, and the paper reports improved inter-laboratory F1 on Bln and Bue. The code is made publicly available.

Significance. The study is potentially valuable for the flow cytometry deep learning community: it provides a broad, controlled comparison of architectures, reports means and standard deviations over multiple runs, uses public datasets, and includes useful ablations (removing attention, increasing inducing points, changing k-NN size). The central architectural insight, that sample-derived inducing points and explicit local graph context can complement global attention, is plausible and worth testing further. If the reported gains survive patient-level evaluation and significance testing, the paper would constitute a useful new state of the art for automated MRD detection. The public code release is a concrete strength.

major comments (4)

[3.3, Tables 5 and 6] The train/validation/test split is described only as a random 50/25/25 split by sample, not by patient. Vie consists of 519 bone-marrow samples from pediatric b-ALL patients, and such patients typically contribute multiple follow-up samples; the paper does not report the number of patients or state that same-patient samples are kept within one split. Since leukemic clones are patient-specific, a sample-level split can put the same patient's samples in both training and test, and the proposed model's reliance on sample-derived inducing points and local graph context may benefit disproportionately from this leakage. Please report patient counts, quantify the overlap across splits, and rerun the key comparisons in Tables 5 and 6 with a patient-stratified split.
[4.2 and 4.4, Tables 3 and 5] The headline comparisons are not accompanied by significance tests, and several margins are small relative to run-to-run variability. For example, ST-FPS versus ST is 0.8369 ± 0.0076 versus 0.8284 ± 0.0117, and GAT-ST-FPS versus GIN-ST-FPS is 0.8465 ± 0.0094 versus 0.8665 ± 0.0083 over at least five runs. If the same split indices are used across models, a paired test or confidence intervals on per-run differences should be reported; otherwise the claimed ordering of methods, and especially the conclusion that GIN is the right local layer, is not established.
[3.2, Figure 2] The proposed GIN-ST-FPS architecture is not specified sufficiently for reproduction. The text states that one GNN layer is used and its output is concatenated with the input feature vectors before three ST-FPS layers, but no equations or tensor-dimension description are given for the concatenation, for how the k-NN graph is constructed in the hybrid model, or for how the FPS-selected inducing points relate to the GNN output. Please provide a precise layer-by-layer specification.
[3.3, 4.4, Table 6] The cross-laboratory experiments are reported without stating how the feature dimensions are aligned across Vie, Bln, and Bue, whose marker panels differ. The text argues that graph construction can incorporate sample-specific panels, but ST-FPS and the combined models expect a fixed input feature dimension for the attention layers. It is therefore unclear which features were used as node features and how the models were evaluated on laboratories with different panels; this should be stated explicitly, including any marker matching, feature harmonization, or imputation.

minor comments (5)

[3.1, 4.1, Introduction] There are several typos: 'Preleminaries' should be 'Preliminaries', 'Tabel 2' should be 'Table 2', and 'cyotmetry' should be 'cytometry'.
[4.2] The sentence 'the performance increase of ST-FPS does stain from sampling feature vectors' should read 'does stem from'.
[4.3 and 4.4] The phrase 'spacial proximity' should be 'spatial proximity'.
[3.3] The paper reports 'med F1' throughout but never defines it; please state explicitly whether it is the median over samples or the median over runs.
[3.3] Table 1 is referenced before it appears; consider moving the table earlier or adjusting the reference.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation; all headline results are held-out empirical comparisons on public datasets.

full rationale

The paper's central claims are empirical: it trains a battery of models under a shared training protocol and reports F1 on held-out test splits of public flow-cytometry datasets (Tables 2-7). The proposed GIN-ST-FPS is a composition of standard, externally published building blocks (Set Transformer, GNN, farthest-point sampling), and its reported improvement over the ST baseline is measured on test data, not derived from the model definition or from the data-generation process. No equation in the paper defines the target metric as an input, and no fitted parameter is renamed as a prediction. The 'current SOTA' designation for ST leans on the authors' own prior work [33], but that baseline is re-implemented and re-evaluated under identical conditions in this paper, so the comparison is falsifiable rather than assumed. The FPS ratio r=0.0005, k=10, and the one-GNN-layer design were selected on development data, which creates a mild model-selection caveat, and the sample-level (not patient-level) 50/25/25 split in Section 3.3 is a potential leakage risk; these are correctness and interpretability concerns, not circularity. No load-bearing step reduces to its own inputs, so the derivation chain is self-contained.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The central empirical claim does not rest on a derivation, so the ledger is dominated by modeling choices and domain assumptions. The most consequential are the sample-level split, the k-NN graph construction, and the 50k-event subsampling. No new physical entities are introduced.

free parameters (4)

FPS sampling ratio r = 0.0005 (about 150 inducing points)
Controls the number of sample-derived inducing points in the ISABs; the value is chosen by the authors, and the paper only tests the induced-point count indirectly via ST-150I, without a sensitivity analysis of r.
k-NN neighborhood size k = 10
Defines the local graph used by the GNN layer; the paper shows k=3 impairs results (Table 4), indicating the result depends on this hand-set value.
Event subsample size per sample = 5e4
Every training and evaluation batch samples 50,000 events per sample; at 0.01% MRD this can contain only about five leukemic cells, yet no stability analysis is provided.
Model depth and width = 4 layers, hidden dim 32, 4 heads
Fixed across all models for comparability; hidden dimension is much smaller than PointNet's 1024, which the authors note as a confound in Table 3.

assumptions (5)

domain assumption Events within an FCM sample are not independent; relative positions of cell populations carry task-relevant information.
Stated in the Introduction and Section 3.1 as motivation for all global- and local-context models.
ad hoc to paper A k-NN graph with k=10 over all events in a sample correctly represents biologically meaningful local cell neighborhoods.
Used to construct graphs for all GNN models; no validation of graph construction against known cell populations is provided.
domain assumption Manual gating by at least two experts yields a correct ground truth.
Ground truth labels are derived from expert gating (Section 3.3); label noise is not modeled.
ad hoc to paper Randomly subsampling 5e4 events per sample preserves the rare leukemic population well enough for training and reliable evaluation.
Adopted silently in Training details; contradicts in spirit the stated MRD frequency as low as 0.01% (Section 3.1).
domain assumption Bln and Bue datasets are representative external laboratories for testing cross-laboratory generalization after training only on Vie.
Used in Section 4.4 and Table 6; differences in cytometer, panel, and patient population are treated as transfer challenges.

how reviews work

0 comments

Cite this review

Pith. "Pith review of On the importance of local and global feature learning for automated measurable residual disease detection in flow cytometry data." pith.science (2026). https://pith.science/paper/FAUFCCZ7

@misc{pith2026241115621,
  author       = {Pith},
  title        = {Pith review of: On the importance of local and global feature learning for automated measurable residual disease detection in flow cytometry data},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FAUFCCZ7}},
  note         = {Machine review of arXiv:2411.15621}
}

read the original abstract

This paper evaluates various deep learning methods for measurable residual disease (MRD) detection in flow cytometry (FCM) data, addressing questions regarding the benefits of modeling long-range dependencies, methods of obtaining global information, and the importance of learning local features. Based on our findings, we propose two adaptations to the current state-of-the-art (SOTA) model. Our contributions include an enhanced SOTA model, demonstrating superior performance on publicly available datasets and improved generalization across laboratories, as well as valuable insights for the FCM community, guiding future DL architecture designs for FCM data analysis. The code is available at \url{https://github.com/lisaweijler/flowNetworks}.

Figures

Figures reproduced from arXiv: 2411.15621 by the authors.

**Figure 2.** This figure shows the general architecture of the proposed local- and globalcontext model. For our experiments, we used one GNN layer and three ISAB with FPS instead of learned query vectors. The prediction head MLP is a linear layer. GAT-ST-FPS: The adapted SOTA architecture based on the findings in Section 4, GAT infused ST with FPS self-attention. We use one GAT layer for local feature extraction and concatenate… view at source ↗

**Figure 3.** Network features of the last layer of each model before the prediction head (a linear layer) are plotted using PCA with two components. Each model had the same FCM sample as input. Healthy cells are denoted in grey, and cancerous cells in red. mated MRD detection, where self-attention based on FPS performs best. Further, methods based solely on local feature learning can perform similarly and, in some cases, even o… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 30 canonical work pages

[1]

Cytometry Part A 95(7), 769–781 (2019)

Abdelaal, T., van Unen, V., Höllt, T., Koning, F., Reinders, M.J., Mahfouz, A.: Predicting cell populations in single cell mass cytometry data. Cytometry Part A 95(7), 769–781 (2019)

work page 2019
[2]

Nature Communications8(14825), 2041–1723 (2017)

Arvaniti,E.,Claassen,M.:Sensitivedetectionofraredisease-associatedcellsubsets via representation learning. Nature Communications8(14825), 2041–1723 (2017)

work page 2017
[3]

Nature biotechnology37(1), 38–44 (2019) Local and global feature learning for FCM data 15

Becht, E., McInnes, L., Healy, J., Dutertre, C.A., Kwok, I.W., Ng, L.G., Ginhoux, F., Newell, E.W.: Dimensionality reduction for visualizing single-cell data using umap. Nature biotechnology37(1), 38–44 (2019) Local and global feature learning for FCM data 15

work page 2019
[4]

Proceed- ings of the National Academy of Sciences111(26), E2770–E2777 (2014)

Bruggner, R.V., Bodenmiller, B., Dill, D.L., Tibshirani, R.J., Nolan, G.P.: Auto- mated identification of stratifying signatures in cellular subpopulations. Proceed- ings of the National Academy of Sciences111(26), E2770–E2777 (2014)

work page 2014
[5]

Hematol- ogy 2010, the American Society of Hematology Education Program Book2010(1), 7–12 (2010)

Campana, D.: Minimal residual disease in acute lymphoblastic leukemia. Hematol- ogy 2010, the American Society of Hematology Education Program Book2010(1), 7–12 (2010)

work page 2010
[6]

Cytometry Part A pp

Cheung, M., Campbell, J.J., Whitby, L., Thomas, R.J., Braybrook, J., Petzing, J.: Current trends in flow cytometry automated data analysis software. Cytometry Part A pp. 1–15 (2021)

work page 2021
[7]

Springer (2020)

Dash, S., Acharya, B.R., Mittal, M., Abraham, A., Kelemen, A.: Deep learning techniques for biomedical and health informatics. Springer (2020)

work page 2020
[8]

Cytometry Part B: Clinical Cytome- try: The Journal of the International Society for Analytical Cytology74(6), 331– 340 (2008)

Dworzak, M.N., Gaipa, G., Ratei, R., Veltroni, M., Schumich, A., Maglia, O., Karawajew, L., Benetello, A., Pötschger, U., Husak, Z., et al.: Standardization of flow cytometric minimal residual disease evaluation in acute lymphoblastic leukemia: Multicentric assessment is feasible. Cytometry Part B: Clinical Cytome- try: The Journal of the International So...

work page 2008

Show all 35 references

[9]

IEEE transactions on pattern analysis and machine intelligence 43(12), 4338–4364 (2020)

Guo, Y., Wang, H., Hu, Q., Liu, H., Liu, L., Bennamoun, M.: Deep learning for 3d point clouds: A survey. IEEE transactions on pattern analysis and machine intelligence 43(12), 4338–4364 (2020)

2020
[10]

Frontiers in immunology12, 787574 (2022)

Hu, Z., Bhattacharya, S., Butte, A.J.: Application of machine learning for cytom- etry data. Frontiers in immunology12, 787574 (2022)

2022
[11]

arXiv preprint arXiv:1609.02907 (2016)

Kipf, T.N., Welling, M.: Semi-supervised classification with graph convolutional networks. arXiv preprint arXiv:1609.02907 (2016)

2016 arXiv
[12]

Kowarsch,F.,Weijler,L.,Wödlinger,M.,Reiter,M.,Maurer-Granofszky,M.,Schu- mich, A., Sajaroff, E.O., Groeneveld-Krentz, S., Rossi, J.G., Karawajew, L., Ratei, R., Dworzak, M.N.: Towards self-explainable transformers for cell classification in flow cytometry data. In: Interpretab...

2022
[13]

Bioinformatics (Oxford, England) 33 (01 2017)

Lee, H.C., Kosoy, R., Becker, C., Kidd, B.: Automated cell type discovery and classification through knowledge transfer. Bioinformatics (Oxford, England) 33 (01 2017)

2017
[14]

In: Interna- tional Conference on Machine Learning

Lee, J., Lee, Y., Kim, J., Kosiorek, A., Choi, S., Teh, Y.W.: Set transformer: A framework for attention-based permutation-invariant neural networks. In: Interna- tional Conference on Machine Learning. pp. 3744–3753. PMLR (2019)

2019
[15]

Cell 162(1), 184—-197 (2015)

Levine, J.H., Simonds, E.F., Bendall, S.C., Davis, K.L., Amir, E.a.D., Tadmor, M.D., Litvin, O., Fienberg, H.G., Jager, A., Zunder, E.R., Finck, R., Gedman, A.L., Radtke, I., Downing, J.R., Pe’er, D., Nolan, G.P.: Data-driven phenotypic dissection of aml reveals progenitor-lik...

2015
[16]

Bioinformatics33(21), 3423–3430 (2017)

Li, H., Shaham, U., Stanton, K.P., Yao, Y., Montgomery, R.R., Kluger, Y.: Gating mass cytometry data by deep learning. Bioinformatics33(21), 3423–3430 (2017)

2017
[17]

In: 2018 24th International Conference on Pattern Recognition (ICPR)

Licandro, R., Schlegl, T., Reiter, M., Diem, M., Dworzak, M., Schumich, A., Langs, G., Kampel, M.: Wgan latent space embeddings for blast identification in child- hood acute myeloid leukaemia. In: 2018 24th International Conference on Pattern Recognition (ICPR). pp. 3868–3873....

2018
[18]

Current protocols in immunology 120(1), 5–1 (2018) 16 L

McKinnon, K.M.: Flow cytometry: an overview. Current protocols in immunology 120(1), 5–1 (2018) 16 L. Weijler et al

2018
[19]

Oncotarget7(44), 71915–71921 (2016)

Ni, W., Hu, B., Zheng, C., Tong, Y., Wang, L., Li, Q.q., Tong, X., Han, Y.: Auto- mated analysis of acute myeloid leukemia minimal residual disease using a support vector machine. Oncotarget7(44), 71915–71921 (2016)

2016
[20]

In: Proceedings of the IEEE conference on computer vision and pattern recognition

Qi, C.R., Su, H., Mo, K., Guibas, L.J.: Pointnet: Deep learning on point sets for 3d classification and segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 652–660 (2017)

2017
[21]

arXiv preprint arXiv:2202.08791 (2022)

Qin, Z., Sun, W., Deng, H., Li, D., Wei, Y., Lv, B., Yan, J., Kong, L., Zhong, Y.: cosformer: Rethinking softmax in attention. arXiv preprint arXiv:2202.08791 (2022)

2022 arXiv
[22]

Ranjan, E., Sanyal, S., Talukdar, P.: Asap: Adaptive structure aware pooling for learninghierarchicalgraphrepresentations.In:ProceedingsoftheAAAIconference on artificial intelligence. vol. 34, pp. 5470–5477 (2020)

2020
[23]

Cytometry Part A95(9), 966–975 (2019)

Reiter, M., Diem, M., Schumich, A., Maurer-Granofszky, M., Karawajew, L., Rossi, J.G., Ratei, R., Groeneveld-Krentz, S., Sajaroff, E.O., Suhendra, S., et al.: Automated flow cytometric mrd assessment in childhood acute b-lymphoblastic leukemia using supervised machine learning...

2019
[24]

a case study: Flow cytometry

Rota, P., Kleber, F., Reiter, M., Groeneveld-Krentz, S., Kampel, M.: The role of machine learning in medical data analysis. a case study: Flow cytometry. In: VISIGRAPP (3: VISAPP). pp. 305–312 (2016)

2016
[25]

In: CEUR WORK- SHOP PROCEEDINGS

Suffian, M., Montagna, S., Bogliolo, A., Ortolani, C., Papa, S., D’Atri, M., et al.: Machine learning for automated gating of flow cytometry data. In: CEUR WORK- SHOP PROCEEDINGS. vol. 3307, pp. 47–56. Sun SITE Central Europe, RWTH Aachen University (2022)

2022
[26]

European Journal of Cancer122, 61–71 (2019)

Testi, A.M., Attarbaschi, A., Valsecchi, M.G., Möricke, A., Cario, G., Niggli, F., Silvestri, D., Bader, P., Kuhlen, M., Parasole, R., et al.: Outcome of adolescent patients with acute lymphoblastic leukaemia aged 10–14 years as compared with those aged 15–17 years: Long-term ...

2019
[27]

In: Advances in neural information processing systems

Vaswani,A.,Shazeer,N.,Parmar,N.,Uszkoreit,J.,Jones,L.,Gomez,A.N.,Kaiser, Ł., Polosukhin, I.: Attention is all you need. In: Advances in neural information processing systems. pp. 5998–6008 (2017)

2017
[28]

In: International Conference on Learning Representations (2018)

Veličković, P., Cucurull, G., Casanova, A., Romero, A., Liò, P., Bengio, Y.: Graph attention networks. In: International Conference on Learning Representations (2018)

2018
[29]

Communica- tions Biology 2(183), 2399–3642 (2019)

Weber, L.M., Nowicka, M., Soneson, C., Robinson, M.D.: diffcyt: Differential dis- covery in high-dimensional cytometry via high-resolution clustering. Communica- tions Biology 2(183), 2399–3642 (2019)

2019
[30]

In: 2020 25th International Conference on Pattern Recognition (ICPR)

Weijler, L., Diem, M., Reiter, M., Maurer-Granofszky, M.: Detecting rare cell pop- ulations in flow cytometry data using umap. In: 2020 25th International Conference on Pattern Recognition (ICPR). pp. 4903–4909 (2021)

2021
[31]

In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision

Weijler, L., Kowarsch, F., Reiter, M., Hermosilla, P., Maurer-Granofszky, M., Dworzak, M.: Fate: Feature-agnostic transformer-based encoder for learning gener- alized embedding spaces in flow cytometry data. In: Proceedings of the IEEE/CVF Winter Conference on Applications of ...

2024
[32]

Cancers14(4) (2022)

Weijler,L.,Kowarsch,F.,Wödlinger,M.,Reiter,M.,Maurer-Granofszky,M.,Schu- mich, A., Dworzak, M.N.: Umap based anomaly detection for minimal residual disease quantification within acute myeloid leukemia. Cancers14(4) (2022)

2022
[33]

Computers in Biology and Medicine p

Wodlinger, M., Reiter, M., Weijler, L., Maurer-Granofszky, M., Schumich, A., Groeneveld-Krentz, S., Ratei, R., Karawajew, L., Sajaroff, E., Rossi, J., Dworzak, Local and global feature learning for FCM data 17 M.N.: Automated identification of cell populations in flow cytometr...

2022
[34]

Advances in Neural Information Processing Systems34, 13266–13279 (2021)

Wu,Z.,Jain,P.,Wright,M.,Mirhoseini,A.,Gonzalez,J.E.,Stoica,I.:Representing long-range context for graph neural networks with global attention. Advances in Neural Information Processing Systems34, 13266–13279 (2021)

2021
[35]

Xu, K., Hu, W., Leskovec, J., Jegelka, S.: How powerful are graph neural networks? In: International Conference on Learning Representations (2019)

2019

Pith tools

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

[1] [1]

Cytometry Part A 95(7), 769–781 (2019)

Abdelaal, T., van Unen, V., Höllt, T., Koning, F., Reinders, M.J., Mahfouz, A.: Predicting cell populations in single cell mass cytometry data. Cytometry Part A 95(7), 769–781 (2019)

work page 2019

[2] [2]

Nature Communications8(14825), 2041–1723 (2017)

Arvaniti,E.,Claassen,M.:Sensitivedetectionofraredisease-associatedcellsubsets via representation learning. Nature Communications8(14825), 2041–1723 (2017)

work page 2017

[3] [3]

Nature biotechnology37(1), 38–44 (2019) Local and global feature learning for FCM data 15

Becht, E., McInnes, L., Healy, J., Dutertre, C.A., Kwok, I.W., Ng, L.G., Ginhoux, F., Newell, E.W.: Dimensionality reduction for visualizing single-cell data using umap. Nature biotechnology37(1), 38–44 (2019) Local and global feature learning for FCM data 15

work page 2019

[4] [4]

Proceed- ings of the National Academy of Sciences111(26), E2770–E2777 (2014)

Bruggner, R.V., Bodenmiller, B., Dill, D.L., Tibshirani, R.J., Nolan, G.P.: Auto- mated identification of stratifying signatures in cellular subpopulations. Proceed- ings of the National Academy of Sciences111(26), E2770–E2777 (2014)

work page 2014

[5] [5]

Hematol- ogy 2010, the American Society of Hematology Education Program Book2010(1), 7–12 (2010)

Campana, D.: Minimal residual disease in acute lymphoblastic leukemia. Hematol- ogy 2010, the American Society of Hematology Education Program Book2010(1), 7–12 (2010)

work page 2010

[6] [6]

Cytometry Part A pp

Cheung, M., Campbell, J.J., Whitby, L., Thomas, R.J., Braybrook, J., Petzing, J.: Current trends in flow cytometry automated data analysis software. Cytometry Part A pp. 1–15 (2021)

work page 2021

[7] [7]

Springer (2020)

Dash, S., Acharya, B.R., Mittal, M., Abraham, A., Kelemen, A.: Deep learning techniques for biomedical and health informatics. Springer (2020)

work page 2020

[8] [8]

Cytometry Part B: Clinical Cytome- try: The Journal of the International Society for Analytical Cytology74(6), 331– 340 (2008)

Dworzak, M.N., Gaipa, G., Ratei, R., Veltroni, M., Schumich, A., Maglia, O., Karawajew, L., Benetello, A., Pötschger, U., Husak, Z., et al.: Standardization of flow cytometric minimal residual disease evaluation in acute lymphoblastic leukemia: Multicentric assessment is feasible. Cytometry Part B: Clinical Cytome- try: The Journal of the International So...

work page 2008