Pith. sign in

REVIEW 3 major objections 6 minor 204 references

Federated deep survival models trained across two unmatched Dutch cohorts beat local-only training on cardiovascular risk without sharing patient records.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · grok-4.5

2026-07-10 04:36 UTC pith:6RKLIRUL

load-bearing objection Real two-cohort federated DeepSurv result with honest modest gains; the Rotterdam improvement is directionally consistent but statistically soft under size-weighted FedAvg and overlapping CIs. the 3 major comments →

arxiv 2607.08595 v1 pith:6RKLIRUL submitted 2026-07-09 cs.LG cs.HC

Federated Deep Learning for Privacy-Preserving Cardiovascular Disease Risk Prediction

Hyunho Mo , Djura Smits , Mahlet A. Birhanu , Maarten J.G. Leening , Daniel Bos , Pim van der Harst , Esther E. Bron This is my paper
classification cs.LG cs.HC
keywords Federated learningCardiovascular diseaseHealthcare AIDeep survival neural networksLifelinesRotterdam StudyDeepSurvFedAvg
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

Cardiovascular risk tools are usually built on one site's data or on pooled individual records, both of which run into privacy barriers when institutions want to collaborate. This paper shows that a deep survival network can be trained jointly on two real population cohorts that never exchange raw data: the large Lifelines study with self-reported events and the smaller Rotterdam Study with digitally linked clinical outcomes. After twenty rounds of federated averaging, discrimination rose in both cohorts, with the clearer gain on the Rotterdam side that has complete follow-up. The result matters because it suggests a practical route to larger, more generalizable risk models while leaving patient-level data behind institutional walls.

Core claim

DeepSurv models trained with Federated Averaging across Lifelines (n=148,230, self-reported outcomes) and the Rotterdam Study (n=10,155, digitally linked outcomes) achieve higher C-statistics than identical models trained only on local data: Rotterdam rises from 0.728 to 0.739 and Lifelines from 0.783 to 0.787, with primary evaluation on Rotterdam because of its complete follow-up.

What carries the argument

Horizontal Federated Averaging (FedAvg) of DeepSurv weights: each site trains the same [12,16,16,1] network for 100 local epochs, then a central aggregator computes a sample-size-weighted average of the weights and redistributes it for the next of 20 rounds; raw records never leave their sites.

Load-bearing premise

That Lifelines' noisier, self-reported events with midpoint-interval dates still carry enough signal under sample-size-weighted averaging to improve a model judged on Rotterdam's precise clinical outcomes, despite large gaps in age, event rate, and outcome definition.

What would settle it

Re-run the same 10-split protocol after deliberately dropping or randomly flipping a substantial fraction of Lifelines event labels and check whether the Rotterdam C-statistic still rises above its local baseline of 0.728; if the gain disappears, the claim that the large noisy cohort helps the precise one fails.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Institutions that cannot pool patient records can still improve local CVD risk models by exchanging only model weights under FedAvg.
  • A large cohort with imperfect outcome ascertainment can still raise discrimination on a smaller cohort that has gold-standard follow-up.
  • Convergence must be monitored site-by-site; sample-size-weighted averaging can produce non-monotonic trajectories when cohort sizes differ by an order of magnitude.
  • Existing GDPR-compliant cloud workspaces and joint-controller agreements already support this style of multi-cohort deep survival training.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Adaptive early-stopping or quality-aware reweighting of client updates would likely reduce the late-training dip seen on the smaller, higher-quality node.
  • The same harmonization-plus-FedAvg pipeline should transfer to other time-to-event endpoints (cancer, dementia) that face identical privacy and label-noise tensions.
  • If competing-risk models replace simple censoring of non-CVD death, absolute risk calibration in older Rotterdam participants would be the natural next stress test.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 6 minor

Summary. The manuscript evaluates federated training of DeepSurv (FedAvg over 20 rounds on Vantage6) across two heterogeneous Dutch cohorts—Lifelines (N=148,230; self-reported CVD with midpoint event times) and the Rotterdam Study (N=10,155; digitally linked clinical outcomes)—without sharing individual-level data. After FHIR-based harmonization of 12 standard CVD predictors, local vs federated discrimination is compared on held-out stratified splits (10 runs, corrected resampled t-test CIs). The primary claim is that federation raises the C-statistic relative to local training: Rotterdam 0.728 (0.717–0.739) → 0.739 (0.728–0.749); Lifelines 0.783 (0.775–0.791) → 0.787 (0.780–0.792), with Rotterdam as the primary evaluation cohort because of complete follow-up.

Significance. If the performance gain is real and transferable, the work is a useful real-world demonstration that federated deep survival models can improve CVD risk prediction across cohorts that differ in size, age, event rate, and outcome ascertainment, while remaining within existing GDPR/governance constraints. Strengths include a genuine multi-site (not pseudo-split) setup, a documented HL7 FHIR harmonization pipeline, fixed architecture/seed, 10 stratified runs with corrected resampled CIs, primary evaluation on the better-labeled cohort, and an unusually candid discussion of size imbalance and label noise. The contribution is incremental relative to the prior MyDigiTwin infrastructure paper, but the multi-cohort evaluation is the right next step for the field.

major comments (3)
  1. [Table 2 / Results §3] Table 2 and §3: The central claim that federation improves discrimination rests on point-estimate increases whose 95% CIs fully overlap (Rotterdam federated point equals the local upper bound; local point equals the federated lower bound). Only separate CIs for each setting are reported; no paired or corrected-resampled test of the federation-minus-local difference is given. Without a formal test of ΔC (or equivalent), the primary claim remains statistically under-supported and should be either tested or substantially softened in the abstract and conclusions.
  2. [Eq. (4) / Figure 2(A) / Discussion] Eq. (4) and Figure 2(A): FedAvg weights updates by sample size, so Lifelines (~15× larger training set) dominates the aggregate. The Rotterdam trajectory peaks near iteration 12 (~0.740) then declines slightly by iteration 20, which the Discussion attributes to growing Lifelines influence. Reporting only the final (iteration-20) model as the federated result, without a pre-specified stopping rule, quality-weighted aggregation, or an ablation that equalizes or down-weights the noisier node, leaves open whether the reported Rotterdam gain is stable transfer of signal or a transient/size-weighting artifact. At minimum, report best-iteration and last-iteration results with a formal comparison, and discuss (or ablate) alternatives to pure nk/n weighting given the known label-quality asymmetry (Section 2.2.2).
  3. [Section 2.2.2 / Discussion] Section 2.2.2 and Discussion: The weakest modeling assumption is that Lifelines self-reported events with midpoint-interval times remain informative under size-weighted FedAvg for a model evaluated on Rotterdam’s precise clinical outcomes, despite large differences in age, event rate (2.6% vs 13.2%), and outcome definition. The paper acknowledges this but does not quantify sensitivity (e.g., event-time noise simulation on Lifelines, or federation with event-type subsets). A sensitivity analysis or explicit bound on how much label noise can be tolerated would make the transfer claim more credible.
minor comments (6)
  1. [Abstract / Table 2] Abstract and Table 2: State explicitly that CIs overlap and that no formal test of the difference was performed (or add the test), so readers are not left to infer statistical significance from non-overlapping language that is not used but easily misread.
  2. [Section 2.3.1] Section 2.3.1: The architecture [12, 16, 16, 1], dropout 0.20, lr 1e-4, batch 1024, 100 local epochs, and 20 rounds are free hyperparameters with no tuning description. Briefly note whether they were fixed a priori from the prior MyDigiTwin work or selected on validation data.
  3. [Section 2.3.3] Section 2.3.3: Median imputation is done independently per site; a short note on missingness rates by cohort would help assess whether imputation differences could affect federation.
  4. [Discussion] Discussion: Competing risks (non-CVD death treated as censoring) is correctly flagged as a limitation for absolute risk in older Rotterdam participants; consider stating that discrimination (C-statistic) is the sole reported metric and is less directly affected than calibration/absolute risk.
  5. [References] References / Katzman et al.: The DeepSurv citation line appears garbled in the reference list (“Ded2008generalepsurv”); please correct the bibliographic entry.
  6. [Figure 2] Figure 2: Add error bars or shaded bands for the 10-run variability at each iteration so the trajectory can be judged against run-to-run noise.

Circularity Check

0 steps flagged

No circularity: federated vs local C-statistic comparison is an external empirical evaluation, not a result forced by definition or self-citation.

full rationale

The paper's central claim is that DeepSurv trained with FedAvg across Lifelines and the Rotterdam Study yields higher held-out C-statistics than the same architecture trained locally (Table 2; Results). The C-statistic is a standard discrimination metric computed on independent local test splits (80/10/10 stratified resampling, 10 runs, corrected resampled t-test CIs); it is not defined by the partial log-likelihood objective (Eq. 3) or by the FedAvg weighted average (Eq. 4). Local baselines use only each node's own data before any aggregation, so the reported gains are not tautological. Self-citations to MyDigiTwin / Cadavid et al. (2025) and the Vantage6 platform describe infrastructure, FHIR harmonization, and a prior pseudo-split feasibility study; they do not supply a uniqueness theorem, ansatz, or fitted parameter that forces the multi-cohort performance numbers. Outcome definitions, predictor sets, and cohort characteristics are stated independently of the claimed improvement. Overlapping CIs and sample-size weighting are statistical/causal concerns, not circular reductions of prediction to input. Score 0 is therefore appropriate.

Axiom & Free-Parameter Ledger

4 free parameters · 5 axioms · 0 invented entities

The claim rests on standard survival and FL machinery plus domain choices about predictors, outcomes, imputation, and aggregation. Free parameters are the usual neural-net and training hyperparameters chosen by the authors; axioms are Cox partial likelihood, FedAvg weighting, and the assumption that self-report labels remain useful under federation. No new physical entities are invented.

free parameters (4)
  • DeepSurv architecture [12, 16, 16, 1]
    Hidden width and depth chosen by authors; not derived from data theory.
  • dropout rate 0.20
    Regularization hyperparameter set by hand.
  • Adam learning rate 1e-4, batch size 1024, 100 local epochs, 20 FedAvg rounds
    Training schedule fixed without reported hyperparameter search; affects final C-statistics.
  • median imputation of missing predictors (per site)
    Simple imputation choice that can shift risk scores.
axioms (5)
  • domain assumption Cox proportional hazards partial log-likelihood is a valid training objective for DeepSurv on these cohorts
    Section 2.3.1 adopts the standard DeepSurv loss; PH and non-competing-risk assumptions are not validated.
  • domain assumption FedAvg sample-size weighting is an appropriate aggregator under large n and label-quality imbalance
    Equation (4); Discussion notes Lifelines dominates later rounds and Rotterdam performance slightly declines after peak.
  • domain assumption Lifelines self-reported CVD with midpoint event times are sufficiently accurate to transfer useful signal
    Section 2.2.2 outcome definition; primary claim depends on this transfer to Rotterdam.
  • domain assumption Horizontal FL with FHIR-harmonized 12 predictors yields comparable feature spaces across sites
    Sections 2.2.1–2.2.3; residual semantic mismatch would undermine federation.
  • standard math Standard neural-net optimization and C-statistic evaluation machinery
    Adam, ReLU, batch norm, concordance index, corrected resampled t-test.

pith-pipeline@v1.1.0-grok45 · 16226 in / 2969 out tokens · 25641 ms · 2026-07-10T04:36:41.348245+00:00 · methodology

0 comments
read the original abstract

Cardiovascular disease risk prediction models often rely on data from a single institution or centrally pooled datasets. Extending these models across institutions could be limited by privacy regulations and constraints on sharing patient-level data. Federated learning enables collaborative model development without transferring sensitive patient data, but its application in healthcare remains challenging because datasets often differ in size, population characteristics, and outcome definitions. In this study, we present a federated deep learning approach for privacy-preserving cardiovascular disease risk prediction that integrates two population-based cohorts with different characteristics: Lifelines, including 148,230 participants meeting the study inclusion criteria with self-reported outcomes, and the Rotterdam Study, including a smaller cohort of 10,155 participants with digitally linked clinical outcomes. Model performance was primarily evaluated on the Rotterdam Study because of its complete follow-up. Deep survival models trained using federated learning achieved higher predictive performance than models trained locally without federation. For the Rotterdam Study, the C-statistic increased from 0.728 (95% CI: 0.717-0.739) to 0.739 (95% CI: 0.728-0.749). For Lifelines, the C-statistic increased from 0.783 (95% CI: 0.775-0.791) to 0.787 (95% CI: 0.780-0.792). These findings suggest that federated deep learning across heterogeneous cohorts can improve cardiovascular disease risk prediction while preserving the privacy of individual-level patient data.

Figures

Figures reproduced from arXiv: 2607.08595 by Daniel Bos, Djura Smits, Esther E. Bron, Hyunho Mo, Maarten J.G. Leening, Mahlet A. Birhanu, Pim van der Harst.

Figure 1
Figure 1. Figure 1: Overview of the federated learning setup using the Vantage6 infrastructure. A dedicated [PITH_FULL_IMAGE:figures/full_fig_p015_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Mean C-statistic across 10 independent runs at each FedAvg update iteration for the [PITH_FULL_IMAGE:figures/full_fig_p015_2.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

204 extracted references · 204 canonical work pages · 9 internal anchors

  1. [1]

    Computers in biology and medicine , volume=

    MyDigiTwin: A privacy-preserving framework for personalized cardiovascular risk prediction and scenario exploration , author=. Computers in biology and medicine , volume=. 2025 , publisher=

    work page 2025

  2. [2]

    2022 , month = sep, langid =

    work page 2022

  3. [3]

    Government of the Netherlands , year =

    What Is a Personal Healthcare Environment? , author =. Government of the Netherlands , year =

    work page

  4. [4]

    Cardiovascular disease risk communication and prevention: a meta-analysis , volume =

    Bakhit, Mina and Fien, Samantha and Abukmail, Eman and Jones, Mark and Clark, Justin and Scott, Anna Mae and Glasziou, Paul and Cardona, Magnolia , urldate =. Cardiovascular disease risk communication and prevention: a meta-analysis , volume =. doi:10.1093/eurheartj/ehae002 , shorttitle =

    work page doi:10.1093/eurheartj/ehae002

  5. [5]

    Hart- en vaatziekten , url =

    work page

  6. [6]

    Studies in Health Technology and Informatics , volume=

    Leveraging FHIR in Federated Learning Environments: A Data Harmonization Framework for Cohort Studies , author=. Studies in Health Technology and Informatics , volume=. 2024 , publisher=

    work page 2024

  7. [7]

    BMC medical research methodology , volume=

    A review of the reporting and handling of missing data in cohort studies with repeated assessment of exposure measures , author=. BMC medical research methodology , volume=. 2012 , publisher=

    work page 2012

  8. [8]

    Zanen, Wouter , urldate =

    work page

  9. [9]

    European Health Data Space , url =

    European Commission , urldate =. European Health Data Space , url =

    work page

  10. [10]

    Logical observation identifier names and codes (

    Forrey,. Logical observation identifier names and codes (. doi:10.1093/clinchem/42.1.81 , pages =

    work page doi:10.1093/clinchem/42.1.81

  11. [11]

    5-step briefing , url =

    work page

  12. [12]

    Communication from the Commission to the European Parliament and the Council: A European Health Data Space: harnessing the power of health data for people, patients and innovation , author =

    work page

  13. [13]

    NPJ digital medicine , volume=

    The health digital twin to tackle cardiovascular disease—a review of an emerging interdisciplinary field , author=. NPJ digital medicine , volume=. 2022 , publisher=

    work page 2022

  14. [14]

    Annual Review of Pharmacology and Toxicology , volume=

    Health digital twins in life science and health care innovation , author=. Annual Review of Pharmacology and Toxicology , volume=. 2024 , publisher=

    work page 2024

  15. [15]

    Chinese Medical Journal , volume=

    Opportunities and challenges of digital twin technology in healthcare , author=. Chinese Medical Journal , volume=. 2023 , publisher=

    work page 2023

  16. [16]

    Challenges and opportunities for digital twins in precision medicine: a complex systems perspective

    Challenges and opportunities for digital twins in precision medicine: a complex systems perspective , author=. arXiv preprint arXiv:2405.09649 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  17. [17]

    Journal of the American Heart Association , volume=

    Building Digital Twins for Cardiovascular Health: From Principles to Clinical Impact , author=. Journal of the American Heart Association , volume=

    work page

  18. [18]

    Transdisciplinary perspectives on complex systems: New findings and approaches , pages=

    Digital twin: Mitigating unpredictable, undesirable emergent behavior in complex systems , author=. Transdisciplinary perspectives on complex systems: New findings and approaches , pages=. 2017 , publisher=

    work page 2017

  19. [19]

    Data Intelligence , volume=

    Distributed analytics on sensitive medical data: the personal health train , author=. Data Intelligence , volume=. 2020 , publisher=

    work page 2020

  20. [20]

    AMIA annual symposium proceedings , volume=

    VANTAGE6: an open source priVAcy preserviNg federaTed leArninG infrastructurE for Secure Insight eXchange , author=. AMIA annual symposium proceedings , volume=. 2020 , organization=

    work page 2020

  21. [21]

    Federated Learning , pages=

    Horizontal federated learning , author=. Federated Learning , pages=. 2020 , publisher=

    work page 2020

  22. [22]

    European journal of public health , volume=

    Benefits and challenges of Big Data in healthcare: an overview of the European initiatives , author=. European journal of public health , volume=. 2019 , publisher=

    work page 2019

  23. [23]

    BMC medical informatics and decision making , volume=

    Learning patient-level prediction models across multiple healthcare databases: evaluation of ensembles for increasing model transportability , author=. BMC medical informatics and decision making , volume=. 2022 , publisher=

    work page 2022

  24. [25]

    Circulation , author =

    Heart. Circulation , author =. 2019 , note =. doi:10.1161/CIR.0000000000000659 , number =

    work page doi:10.1161/cir.0000000000000659 2019

  25. [26]

    professional.heart.org , month = sep, year =

    work page

  26. [27]

    Diabetologia , author =

    Cardiovascular risk prediction in type 2 diabetes: a comparison of 22 risk scores in primary care settings , volume =. Diabetologia , author =. 2022 , pmid =. doi:10.1007/s00125-021-05640-y , abstract =

    work page doi:10.1007/s00125-021-05640-y 2022

  27. [28]

    New England Journal of Medicine , volume=

    Estimating glomerular filtration rate from serum creatinine and cystatin C , author=. New England Journal of Medicine , volume=. 2012 , publisher=

    work page 2012

  28. [29]

    2022 , publisher=

    Test driven development: By example , author=. 2022 , publisher=

    work page 2022

  29. [30]

    Foundations and trends

    Advances and open problems in federated learning , author=. Foundations and trends. 2021 , publisher=

    work page 2021

  30. [31]

    ACM Transactions on Intelligent Systems and Technology (TIST) , volume=

    Federated machine learning: Concept and applications , author=. ACM Transactions on Intelligent Systems and Technology (TIST) , volume=. 2019 , publisher=

    work page 2019

  31. [32]

    Artificial intelligence and statistics , pages=

    Communication-efficient learning of deep networks from decentralized data , author=. Artificial intelligence and statistics , pages=. 2017 , organization=

    work page 2017

  32. [33]

    2008 , publisher=

    Applied survival analysis: regression modeling of time-to-event data , author=. 2008 , publisher=

    work page 2008

  33. [34]

    Journal of the Royal Statistical Society: Series B (Methodological) , volume=

    Regression models and life-tables , author=. Journal of the Royal Statistical Society: Series B (Methodological) , volume=. 1972 , publisher=

    work page 1972

  34. [35]

    Jama , volume=

    Evaluating the yield of medical tests , author=. Jama , volume=. 1982 , publisher=

    work page 1982

  35. [36]

    Archives of internal medicine , volume=

    Decade-long trends and factors associated with time to hospital presentation in patients with acute myocardial infarction: the Worcester Heart Attack study , author=. Archives of internal medicine , volume=. 2000 , publisher=

    work page 2000

  36. [37]

    ACM Computing Surveys , volume=

    Heterogeneous federated learning: State-of-the-art and research challenges , author=. ACM Computing Surveys , volume=. 2023 , publisher=

    work page 2023

  37. [38]

    Proceedings of the 3rd International Workshop on Distributed Machine Learning , pages=

    Rethinking normalization methods in federated learning , author=. Proceedings of the 3rd International Workshop on Distributed Machine Learning , pages=

    work page

  38. [39]

    2023 IEEE International Conference on Internet of Things and Intelligence Systems (IoTaIS) , pages=

    Federated Learning Algorithm Handling Missing Attributes , author=. 2023 IEEE International Conference on Internet of Things and Intelligence Systems (IoTaIS) , pages=. 2023 , organization=

    work page 2023

  39. [40]

    Advances in neural information processing systems , volume=

    Inference for the generalization error , author=. Advances in neural information processing systems , volume=

    work page

  40. [41]

    International journal of cardiology , volume=

    The LifeLines Cohort Study: prevalence and treatment of cardiovascular disease and risk factors , author=. International journal of cardiology , volume=. 2017 , publisher=

    work page 2017

  41. [42]

    International journal of epidemiology , volume=

    Cohort Profile: LifeLines, a three-generation cohort study and biobank , author=. International journal of epidemiology , volume=. 2015 , publisher=

    work page 2015

  42. [43]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Local learning matters: Rethinking data heterogeneity in federated learning , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    work page

  43. [44]

    Fast Healthcare Interoperability Resources (FHIR) Specification , year =

    work page

  44. [45]

    Nature Medicine , volume=

    Federated learning in healthcare: facilitating multi-institutional collaborations without sharing patient data , author=. Nature Medicine , volume=. 2021 , doi=

    work page 2021

  45. [46]

    Mandemakers and Harold Snieder and Ronald P

    Bart Klijs and Salome Scholtens and Jornt J. Mandemakers and Harold Snieder and Ronald P. Stolk and Nynke Smidt , title =. PLOS ONE , volume =. 2015 , doi =

    work page 2015

  46. [47]

    Procedia Computer Science , volume=

    Homomorphic encryption , author=. Procedia Computer Science , volume=. 2013 , publisher=

    work page 2013

  47. [48]

    Electronics , volume=

    Reviewing federated learning aggregation algorithms; strategies, contributions, limitations and future perspectives , author=. Electronics , volume=. 2023 , publisher=

    work page 2023

  48. [49]

    International colloquium on automata, languages, and programming , pages=

    Differential privacy , author=. International colloquium on automata, languages, and programming , pages=. 2006 , organization=

    work page 2006

  49. [50]

    Biometrics , volume=

    Time-dependent ROC curves for censored survival data and a diagnostic marker , author=. Biometrics , volume=. 2000 , publisher=

    work page 2000

  50. [51]

    Proceedings of Machine learning and systems , volume=

    Federated optimization in heterogeneous networks , author=. Proceedings of Machine learning and systems , volume=

    work page

  51. [52]

    FedBN: Federated Learning on Non-IID Features via Local Batch Normalization

    Fedbn: Federated learning on non-iid features via local batch normalization , author=. arXiv preprint arXiv:2102.07623 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  52. [53]

    Advances in neural information processing systems , volume=

    An efficient framework for clustered federated learning , author=. Advances in neural information processing systems , volume=

    work page

  53. [54]

    IEEE transactions on neural networks and learning systems , volume=

    Clustered federated learning: Model-agnostic distributed multitask optimization under privacy constraints , author=. IEEE transactions on neural networks and learning systems , volume=. 2020 , publisher=

    work page 2020

  54. [55]

    Statistics in medicine , volume=

    A neural network model for survival data , author=. Statistics in medicine , volume=. 1995 , publisher=

    work page 1995

  55. [56]

    Adam: A Method for Stochastic Optimization

    Adam: A method for stochastic optimization , author=. arXiv preprint arXiv:1412.6980 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  56. [57]

    1997 , publisher=

    Analysis of incomplete multivariate data , author=. 1997 , publisher=

    work page 1997

  57. [58]

    Article Title

    Author1 LastName1 and Author2 LastName2 and Author3 LastName3. Article Title. Frontiers in Neuroscience. 2013. doi:10.3389/fnins.2013.12345

    work page doi:10.3389/fnins.2013.12345 2013

  58. [59]

    Author Name , title =

    work page

  59. [60]

    The title of the book , publisher =

    Bauthor Surname , title =. The title of the book , publisher =

    work page

  60. [61]

    The title of the conference proceedings , year = 1996, publisher =

    Cauthor Name and Dauthor Surname and Fauthor LastName , title =. The title of the conference proceedings , year = 1996, publisher =

    work page 1996

  61. [62]

    Gauthor Name1 , title =

    work page

  62. [63]

    Hauthor Surname1 , title =

    work page

  63. [64]

    Data Title

    Author1 LastName1 and Author2 LastName2 and Author3 LastName3. Data Title. 2011. doi:10.000/55555

    work page 2011

  64. [65]

    The Lancet , volume =

    Kathy Leung and Joseph T Wu and Di Liu and Gabriel M Leung , title =. The Lancet , volume =

    work page

  65. [66]

    The Lancet Public Health , volume =

    Benjamin J Cowling and Sheikh Taslim Ali and Tiffany W Y Ng and Tim K Tsang and Julian C M Li and Min Whui Fong and Qiuyan Liao and Mike YW Kwan and So Lun Lee and Susan S Chiu and Joseph T Wu and Peng Wu and Gabriel M Leung , title =. The Lancet Public Health , volume =. doi:10.1016/s2468-2667(20)30090-6 , url =

    work page doi:10.1016/s2468-2667(20)30090-6

  66. [67]

    Cardiovascular Imaging , volume=

    When does a calcium score equate to secondary prevention? Insights from the multinational CONFIRM registry , author=. Cardiovascular Imaging , volume=. 2023 , publisher=

    work page 2023

  67. [68]

    Heart, Lung and Circulation , volume=

    Cardiac Society of Australia and New Zealand position statement: coronary artery calcium scoring , author=. Heart, Lung and Circulation , volume=. 2017 , publisher=

    work page 2017

  68. [69]

    Canadian journal of cardiology , volume=

    2021 Canadian Cardiovascular Society guidelines for the management of dyslipidemia for the prevention of cardiovascular disease in adults , author=. Canadian journal of cardiology , volume=. 2021 , publisher=

    work page 2021

  69. [70]

    Greenwade

    George D. Greenwade. The C omprehensive T ex A rchive N etwork ( CTAN ). TUGBoat. 1993

    work page 1993

  70. [71]

    Circulation , volume=

    2013 AHA/ACC guideline on lifestyle management to reduce cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines , author=. Circulation , volume=. 2014 , publisher=

    work page 2013

  71. [72]

    PloS one , volume=

    Cardiovascular disease risk prediction using automated machine learning: A prospective study of 423,604 UK Biobank participants , author=. PloS one , volume=. 2019 , publisher=

    work page 2019

  72. [73]

    International Journal of Epidemiology , volume=

    Predicting cardiovascular risk from national administrative databases using a combined survival analysis and deep learning approach , author=. International Journal of Epidemiology , volume=. 2022 , publisher=

    work page 2022

  73. [74]

    Circulation , volume=

    General cardiovascular risk profile for use in primary care: the Framingham Heart Study , author=. Circulation , volume=. 2008 , publisher=

    work page 2008

  74. [75]

    Advances in neural information processing systems , volume=

    Attention is all you need , author=. Advances in neural information processing systems , volume=

    work page

  75. [76]

    MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications

    Mobilenets: Efficient convolutional neural networks for mobile vision applications , author=. arXiv preprint arXiv:1704.04861 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  76. [77]

    European heart journal , volume=

    Estimation of ten-year risk of fatal cardiovascular disease in Europe: the SCORE project , author=. European heart journal , volume=. 2003 , publisher=

    work page 2003

  77. [78]

    European journal of preventive cardiology , volume=

    2021 ESC Guidelines on cardiovascular disease prevention in clinical practice: Developed by the Task Force for cardiovascular disease prevention in clinical practice with representatives of the European Society of Cardiology and 12 medical societies With the special contribution of the European Association of Preventive Cardiology (EAPC) , author=. Europe...

    work page 2021

  78. [79]

    Journal of the American College of Cardiology , volume=

    2013 ACC/AHA guideline on the assessment of cardiovascular risk , author=. Journal of the American College of Cardiology , volume=. 2014 , publisher=

    work page 2013

  79. [80]

    bmj , volume=

    Prediction models for cardiovascular disease risk in the general population: systematic review , author=. bmj , volume=. 2016 , publisher=

    work page 2016

  80. [81]

    Journal of Healthcare Informatics Research , volume=

    Multi-cohort federated learning shows synergy in mortality prediction for MRI-based and metabolomics-based age scores , author=. Journal of Healthcare Informatics Research , volume=. 2025 , publisher=

    work page 2025

Showing first 80 references.