7. To Go Further

7.1. Model Implementation

7.1.1. DAG Intuition and Structure for Models

7.2. What kind of scientific question could be answered with leaspy?

Different papers have been published trying to answer different scientific questions using the software.

7.2.1. Used in different context

Different chronic diseases: The model has been used to describe very different chronic diseases as Hungtington [KDBT+22], Alzheimer [MKO+23], Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts Leukoencephalopathy (CADASIL) [KHJ+25], Amyothrophic Lateral Sclerosis [OPTdMD23], Ataxia [MPP+23], Parkinson [CVC+19, PD23]. ALS

Many types of data: Different types of data have been analysed from clinical scores to biomarkers such as clinical scores and brain markers [KBL+21] and events [ODM25]. For longitudinal data progression where used from linear [REF?] and logistic [KHJ+25, OPTdMD23] to ordinal [MPP+23]. The model has been shown quite robust to missing data [CVC+19].

7.2.2. Used for different tasks

Describe the joint progression of multiple outcomes: This package has been extensively used to describe the progression of multiple outcomes [KBL+21, OPTdMD23] up to 14 clinical outcomes have been studied at the same time [KHJ+25].

Describe disease heterogeneity: Post-hoc analysis of the individual variability to describe disease heterogeneity were conducted using a supervised approach for Amyotrophic Lateral Sclerosis [OPTdMD23] and Ataxia [MPP+23] as well as an unsupervised approach for CADASIL [KHJ+25].

Improve clinical trials: The model has been shown useful to select patients for clinical trials in order to increase the sensibility of the trial [MKO+23]. The model’s predictions could also be integrated to clinical trials, through prognostic score’s methods (such as Prognostic Covariate Adjustment or Prediction-Powered inference for Clinical Trials) to increase the statistical power of the trials [PTM+25].

Make predictions: Leaspy outperformed the 56 alternative methods for predicting cognitive decline in the framework of the TADPOLE challenge [MOY+19] and was more generally used for diverse applications [KDBT+22, MKO+23]

7.3. References

[CVC+19] (1,2)

Raphael Couronne, Marie Vidailhet, Jean Christophe Corvol, Stephane Lehericy, and Stanley Durrleman. Learning Disease Progression Models With Longitudinal Data and Missing Values. In 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019), 1033–1037. April 2019. ISSN: 1945-8452. URL: https://ieeexplore.ieee.org/abstract/document/8759198 (visited on 2025-05-07), doi:10.1109/ISBI.2019.8759198.

[KHJ+25] (1,2,3,4)

Sofia Kaisaridi, Dominique Herve, Aude Jabouley, Sonia Reyes, Carla Machado, Stéphanie Guey, Abbas Taleb, Fanny Fernandes, Hugues Chabriat, and Sophie Tezenas Du Montcel. Determining Clinical Disease Progression in Symptomatic Patients With CADASIL. Neurology, 104(1):e210193, January 2025. doi:10.1212/WNL.0000000000210193.

[KBL+21] (1,2)

Igor Koval, Alexandre Bône, Maxime Louis, Thomas Lartigue, Simona Bottani, Arnaud Marcoux, Jorge Samper-González, Ninon Burgos, Benjamin Charlier, Anne Bertrand, Stéphane Epelbaum, Olivier Colliot, Stéphanie Allassonnière, and Stanley Durrleman. AD Course Map charts Alzheimer's disease progression. Scientific Reports, 11(1):8020, April 2021. doi:10.1038/s41598-021-87434-1.

[KDBT+22] (1,2)

Igor Koval, Thomas Dighiero-Brecht, Allan J. Tobin, Sarah J. Tabrizi, Rachael I. Scahill, Sophie Tezenas du Montcel, Stanley Durrleman, and Alexandra Durr. Forecasting individual progression trajectories in Huntington disease enables more powered clinical trials. Scientific Reports, 12(1):18928, November 2022. Publisher: Nature Publishing Group. URL: https://www.nature.com/articles/s41598-022-18848-8 (visited on 2025-05-07), doi:10.1038/s41598-022-18848-8.

[MKO+23] (1,2,3)

Etienne Maheux, Igor Koval, Juliette Ortholand, Colin Birkenbihl, Damiano Archetti, Vincent Bouteloup, Stéphane Epelbaum, Carole Dufouil, Martin Hofmann-Apitius, and Stanley Durrleman. Forecasting individual progression trajectories in Alzheimer’s disease. Nature Communications, 14(1):761, February 2023. Publisher: Nature Publishing Group. URL: https://www.nature.com/articles/s41467-022-35712-5 (visited on 2025-05-07), doi:10.1038/s41467-022-35712-5.

[MOY+19]

Răzvan V. Marinescu, Neil P. Oxtoby, Alexandra L. Young, Esther E. Bron, Arthur W. Toga, Michael W. Weiner, Frederik Barkhof, Nick C. Fox, Polina Golland, Stefan Klein, and Daniel C. Alexander. TADPOLE Challenge: Accurate Alzheimer’s disease prediction through crowdsourced forecasting of future data. PRedictive Intelligence in MEdicine. PRIME (Workshop), 11843:1–10, October 2019. URL: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7315046/ (visited on 2025-05-07), doi:10.1007/978-3-030-32281-6_1.

[MPP+23] (1,2,3)

Paul Moulaire, Pierre Emmanuel Poulet, Emilien Petit, Thomas Klockgether, Alexandra Durr, Tetsuo Ashisawa, Sophie Tezenas du Montcel, and READISCA Consortium. Temporal Dynamics of the Scale for the Assessment and Rating of Ataxia in Spinocerebellar Ataxias. Movement Disorders: Official Journal of the Movement Disorder Society, 38(1):35–44, January 2023. doi:10.1002/mds.29255.

[ODM25]

Juliette Ortholand, Stanley Durrleman, and Sophie Tezenas du Montcel. A joint spatiotemporal model for multiple longitudinal markers and competing events. January 2025. arXiv:2501.08960 [stat]. URL: http://arxiv.org/abs/2501.08960 (visited on 2025-05-07), doi:10.48550/arXiv.2501.08960.

[OPTdMD23] (1,2,3,4)

Juliette Ortholand, Pierre-François Pradat, Sophie Tezenas du Montcel, and Stanley Durrleman. Interaction of sex and onset site on the disease trajectory of amyotrophic lateral sclerosis. Journal of Neurology, 270(12):5903–5912, December 2023. doi:10.1007/s00415-023-11932-7.

[PD23]

Pierre-Emmanuel Poulet and Stanley Durrleman. Multivariate disease progression modeling with longitudinal ordinal data. Statistics in Medicine, 42(18):3164–3183, 2023. _eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1002/sim.9770. URL: https://onlinelibrary.wiley.com/doi/abs/10.1002/sim.9770 (visited on 2025-05-07), doi:10.1002/sim.9770.

[PTM+25]

Pierre-Emmanuel Poulet, Maylis Tran, Sophie Tezenas du Montcel, Bruno Dubois, Stanley Durrleman, and Bruno Jedynak. Prediction-powered Inference for Clinical Trials. January 2025. Pages: 2025.01.15.25320578. URL: https://www.medrxiv.org/content/10.1101/2025.01.15.25320578v1 (visited on 2025-05-14), doi:10.1101/2025.01.15.25320578.