Pith. sign in

REVIEW 4 major objections 4 minor 55 references

SepsisCalc: Integrating Clinical Calculators into Early Sepsis Prediction via Dynamic Temporal Graph Construction

T0 review · 4 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read SepsisCalc claims that dynamically estimating clinical calculators, adding only the confident estimates to a temporal graph, and jointly predicting organ dysfunction and sepsis risk improves early sepsis prediction over existing baselines.

desk verdict Solid empirical paper with a genuinely new architecture, but the confidence-gating mechanism is trained in a different missingness regime than the one it operates in, so the core claim needs one more experiment. read the letter →

arxiv 2501.00190 v2 pith:RW2ZQOBG submitted 2024-12-31 cs.LG cs.AIcs.HC

Changchang Yin , Shihan Fu , Bingsheng Yao , Thai-Hoang Pham , Weidan Cao , Dakuo Wang , Jeffrey Caterino , Ping Zhang This is my paper · ORCID
classification cs.LGcs.AIcs.HC
keywords earlysepsispredictionclinicalcalculatorsSOFAscoretemporalheterogeneousgraphneuralnetworkelectronichealthrecordsmissingdataorgandysfunction
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 tries to establish that an early sepsis predictor should mirror the clinician's workflow rather than output one opaque risk score. SepsisCalc first builds a temporal heterogeneous graph of a patient's observed variables, then estimates the standard clinical calculators clinicians rely on (SOFA, qSOFA, SIRS, NEWS, MEWS, APACHE II), and dynamically adds only the estimates it is confident about as new graph nodes. From that augmented graph it predicts both the six organ-dysfunction risks and the overall sepsis risk. In experiments on two public ICU datasets and one proprietary hospital cohort, SepsisCalc reports AUCs of 0.839, 0.848 and 0.918 against best-baseline AUCs of 0.822, 0.828 and 0.894, and the ablations attribute part of the gain to the confidence-gated calculator nodes rather than to imputation. If the claim is right, sepsis warning systems could give clinicians organ-level evidence alongside the risk score, which is the information needed for timely intervention.

What carries the argument

The load-bearing mechanism is the dynamic temporal heterogeneous graph with four node types: collection nodes, observed clinical variables, organ nodes, and calculator nodes; and three edge types: directed edges linking successive observations of the same variable (with the elapsed time as an edge attribute), undirected edges connecting variables observed at the same timestamp, and directed clinical-event-interaction edges (for example, vasopressor use linked to low blood pressure). The calculator estimator reads the collection-level embedding and produces both an estimated calculator score and a confidence value $p^c_t$, and the confidence gate discards estimates with confidence below 0.5 before adding a calculator node. New edges are generated between the added calculator and its component variables. A temporal heterogeneous message-passing network with multi-head attention propagates information through this graph, and the total loss combines sepsis prediction, organ-dysfunction prediction, calculator estimation, and confidence-gating objectives.

What would settle it

Take a cohort where true SOFA can be recovered for many time points despite the usual high missingness, for example by chart review or by collecting the missing components on a random sample. Compare the squared error of SepsisCalc's estimated SOFA scores for the collections the confidence gate admits (confidence at least 0.5) with the error for collections it rejects. If the admitted estimates are not systematically more accurate, or if the confidence score does not rank errors at realistic missing rates, the gate that the ablation credits for the gains is not doing the work attributed to it and the reported improvement would not be expected to transfer to deployment.

Watch

Extended reading notes

Core claim

SepsisCalc claims that a dynamic temporal heterogeneous graph can carry the information that matters for sepsis prediction while staying faithful to how sepsis is actually diagnosed. The model estimates clinical calculators from the current collection-level representation, assigns each estimate a confidence value, and inserts the calculator as a node only when the confidence is at least 0.5. It then re-reads the augmented graph with a temporal heterogeneous message-passing network and jointly outputs a sepsis risk score and six organ-dysfunction risk scores. The paper's experiments show SepsisCalc outperforming the strongest baselines by roughly 0.017 to 0.024 area under the ROC curve across the three ICU cohorts, and the ablations show that removing the dynamic calculator construction or replacing it with imputation lowers performance. The intended payoff is a model whose predictions are tied to the same organ-failure assessments clinicians already use.

Load-bearing premise

The confidence classifier that decides which estimated calculators enter the graph is trained only on the rare collections where every calculator component is observed (fewer than 6% of SOFA collections), and the paper assumes this low-missingness training transfers to deployment, where those components are missing about 94% of the time.

Editorial extensions

If this is right

  • If the claim holds, an early-warning system can report a sepsis risk score and the six organ-dysfunction risks from the same model, so clinicians see which organ signals are driving the risk.
  • The graph representation removes the need to impute missing lab values: only observed variables and confidently estimated calculators enter the model, which should reduce imputation bias in high-missing-rate settings.
  • The confidence gate formalizes when a calculator like SOFA is trustworthy: when its components are too sparse, the model withholds the score instead of guessing it.
  • Because calculator definitions enter only through component variables and supervision, adding new calculators or organ scores would not require redesigning the architecture.
  • Deployed in an EHR system, the model's organ-specific outputs can be shown alongside the estimated SOFA score, letting clinicians prepare interventions for the organ that is failing.

Reading between the lines

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

  • Beyond the reported experiments, the central risk to the claim is deployment shift: the confidence gate is trained on the rare fully observed collections, and its behavior when the graph is fed mostly missing SOFA components is not measured; a direct calibration test under realistic missingness would settle whether the reported gains survive.
  • Read as a missing-data strategy, SepsisCalc is a task-aware alternative to imputation: it reconstructs the clinical summary clinicians would compute rather than the underlying lab values, which may be a more stable target when labs are missing not at random.
  • The same dynamic-node recipe extends naturally beyond sepsis calculators to other composite severity scores, since the estimator learns their computation rules from observed components and then generalizes to partial observations.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 4 minor

Summary. This paper proposes SepsisCalc, an early sepsis prediction model that converts EHRs into temporal heterogeneous graphs, estimates clinical calculators (SOFA, qSOFA, NEWS, and related organ-specific scores) from the observed graph, and dynamically adds only those estimated calculator nodes whose confidence score p_c exceeds 0.5 before a second round of temporal heterogeneous message passing. The model is trained with a weighted loss combining sepsis risk, organ dysfunction risk, calculator score estimation, and confidence gating. Across MIMIC-III, AmsterdamUMCdb, and OSUWMC, SepsisCalc reports AUC 0.839/0.848/0.918 versus best baseline AUCs 0.822/0.828/0.894, with ablations showing contributions from event interactions, dynamic calculator nodes, and the organ-dysfunction auxiliary task. The paper also describes a deployed Epic EHR interface for organ dysfunction and sepsis risk review.

Significance. The contribution is potentially valuable: injecting clinically validated calculators as explicit graph nodes is a plausible way to combine clinical workflow knowledge with learned representations, and the multi-dataset evaluation with ablations, code release, and mask-based calculator estimation experiments is a strength. If the confidence-gating mechanism transfers to real high-missingness settings, the reported gains are meaningful and the system is clinically relevant. However, the central transfer assumption is not yet validated, and the organ-dysfunction label definition is missing; these are load-bearing gaps that can be addressed with additional experiments.

major comments (4)
  1. [§3.4.2, Eq. (10), Table 8] Eq. (10) trains the confidence head only on collections with M_{t,i}=1, i.e., where the ground-truth calculator is computable because all component variables are observed. Table 8 shows that for SOFA this holds in only about 5–6% of collections on all three datasets, so the confidence classifier is trained almost entirely in the near-complete-observation regime, while deployment is dominated by the 94–95% missingness regime. The input representation h_L_t in Eq. (7) has systematically fewer variable nodes and edges when components are missing, so the feature distribution at the gating head differs between training and deployment. Since p_c determines which calculator nodes enter the dynamic graph, an unvalidated transfer assumption is load-bearing for the paper's central claim. Please report calibration curves for p_c on observed versus missing collections, the distribution and inclusion rate of calculator nodes under natural missingness, and a version of the gating head trained with masking augmentation or a missingness-robust representation.
  2. [§5.2.2, Fig. 7, Table 8] The mask-observation evaluation randomly masks 70% of calculator component variables. This is MCAR and does not reproduce the empirical missingness pattern in Table 8, which ranges from 23% (urine output) to 99% (Bands, C-reactive protein) and is likely informative because test ordering depends on acuity. The claim that SepsisCalc naturally handles missing values in deployment therefore rests on an untested equivalence between random masking and real missingness. Please evaluate calculator estimation and gating under an empirical mask distribution matched to per-variable missing rates, and, if possible, use naturally missing collections as a held-out test set.
  3. [Eq. (10), §3.4.2] The binary confidence label y_c = I[(e_c - ehat_c)^2 < 0.01] is an exact-match criterion for integer-valued calculators such as SOFA: an estimated score that differs by 1 point is labeled inaccurate even though one-point SOFA error is clinically tolerable. The fixed inclusion threshold 0.5 in Eq. (9) is also presented without sensitivity analysis. Please add a sweep over the threshold and consider a tolerance-aware label (e.g., absolute error <= 1) so that 'accurately estimated' matches clinical accuracy rather than exact equality.
  4. [§3.5, Eq. (12), Table 9] The organ dysfunction auxiliary task is part of the method and of the SepsisCalc-o ablation, but the paper never defines how the ground-truth organ dysfunction labels y_o,i_t are generated from the EHR data (e.g., which SOFA component threshold or charting source is used). Without this definition, Table 9 and the multi-task loss are not reproducible. Please state the label construction explicitly.
minor comments (4)
  1. [§4.3 / A.4.4] The '10-fold cross-validation' description says 7 sets are used for training, 1 for validation, and 2 for testing; this is a 70/10/20 split repeated 10 times rather than standard 10-fold CV. Please clarify the protocol and what the standard deviations in Table 2 represent.
  2. [Eq. (12)] The sepsis loss as printed is '-y_t log(p_t) - (1-y_t) log p_t'; the second term should be log(1-p_t).
  3. [§3.4.2 heading] Typo: 'Constrution' should be 'Construction'; also several affiliation and author strings misspell 'Northeastern' as 'Northestern'.
  4. [Algorithm 1] Algorithm 1 is described as returning h^L_t and h^{L,i}_t, but the pseudocode does not show how the organ node features are separated from the collection node features; a sentence or an explicit line in the pseudocode would improve reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central claim is an empirical benchmark against external sepsis labels; auxiliary calculator and confidence modules are supervised by external targets.

full rationale

The paper's central claim is an empirical outcome comparison: SepsisCalc's sepsis risk AUC/F1/Recall are measured against held-out sepsis-3 labels and compared with baseline models (Table 2). The calculator estimator is trained with MSE (Eq. 8) toward deterministic calculator ground truths, which are external to the sepsis label; this is a fitted auxiliary task, not a prediction that reduces to its input. The confidence-gating module in Eqs. 9-10 uses a label y^c defined from the estimator's own error, which is a self-referential supervision signal for the gating head, but it does not define the sepsis prediction target and does not force the reported performance differences, which are evaluated on external labels. The dynamic graph construction does not import a uniqueness theorem or a load-bearing self-citation: self-citations [47,49] are used only for value/time embedding techniques. The identified weakness (the confidence module is trained almost entirely on near-complete calculator observations, Table 8, while deployment has high missingness) is a generalization/calibration risk, not a circularity. No derivation step equivalently reduces to its own inputs by construction.

Assumptions & free parameters 6 free parameters · 7 assumptions · 0 invented entities

All quantities the central claim depends on are standard ML hyperparameters plus domain assumptions inherited from sepsis definitions and calculator validity. The most fragile item is the implicit transfer of the confidence classifier from fully observed training examples to a deployment setting with more than 90% missingness; this is an ad hoc assumption of the dynamic graph module. No new physical or causal entities are introduced.

free parameters (6)
  • Calculator confidence threshold = 0.5
    Equation 9 filters calculator nodes with p_c < 0.5; no sensitivity analysis is reported for this threshold.
  • Calculator accuracy threshold = 0.01
    Equation 10 defines y_c = I[(e_c - ehat_c)^2 < 0.01], used as the ground truth for confidence training.
  • Loss weights alpha_o, alpha_e, alpha_d = 1.0 after grid search over 0.1 to 10
    Appendix A.5.3 selected alpha* = 1 because performance was stable for 0.3 <= alpha* <= 3.
  • Embedding dimension d = 512
    Implementation detail in Section 4.3; central to model capacity.
  • Number of discretization bins n for value embedding = 1000
    Appendix A.3.2 uses n = 1000 sub-ranges to map observed values into embeddings.
  • GNN depth L and attention heads h = not reported in the paper
    Algorithm 1 iterates over L layers and h heads, but their values are not given in the visible text, affecting architecture and reproducibility.
assumptions (7)
  • domain assumption Sepsis-3 criteria and the clinician-defined 'suspected infection' cohort define ground-truth sepsis labels.
    Used in Section 4.1 to split cases and controls; label noise in EHR-based Sepsis-3 phenotyping propagates to AUC estimates.
  • domain assumption Clinical calculators such as SOFA, qSOFA, NEWS, and MEWS are valid, evidence-based organ-dysfunction assessments whose inclusion should improve prediction.
    Central motivation in Sections 1 and 5.1; the method's value depends on calculators encoding clinically meaningful information.
  • domain assumption A temporal graph containing only observed variables can represent EHRs without imputation, and missingness does not need to be modeled explicitly.
    Sections 2 and A.3.1 claim graph representation avoids imputation bias; this assumes missing-at-random or ignorable missingness for prediction.
  • ad hoc to paper The calculator-confidence classifier trained on fully observed collections transfers to the high-missingness deployment setting.
    Equations 8 to 10 train only when calculator ground truth is available, fewer than 6% of SOFA collections per Table 8, yet inference gates calculators under more than 90% missingness.
  • ad hoc to paper Synthetic random masking of 70% of calculator components mimics real-world missingness patterns.
    Section 5.2.2 uses random masks; clinical missingness is driven by ordering behavior and severity, which is likely informative.
  • domain assumption Hand-defined clinical event interaction edges, such as vasopressor-MAP, antibiotic-WBC, and ventilation-SpO2, capture clinically meaningful dependencies.
    Appendix A.4.6 adds these edges following the Surviving Sepsis Campaign; if edges are noisy, message passing could propagate spurious information.
  • standard math Standard deep-learning building blocks such as attention, softmax, backpropagation, and Adam behave as expected.
    Used throughout Section 3; no formal correctness proof is needed or provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of SepsisCalc: Integrating Clinical Calculators into Early Sepsis Prediction via Dynamic Temporal Graph Construction." pith.science (2026). https://pith.science/paper/RW2ZQOBG

@misc{pith2026250100190,
  author       = {Pith},
  title        = {Pith review of: SepsisCalc: Integrating Clinical Calculators into Early Sepsis Prediction via Dynamic Temporal Graph Construction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RW2ZQOBG}},
  note         = {Machine review of arXiv:2501.00190}
}
read the original abstract

Sepsis is an organ dysfunction caused by a deregulated immune response to an infection. Early sepsis prediction and identification allow for timely intervention, leading to improved clinical outcomes. Clinical calculators (e.g., the six-organ dysfunction assessment of SOFA) play a vital role in sepsis identification within clinicians' workflow, providing evidence-based risk assessments essential for sepsis diagnosis. However, artificial intelligence (AI) sepsis prediction models typically generate a single sepsis risk score without incorporating clinical calculators for assessing organ dysfunctions, making the models less convincing and transparent to clinicians. To bridge the gap, we propose to mimic clinicians' workflow with a novel framework SepsisCalc to integrate clinical calculators into the predictive model, yielding a clinically transparent and precise model for utilization in clinical settings. Practically, clinical calculators usually combine information from multiple component variables in Electronic Health Records (EHR), and might not be applicable when the variables are (partially) missing. We mitigate this issue by representing EHRs as temporal graphs and integrating a learning module to dynamically add the accurately estimated calculator to the graphs. Experimental results on real-world datasets show that the proposed model outperforms state-of-the-art methods on sepsis prediction tasks. Moreover, we developed a system to identify organ dysfunctions and potential sepsis risks, providing a human-AI interaction tool for deployment, which can help clinicians understand the prediction outputs and prepare timely interventions for the corresponding dysfunctions, paving the way for actionable clinical decision-making support for early intervention.

Figures

Figures reproduced from arXiv: 2501.00190 by the authors.

Figure 1
Figure 1. Workflows of clinicians and AI for sepsis identifica [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 3
Figure 3. Framework of SepsisCalc. (A) Dynamic temporal graph construction. (B) Sepsis prediction framework. methods achieved superior performance, they face a critical chal￾lenge due to the data representation. The models need to com￾pletely observe a list of variables (including vital signs and lab tests), while many variables are missing in real-world data. Existing studies [15, 18, 49] usually impute the missing values be… view at source ↗
Figure 4
Figure 4. Setting of sepsis onset prediction. 4 EXPERIMENT To demonstrate the effectiveness of the proposed SepsisCalc, we conducted extensive experiments on multiple real-world datasets. 4.1 Datasets and Setup Datasets. We validated our model on two publicly available datasets (MIMIC-III1 and AmsterdamUMCdb2 ) and one proprietary dataset extracted from OSUWMC3 . We first extracted all the sepsis pa￾tients with suspected infe… view at source ↗
Figures from the paper (5 more)
Figure 5
Figure 5. Figure 5: Sepsis risk prediction performance in both full and [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: Average alert time before sepsis and recall. [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]
Figure 8
Figure 8. Figure 8: Calculator estimation error over confidence levels. [PITH_FULL_IMAGE:figures/full_fig_p007_8.png]
Figure 9
Figure 9. Figure 9: User Interface of SepsisCalc System. (A) Patient list with sepsis risk score. (B) Demographic information. (C) Overall SOFA scores. (D) Organ-specific SOFA score. (E) Vital signs and lab test results related to the specific organ. temporal graphs and effectively extrac…
Figure 10
Figure 10. Figure 10: (A) List view of clinical variables for organ status. [PITH_FULL_IMAGE:figures/full_fig_p014_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

55 extracted references · 51 canonical work pages

  1. [1]

    Poushali Bhattacharjee, Dana P Edelson, and Matthew M Churpek. 2017. Identi- fying patients with sepsis on the hospital wards. Chest 151, 4 (2017), 898–907

    work page 2017

  2. [2]

    Roger C Bone, Robert A Balk, Frank B Cerra, R Phillip Dellinger, Alan M Fein, William A Knaus, Roland MH Schein, and William J Sibbald. 1992. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Chest 101, 6 (1992), 1644–1655

    work page 1992

  3. [3]

    Kyunghyun Cho, Bart van Merrienboer, Dzmitry Bahdanau, and Yoshua Ben- gio. 2014. On the Properties of Neural Machine Translation: Encoder-Decoder Approaches

    work page 2014

  4. [4]

    Edward Choi, Mohammad Taha Bahadori, Le Song, et al. 2017. GRAM: Graph- based Attention Model for Healthcare Representation Learning. In Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

    work page 2017

  5. [5]

    Edward Choi, Mohammad Taha Bahadori, Jimeng Sun, et al. 2016. RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechanism. In Advances in Neural Information Processing Systems 29: Annual Conference on Neural Information Processing Systems

    work page 2016

  6. [6]

    Shaika Chowdhury, Yongbin Chen, Andrew Wen, Xiao Ma, Qiying Dai, Yue Yu, Sunyang Fu, Xiaoqian Jiang, and Nansu Zong. 2023. Predicting physiological response in heart failure management: A graph representation learning approach using electronic health records. medRxiv (2023)

    work page 2023

  7. [7]

    Maia Dorsett, Melissa Kroll, Clark S Smith, Phillip Asaro, Stephen Y Liang, and Hawnwan P Moy. 2017. qSOFA has poor sensitivity for prehospital identification of severe sepsis and septic shock.Prehospital emergency care 21, 4 (2017), 489–497

    work page 2017

  8. [8]

    Yongrui Duan, Jiazhen Huo, Mingzhou Chen, et al . 2023. Early prediction of sepsis using double fusion of deep features and handcrafted features. Applied Intelligence 53, 14 (2023), 17903–17919

    work page 2023

Show all 55 references
  1. [9]

    Mikhail A Dziadzko, Ognjen Gajic, Brian W Pickering, and Vitaly Herasevich

  2. [10]

    Evangelos J Giamarellos-Bourboulis, Thomas Tsaganos, I Tsangaris, M Lada, C Routsi, D Sinapidis, M Koupetori, M Bristianou, G Adamis, K Mandragos, et al

  3. [11]

    Tim A Green, Stevan Whitt, Jeffery L Belden, Sanda Erdelez, and Chi-Ren Shyu

  4. [12]

    KDIGO Group et al. 2012. KDIGO clinical practice guideline for acute kidney injury. Kidney Int. Suppl. 2 (2012), 1. Conference acronym ’XX, June 03–05, 2018, Woodstock, NY Yin et al

    2012

  5. [13]

    Yuanfang Guan, Hongyang Li, Daiyao Yi, Dongdong Zhang, Changchang Yin, Keyu Li, and Ping Zhang. 2021. A survival model generalized to regression learning algorithms. Nature computational science 1, 6 (2021), 433–440

    2021

  6. [14]

    Sepp Hochreiter and Jürgen Schmidhuber. 1997. Long Short-Term Memory. Neural Computation 8 (1997)

    1997

  7. [15]

    Md Mohaimenul Islam, Tahmina Nasrin, Bruno Andreas Walther, Chieh-Chen Wu, Hsuan-Chia Yang, and Yu-Chuan Li. 2019. Prediction of sepsis patients using machine learning approach: a meta-analysis. Computer methods and programs in biomedicine 170 (2019), 1–9

    2019

  8. [16]

    Qiao Jin, Zhizheng Wang, et al. 2024. AgentMD: Empowering Language Agents for Risk Prediction with Large-Scale Clinical Tool Learning. arXiv preprint arXiv:2402.13225 (2024)

    2024 arXiv

  9. [17]

    Johnson, Tom J

    Alistair E.W. Johnson, Tom J. Pollard, Lu Shen, et al. 2016. MIMIC-III, a freely accessible critical care database. (2016)

    2016

  10. [18]

    Sundreen Asad Kamal, Changchang Yin, Buyue Qian, and Ping Zhang. 2020. An interpretable risk prediction model for healthcare with pattern attention. BMC Medical Informatics and Decision Making 20 (2020), 1–10

    2020

  11. [19]

    Patrick S Kamath and W Ray Kim. 2007. The model for end-stage liver disease (MELD). Hepatology 45, 3 (2007), 797–805

    2007

  12. [20]

    William A Knaus, Elizabeth A Draper, Douglas P Wagner, and Jack E Zimmerman

  13. [21]

    Matthieu Komorowski, Leo A Celi, Omar Badawi, Anthony C Gordon, and A Aldo Faisal. 2018. The artificial intelligence clinician learns optimal treatment strategies for sepsis in intensive care. Nature medicine 24, 11 (2018), 1716–1720

    2018

  14. [22]

    Mitchell M Levy, Laura E Evans, and Andrew Rhodes. 2018. The surviving sepsis campaign bundle: 2018 update. Intensive care medicine 44 (2018), 925–928

    2018

  15. [23]

    Sicen Liu, Tao Li, Haoyang Ding, Buzhou Tang, Xiaolong Wang, Qingcai Chen, Jun Yan, and Yi Zhou. 2020. A hybrid method of recurrent neural network and graph neural network for next-period prescription prediction. International Journal of Machine Learning and Cybernetics 11 (20...

    2020

  16. [24]

    Escobar, et al

    Vincent Liu, Gabriel J. Escobar, et al . 2014. Hospital Deaths in Patients With Sepsis From 2 Independent Cohorts. JAMA 312, 1 (07 2014), 90–92

    2014

  17. [25]

    Vincent X Liu, Vikram Fielding-Singh, John D Greene, Jennifer M Baker, Theodore J Iwashyna, Jay Bhattacharya, and Gabriel J Escobar. 2017. The tim- ing of early antibiotics and hospital mortality in sepsis. American journal of respiratory and critical care medicine 196, 7 (201...

    2017

  18. [26]

    Chang Lu, Chandan Reddy, Prithwish Chakraborty, Samantha Kleinberg, and Yue Ning. 2021. Collaborative Graph Learning with Auxiliary Text for Temporal Event Prediction in Healthcare. In IJCAI. 3529–3535

    2021

  19. [27]

    Yuan Luo, Peter Szolovits, Anand Dighe, and Jason Baron. 2018. 3D-MICE: integration of cross-sectional and longitudinal imputation for multi-analyte longitudinal clinical data. JAMIA 25, 6 (2018), 645–653

    2018

  20. [28]

    Fenglong Ma, Radha Chitta, Jing Zhou, Quanzeng You, Tong Sun, and Jing Gao

  21. [29]

    Fenglong Ma, Quanzeng You, et al. 2018. KAME: Knowledge-based Attention Model for Diagnosis Prediction in Healthcare. In Proceedings of the 27th ACM International Conference on Information and Knowledge Management, CIKM

    2018

  22. [30]

    Konstantinos Makris and Loukia Spanou. 2016. Acute kidney injury: diagnostic approaches and controversies. The Clinical Biochemist Reviews 37, 4 (2016), 153

    2016

  23. [31]

    Paul E Marik and Abdalsamih M Taeb. 2017. SIRS, qSOFA and new sepsis definition. Journal of thoracic disease 9, 4 (2017), 943

    2017

  24. [32]

    John G O’Grady, Graeme JM Alexander, Karen M Hayllar, and Roger Williams

  25. [33]

    In Proceedings of the 23rd ACM SIGKDD international conference on knowledge discovery and data mining

    Dipole: Diagnosis prediction in healthcare via attention-based bidirectional recurrent neural networks. In Proceedings of the 23rd ACM SIGKDD international conference on knowledge discovery and data mining . 1903–1911

    1903

  26. [34]

    Xia Qiu, Yu-Peng Lei, and Rui-Xi Zhou. 2023. SIRS, SOFA, qSOFA, and NEWS in the diagnosis of sepsis and prediction of adverse outcomes: a systematic review and meta-analysis. Expert review of anti-infective therapy 21, 8 (2023), 891–900

    2023

  27. [35]

    Matthew A Reyna, Christopher S Josef, et al. 2019. Early prediction of sepsis from clinical data: the PhysioNet/Computing in Cardiology Challenge 2019. Critical Care Medicine (2019)

    2019

  28. [36]

    Mervyn Singer, Clifford S Deutschman, et al . 2016. The third international consensus definitions for sepsis and septic shock (Sepsis-3). Jama 315, 8 (2016), 801–810

    2016

  29. [37]

    Gary B Smith, David R Prytherch, Paul Meredith, Paul E Schmidt, and Peter I Featherstone. 2013. The ability of the National Early Warning Score (NEWS) to discriminate patients at risk of early cardiac arrest, unanticipated intensive care unit admission, and death. Resuscitatio...

    2013

  30. [38]

    Christian P Subbe, M Kruger, et al. 2001. Validation of a modified Early Warning Score in medical admissions. Qjm 94, 10 (2001), 521–526

    2001

  31. [39]

    Carly J Paoli, Mark A Reynolds, et al. 2018. Epidemiology and costs of sepsis in the United States—an analysis based on timing of diagnosis and severity level. Critical care medicine 46, 12 (2018), 1889

    2018

  32. [40]

    Patrick Thoral, Jan Peppink, Ronald Driessen, et al. 2020. AmsterdamUMCdb: The First Freely Accessible European Intensive Care Database from the ESICM Data Sharing Initiative. (2020). https://doi.org/10.1109/JBHI.2020.2995139 access: https://www.amsterdammedicaldatascience.nl

    2020

  33. [41]

    Andrea Tsoris and Clinton A Marlar. 2019. Use of the Child Pugh score in liver disease. (2019)

    2019

  34. [42]

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need. In Advances in neural information processing systems . 5998–6008

    2017

  35. [43]

    J L Vincent, Rui Moreno, Jukka Takala, Sheila Willatts, Arnaldo De Mendonça, Hajo Bruining, CK Reinhart, PeterM Suter, and Lambertius G Thijs. 1996. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunc- tion/failure: On behalf of the Working Group...

    1996

  36. [44]

    Hideo Wada, Takeshi Matsumoto, and Yoshiki Yamashita. 2014. Diagnosis and treatment of disseminated intravascular coagulation (DIC) according to four DIC guidelines. Journal of Intensive Care 2 (2014), 1–8

    2014

  37. [45]

    Carlos Sánchez, Orlando Pérez-Nieto, and Eder Zamarrón. 2023. Chapter 16 - Mechanical Ventilation in Sepsis. In The Sepsis Codex , Marcio Borges, Jorge Hidalgo, and Javier Perez-Fernandez (Eds.). Elsevier, 135–138. https://doi.org/10. 1016/B978-0-323-88271-2.00009-2

    2023

  38. [46]

    Kai Yang, Yongxin Xu, Peinie Zou, Hongxin Ding, Junfeng Zhao, Yasha Wang, and Bing Xie. 2023. KerPrint: Local-Global Knowledge Graph Enhanced Diagnosis Prediction for Retrospective and Prospective Interpretations. AAAI 37, 4 (2023)

    2023

  39. [47]

    Changchang Yin, Ruoqi Liu, Dongdong Zhang, and Ping Zhang. 2020. Identifying sepsis subphenotypes via time-aware multi-modal auto-encoder. In Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery & data mining. 862–872

    2020

  40. [48]

    Changchang Yin, Rongjian Zhao, Buyue Qian, Xin Lv, and Ping Zhang. 2019. Domain Knowledge guided deep learning with electronic health records. In 2019 IEEE International Conference on Data Mining (ICDM) . IEEE, 738–747

    2019

  41. [49]

    Dongdong Zhang, Changchang Yin, Katherine M Hunold, Xiaoqian Jiang, Jef- frey M Caterino, and Ping Zhang. 2021. An interpretable deep-learning model for early prediction of sepsis in the emergency department. Patterns 2, 2 (2021). A APPENDIX A.1 Important Notations We summariz...

    2021

  42. [51]

    Chao Yan, Cheng Gao, Xinmeng Zhang, et al. 2019. Deep Imputation of Temporal Data. In 2019 IEEE International Conference on Healthcare Informatics, ICHI 2019, Xi’an, China, June 10-13, 2019 . 1–3

    2019

  43. [1985]

    APACHE II: a severity of disease classification system.Critical care medicine 13, 10 (1985), 818–829

    1985

  44. [1989]

    Gastroenterology 97, 2 (1989), 439–445

    Early indicators of prognosis in fulminant hepatic failure. Gastroenterology 97, 2 (1989), 439–445

    1989

  45. [2016]

    Computer methods and programs in biomedicine 133 (2016), 1–6

    Clinical calculators in hospital medicine: availability, classification, and needs. Computer methods and programs in biomedicine 133 (2016), 1–6

    2016

  46. [2017]

    Clinical Microbiology and Infection 23, 2 (2017), 104–109

    Validation of the new Sepsis-3 definitions: proposal for improvement in early risk identification. Clinical Microbiology and Infection 23, 2 (2017), 104–109

    2017

  47. [2019]

    Computer Methods and Programs in Biomedicine 179 (2019), 105002

    Medical calculators: prevalence, and barriers to use. Computer Methods and Programs in Biomedicine 179 (2019), 105002

    2019

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.