CoChat
Sign Up
PreprintCited by 40Multiple Sclerosis Research StudiesMachine Learning in HealthcareStatistical Methods in Clinical Trials

Generating Digital Twins with Multiple Sclerosis Using Probabilistic Neural Networks

Open access
Walsh, Jonathan R., Smith, Aaron M., Pouliot, Yannick, Li-Bland, David, Loukianov, Anton, Fisher, Charles K., for the Multiple Sclerosis Outcome Assessments Consortium·2020·bioRxiv (Cold Spring Harbor Laboratory)
DOI 10.1101/2020.02.04.934679
Open this paper in CoChat — free accountOpen PDF

Abstract

Multiple Sclerosis (MS) is a neurodegenerative disorder characterized by a complex set of clinical assessments. We use an unsupervised machine learning model called a Conditional Restricted Boltzmann Machine (CRBM) to learn the relationships between covariates commonly used to characterize subjects and their disease progression in MS clinical trials. A CRBM is capable of generating digital twins, which are simulated subjects having the same baseline data as actual subjects. Digital twins allow for subject-level statistical analyses of disease progression. The CRBM is trained using data from 2395 subjects enrolled in the placebo arms of clinical trials across the three primary subtypes of MS. We discuss how CRBMs are trained and show that digital twins generated by the model are statistically indistinguishable from their actual subject counterparts along a number of measures.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1101/2020.02.04.934679

Cites

34 of 34 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Unlearn.AI has developed and tested digital twin models that leverage diverse sets of clinical data to enhance clinical trials for Alzheimer's disease and multiple sclerosis.

    “From a commercial point of view, Unlearn.AI has developed and tested digital twin models that leverage diverse sets of clinical data to enhance clinical trials for Alzheimer's disease and multiple sclerosis 70,71 .”

    From Multimodal biomedical AI (2022) · cites this paper as [71]

    The Abstract and Discussion report that the authors developed and statistically tested digital twins using varied clinical data from multiple-sclerosis trials, and describe potential uses in trial analysis. The Discussion mentions earlier work by the authors on Alzheimer’s disease, but this paper does not itself describe or test that model. It also presents trial enhancement as a potential application, not a demonstrated outcome. (LLM, conf=0.84)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 24

  1. N. Holford, S. Ma, and B. Ploeger, Clinical Pharmacology & Therapeutics 88, 166 (2010).
  2. C. K. Fisher, A. M. Smith, and J. R. Walsh, Scientific Reports 9, 1 (2019). · DOI · Paper page
  3. G. Carrigan, S. Whipple, W. B. Capra, M. D. Taylor, J. S. Brown, M. Lu, B. Arnieri, R. Cop- ping, and K. J. Rothman, Clinical Pharmacology & Therapeutics 107, 369 (2020).
  4. C. H. Polman, S. Reingold, F. Barkhof, P. Calabresi, M. Clanet, J. Cohen, G. Cutter, M. Freed- man, L. Kappos, F. Lublin, et al., Neurology 70, 1134 (2008).
  5. B. M. Uitdehaag, F. Barkhof, P. K. Coyle, J. D. Gardner, D. R. Jeffery, and D. D. Mikol, Current medical research and opinion 27, 1529 (2011).
  6. M. Grieves and J. Vickers, in Transdisciplinary Perspectives on Complex Systems: New Find- ings and Approaches, edited by F.-J. Kahlen, S. Flumerfelt, and A. Alves (Springer, 2016).
  7. G. W. Taylor and G. E. Hinton, in Proceedings of the 26th Annual International Conference on Machine Learning, ICML '09 (ACM, New York, NY, USA, 2009) pp. 1025-1032, event-place: Montreal, Quebec, Canada. · DOI
  8. G. W. Taylor, G. E. Hinton, and S. T. Roweis, in Advances in Neural Information Processing Systems 19 , edited by B. Schölkopf, J. C. Platt, and T. Hoffman (MIT Press, 2007) pp. 1345-1352.
  9. M. M. Goldenberg, Pharmacy and Therapeutics 37, 175 (2012).
  10. D. H. Ackley, G. E. Hinton, and T. J. Sejnowski, Cognitive science 9, 147 (1985).
  11. G. E. Hinton and R. R. Salakhutdinov, Science 313, 504 (2006).
  12. R. Salakhutdinov and G. Hinton, in Artificial Intelligence and Statistics (2009) pp. 448-455.
  13. G. Hinton, Momentum 9, 926 (2010).
  14. I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, and Y. Bengio, in Advances in neural information processing systems (2014) pp. 2672-2680.
  15. W. R. Kubick, S. Ruberg, and E. Helton, Drug Information Journal (2016). · DOI
  16. H. Wickham et al., Journal of Statistical Software 59, 1 (2014).
  17. H. Theil, Nederl. Akad. Wetensch., Proc. 53, 386 (1950).
  18. P. K. Sen, Journal of the American Statistical Association 63, 1379 (1968). · DOI
  19. D. B. Rubin, Scandinavian Journal of Statistics 31, 161 (2004).
  20. G. W. Imbens and D. B. Rubin, in Microeconometrics (Springer, 2010) pp. 229-241.
  21. S. R. Künzel, J. S. Sekhon, P. J. Bickel, and B. Yu, Proceedings of the national academy of sciences 116, 4156 (2019).
  22. C. Gouriéroux and J. Jasiak, Journal of Forecasting 25, 129.
  23. G. Desjardins, A. Courville, Y. Bengio, P. Vincent, and O. Delalleau, in Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics (MIT Press Cam- bridge, MA, 2010) pp. 145-152.
  24. G. Desjardins, A. Courville, Y. Bengio, P. Vincent, and O. Delalleau, in Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics (2010) pp. 145- 152.

Cited by

Sorted by citations

Related

Authors
Walsh, J. R., Smith, A. M., Pouliot, Y., Li-Bland, D., Loukianov, A., Fisher, C. K., & for the Multiple Sclerosis Outcome Assessments Consortium
Corresponding author
Walsh, Jonathan R.
ORCID
Venue
bioRxiv (Cold Spring Harbor Laboratory)
Publisher
openRxiv
Published
2020-02-05
DOI
10.1101/2020.02.04.934679
OpenAlex
W3004601859
Cited by
40
References
34
Citations by year
2026: 6 · 2025: 4 · 2024: 5 · 2023: 10 · 2022: 5 · 2021: 3 · 2020: 7
Type
posted content
Language
en
Open access
Licence
Locations
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by)
  • arXiv (Cornell University) (submittedVersion)
  • arXiv (Cornell University)
  • TIB Data Manager
Topics
  • Health Sciences › Medicine › Pathology and Forensic Medicine › Multiple Sclerosis Research Studies
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
  • Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Clinical Trials
Keywords
multiple sclerosis, digital twin, conditional restricted Boltzmann machine, probabilistic neural network, disease progression simulation, unsupervised learning, neurodegenerative diseases

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review