Diego Valderrama
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Diego Valderrama has a total of 32 co-authors. Here are the top co-authors by number of shared publications:
- Holger Fröhlich
Fraunhofer SCAI, Sankt Augustin, Germany
10 shared publications - Sven Stodtmann
Clinical Pharmacology AbbVie Deutschland GmbH & Co. KG Ludwigshafen am Rhein Germany
4 shared publications - Elena Trunz
Institute of Computer Science II, Visual Computing University of Bonn Bonn Germany
3 shared publications - Luca Marie Koltermann
Department of Clinical Pharmacy, Institute of Pharmacy, University of Bonn, Bonn, Germany
3 shared publications - Olga Teplytska
Department of Clinical Pharmacy, Institute of Pharmacy, University of Bonn, Bonn, Germany
3 shared publications - Ulrich Jaehde
Department of Clinical Pharmacy, Institute of Pharmacy, University of Bonn, Bonn, Germany
3 shared publications
Latest Publications
- Open Access
- 20 November 2026
- Published by Elsevier in Journal of Biomedical Informatics
MultiNSDE: treatment-aware neural SDEs for generative modeling of real and counterfactual patient trajectories
- 20 November 2026
- Published by Elsevier in Journal of Biomedical Informatics
Open AccessMultiNSDE: treatment-aware neural SDEs for generative modeling of real and counterfactual patient trajectories
- Open Access
- 29 July 2026
- Published by Elsevier in SSRN Electronic Journal
MultiNSDE: Treatment-Aware Neural SDEs for Generative Modelingof Real and Counterfactual Patient Trajectories
Background: Synthetic patient data may facilitate biomedical research, trial planning, and the construction of synthetic control arms, but there is a lack of methods that enable the realistic generation of mixed-type longitudinal clinical trial data, specifically in the presence of treatment effects. Methods: We developed MultiNSDE, a neural stochastic differential equation framework for modeling high-dimensional static and longitudinal clinical study data, including continuous, categorical, and time-to-event outcomes. The model separates treatment- and placebo-associated dynamics, enabling individual trajectory simulation and counterfactual treatment assignment. Results: MultiNSDE was evaluated on three neurodegenerative disease trial datasets: A4 in Alzheimer’s disease, PRO-ACT in amyotrophic lateral sclerosis, and DATATOP in Parkinson’s disease. The model reproduced static and longitudinal data distributions, extrapolated clinically plausible trajectories, and generated coherent counterfactual outcomes under treatment-to-placebo and placebo-to-treatment shifts. Synthetic data generated by MultiNSDE preserved average treatment effect estimates observed in the real datasets. Conclusions: MultiNSDE enables generation of realistic mixed-type longitudinal clinical trial data while preserving clinically meaningful treatment-effect structure. It may support disease modeling, counterfactual simulation, and synthetic control generation in neurodegenerative disease research.
- 29 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessMultiNSDE: Treatment-Aware Neural SDEs for Generative Modelingof Real and Counterfactual Patient Trajectories
Background: Synthetic patient data may facilitate biomedical research, trial planning, and the construction of synthetic control arms, but there is a lack of methods that enable the realistic generation of mixed-type longitudinal clinical trial data, specifically in the presence of treatment effects. Methods: We developed MultiNSDE, a neural stochastic differential equation framework for modeling high-dimensional static and longitudinal clinical study data, including continuous, categorical, and time-to-event outcomes. The model separates treatment- and placebo-associated dynamics, enabling individual trajectory simulation and counterfactual treatment assignment. Results: MultiNSDE was evaluated on three neurodegenerative disease trial datasets: A4 in Alzheimer’s disease, PRO-ACT in amyotrophic lateral sclerosis, and DATATOP in Parkinson’s disease. The model reproduced static and longitudinal data distributions, extrapolated clinically plausible trajectories, and generated coherent counterfactual outcomes under treatment-to-placebo and placebo-to-treatment shifts. Synthetic data generated by MultiNSDE preserved average treatment effect estimates observed in the real datasets. Conclusions: MultiNSDE enables generation of realistic mixed-type longitudinal clinical trial data while preserving clinically meaningful treatment-effect structure. It may support disease modeling, counterfactual simulation, and synthetic control generation in neurodegenerative disease research.
- Open Access
- 1 December 2025
- Published by Elsevier in The Journal Of Prevention of Alzheimer's Disease
- Vol. 13 (1) , 100420
A benchmark of text embedding models for semantic harmonization of Alzheimer's disease cohorts
- 1 December 2025
- Published by Elsevier in The Journal Of Prevention of Alzheimer's Disease
- Vol. 13 (1) , 100420
Open AccessA benchmark of text embedding models for semantic harmonization of Alzheimer's disease cohorts
- Open Access
- 13 November 2025
- Published by Wiley in CPT: Pharmacometrics & Systems Pharmacology
- Vol. 15 (1)
Predictive AI in Clinical Pharmacology: A Call to Action to Develop Robust Benchmarking Practices
- 13 November 2025
- Published by Wiley in CPT: Pharmacometrics & Systems Pharmacology
- Vol. 15 (1)
Open AccessPredictive AI in Clinical Pharmacology: A Call to Action to Develop Robust Benchmarking Practices
- Open Access
- 7 February 2025
- Published by Wiley in CPT: Pharmacometrics & Systems Pharmacology
- Vol. 14 (4) , 759-769
Comparing Scientific Machine Learning With Population Pharmacokinetic and Classical Machine Learning Approaches for Prediction of Drug Concentrations
A variety of classical machine learning (ML) approaches has been developed over the past decade aiming to individualize drug dosages based on measured plasma concentrations. However, the interpretability of these models is challenging as they do not incorporate information on pharmacokinetic (PK) drug disposition. In this work we compare drug plasma concentraton predictions of well‐known population PK (PopPK) modeling with classical machine learning models and a newly proposed scientific machine learning (MMPK‐SciML) framework. MMPK‐SciML allows to estimate PopPK parameters and their inter‐individual variability (IIV) using multimodal covariate data of each patient and does not require assumptions about the underlying covariate relationships. A dataset of 541 fluorouracil (5FU) plasma concentrations as example for an intravenously administered drug and a dataset of 302 sunitinib and its active metabolite concentrations each as example for an orally administered drug were used for analysis. Whereas classical ML models were not able to describe the data sufficiently, MMPK‐SciML allowed us to obtain accurate drug plasma concentration predictions for test patients. In case of 5FU, goodness‐of‐fit shows that the MMPK‐SciML approach predicts drug plasma concentrations more accurately than PopPK models. For sunitinib, we observed slightly less accurate drug concentration predictions compared to PopPK. Overall, MMPK‐SciML has shown promising results and should therefore be further investigated as a valuable alternative to classical PopPK modeling, provided there is sufficient training data.
- 7 February 2025
- Published by Wiley in CPT: Pharmacometrics & Systems Pharmacology
- Vol. 14 (4) , 759-769
Open AccessComparing Scientific Machine Learning With Population Pharmacokinetic and Classical Machine Learning Approaches for Prediction of Drug Concentrations
A variety of classical machine learning (ML) approaches has been developed over the past decade aiming to individualize drug dosages based on measured plasma concentrations. However, the interpretability of these models is challenging as they do not incorporate information on pharmacokinetic (PK) drug disposition. In this work we compare drug plasma concentraton predictions of well‐known population PK (PopPK) modeling with classical machine learning models and a newly proposed scientific machine learning (MMPK‐SciML) framework. MMPK‐SciML allows to estimate PopPK parameters and their inter‐individual variability (IIV) using multimodal covariate data of each patient and does not require assumptions about the underlying covariate relationships. A dataset of 541 fluorouracil (5FU) plasma concentrations as example for an intravenously administered drug and a dataset of 302 sunitinib and its active metabolite concentrations each as example for an orally administered drug were used for analysis. Whereas classical ML models were not able to describe the data sufficiently, MMPK‐SciML allowed us to obtain accurate drug plasma concentration predictions for test patients. In case of 5FU, goodness‐of‐fit shows that the MMPK‐SciML approach predicts drug plasma concentrations more accurately than PopPK models. For sunitinib, we observed slightly less accurate drug concentration predictions compared to PopPK. Overall, MMPK‐SciML has shown promising results and should therefore be further investigated as a valuable alternative to classical PopPK modeling, provided there is sufficient training data.
- Open Access
- 22 August 2024
- Published by Public Library of Science (PLoS) in PLOS Global Public Health
- Vol. 4 (8) , e0003058
A dynamic ensemble model for short-term forecasting in pandemic situations
During the COVID-19 pandemic, many hospitals reached their capacity limits and could no longer guarantee treatment of all patients. At the same time, governments endeavored to take sensible measures to stop the spread of the virus while at the same time trying to keep the economy afloat. Many models extrapolating confirmed cases and hospitalization rate over short periods of time have been proposed, including several ones coming from the field of machine learning. However, the highly dynamic nature of the pandemic with rapidly introduced interventions and new circulating variants imposed non-trivial challenges for the generalizability of such models. In the context of this paper, we propose the use of ensemble models, which are allowed to change in their composition or weighting of base models over time and could thus better adapt to highly dynamic pandemic or epidemic situations. In that regard, we also explored the use of secondary metadata—Google searches—to inform the ensemble model. We tested our approach using surveillance data from COVID-19, Influenza, and hospital syndromic surveillance of severe acute respiratory infections (SARI). In general, we found ensembles to be more robust than the individual models. Altogether we see our work as a contribution to enhance the preparedness for future pandemic situations.
- 22 August 2024
- Published by Public Library of Science (PLoS) in PLOS Global Public Health
- Vol. 4 (8) , e0003058
Open AccessA dynamic ensemble model for short-term forecasting in pandemic situations
During the COVID-19 pandemic, many hospitals reached their capacity limits and could no longer guarantee treatment of all patients. At the same time, governments endeavored to take sensible measures to stop the spread of the virus while at the same time trying to keep the economy afloat. Many models extrapolating confirmed cases and hospitalization rate over short periods of time have been proposed, including several ones coming from the field of machine learning. However, the highly dynamic nature of the pandemic with rapidly introduced interventions and new circulating variants imposed non-trivial challenges for the generalizability of such models. In the context of this paper, we propose the use of ensemble models, which are allowed to change in their composition or weighting of base models over time and could thus better adapt to highly dynamic pandemic or epidemic situations. In that regard, we also explored the use of secondary metadata—Google searches—to inform the ensemble model. We tested our approach using surveillance data from COVID-19, Influenza, and hospital syndromic surveillance of severe acute respiratory infections (SARI). In general, we found ensembles to be more robust than the individual models. Altogether we see our work as a contribution to enhance the preparedness for future pandemic situations.
- Open Access
- 17 August 2024
- Published by Springer Nature in Clinical Pharmacokinetics
- Vol. 63 (9) , 1221-1237
Machine Learning Methods for Precision Dosing in Anticancer Drug Therapy: A Scoping Review
Introduction In the last decade, various Machine Learning techniques have been proposed aiming to individualise the dose of anticancer drugs mostly based on a presumed drug effect or measured effect biomarkers. The aim of this scoping review was to comprehensively summarise the research status on the use of Machine Learning for precision dosing in anticancer drug therapy. Methods This scoping review was conducted in accordance with the interim guidance by Cochrane and the Joanna Briggs Institute. We systematically searched the databases Medline (via PubMed), Embase and the Cochrane Library for research articles and reviews including results published after 2016. Results were reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) checklist. Results A total of 17 relevant studies was identified. In 12 of the included studies, Reinforcement Learning methods were used, including Classical, Deep, Double Deep and Conservative Q-Learning and Fuzzy Reinforcement Learning. Furthermore, classical Machine Learning methods were compared in terms of their performance and an artificial intelligence platform based on parabolic equations was used to guide dosing prospectively and retrospectively, albeit only in a limited number of patients. Due to the significantly different algorithm structures, a meaningful comparison between the various Machine Learning approaches was not possible. Conclusion Overall, this review emphasises the clinical relevance of Machine Learning methods for anticancer drug dose optimisation, as many algorithms have shown promising results enabling model-free predictions with the potential to maximise efficacy and minimise toxicity when compared to standard protocols.
- 17 August 2024
- Published by Springer Nature in Clinical Pharmacokinetics
- Vol. 63 (9) , 1221-1237
Open AccessMachine Learning Methods for Precision Dosing in Anticancer Drug Therapy: A Scoping Review
Introduction In the last decade, various Machine Learning techniques have been proposed aiming to individualise the dose of anticancer drugs mostly based on a presumed drug effect or measured effect biomarkers. The aim of this scoping review was to comprehensively summarise the research status on the use of Machine Learning for precision dosing in anticancer drug therapy. Methods This scoping review was conducted in accordance with the interim guidance by Cochrane and the Joanna Briggs Institute. We systematically searched the databases Medline (via PubMed), Embase and the Cochrane Library for research articles and reviews including results published after 2016. Results were reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) checklist. Results A total of 17 relevant studies was identified. In 12 of the included studies, Reinforcement Learning methods were used, including Classical, Deep, Double Deep and Conservative Q-Learning and Fuzzy Reinforcement Learning. Furthermore, classical Machine Learning methods were compared in terms of their performance and an artificial intelligence platform based on parabolic equations was used to guide dosing prospectively and retrospectively, albeit only in a limited number of patients. Due to the significantly different algorithm structures, a meaningful comparison between the various Machine Learning approaches was not possible. Conclusion Overall, this review emphasises the clinical relevance of Machine Learning methods for anticancer drug dose optimisation, as many algorithms have shown promising results enabling model-free predictions with the potential to maximise efficacy and minimise toxicity when compared to standard protocols.
- Open Access
Comparing Scientific Machine Learning with Population Pharmacokinetic and Classical Machine Learning Approaches for Prediction of Drug Concentrations
A variety of classical machine learning (ML) approaches have been developed over the past decade aiming to individualize drug dosages based on measured plasma concentrations. However, the interpretability of these models is challenging as they do not incorporate information on pharmacokinetic (PK) drug disposition. In this work we compare drug plasma concentraton predictions of well-known population PK (PopPK) modeling with classical machine learning models and a newly proposed scientific machine learning (MMPK-SciML) framework. MMPK-SciML lets us estimate PopPK parameters and their inter-individual variability (IIV) using multimodal covariate data of each patient and does not require assumptions about the underlying covariate relationships. A dataset of 541 fluorouracil (5FU) plasma concentrations as example for an intravenously administered drug and a dataset of 302 sunitinib and its active metabolite concentrations each, as example for an orally administered drug were used for analysis. Whereas classical machine learning models were not able to describe the data sufficiently, MMPK-SciML allowed us to obtain accurate drug plasma concentration predictions for test patients. In case of 5FU, goodness-of-fit shows that the MMPK-SciML approach predicts drug plasma concentrations more accurately than PopPK models. For sunitinib, we observed slightly less accurate drug concentration predictions compared to PopPK. Overall, MMPK-SciML has shown promising results and should therefore be further investigated as a valuable alternative to classical PopPK modeling, provided there is sufficient training data.
Open AccessComparing Scientific Machine Learning with Population Pharmacokinetic and Classical Machine Learning Approaches for Prediction of Drug Concentrations
A variety of classical machine learning (ML) approaches have been developed over the past decade aiming to individualize drug dosages based on measured plasma concentrations. However, the interpretability of these models is challenging as they do not incorporate information on pharmacokinetic (PK) drug disposition. In this work we compare drug plasma concentraton predictions of well-known population PK (PopPK) modeling with classical machine learning models and a newly proposed scientific machine learning (MMPK-SciML) framework. MMPK-SciML lets us estimate PopPK parameters and their inter-individual variability (IIV) using multimodal covariate data of each patient and does not require assumptions about the underlying covariate relationships. A dataset of 541 fluorouracil (5FU) plasma concentrations as example for an intravenously administered drug and a dataset of 302 sunitinib and its active metabolite concentrations each, as example for an orally administered drug were used for analysis. Whereas classical machine learning models were not able to describe the data sufficiently, MMPK-SciML allowed us to obtain accurate drug plasma concentration predictions for test patients. In case of 5FU, goodness-of-fit shows that the MMPK-SciML approach predicts drug plasma concentrations more accurately than PopPK models. For sunitinib, we observed slightly less accurate drug concentration predictions compared to PopPK. Overall, MMPK-SciML has shown promising results and should therefore be further investigated as a valuable alternative to classical PopPK modeling, provided there is sufficient training data.
- Open Access
A dynamic ensemble model for short-term forecasting in pandemic situations
During the COVID-19 pandemic, many hospitals reached their capacity limits and could no longer guarantee treatment of all patients. At the same time, governments endeavored to take sensible measures to stop the spread of the virus while at the same time trying to keep the economy afloat. Many models extrapolating confirmed cases and hospitalization rate over short periods of time have been proposed, including several ones coming from the field of machine learning. However, the highly dynamic nature of the pandemic with rapidly introduced interventions and new circulating variants imposed non-trivial challenges for the generalizability of such models. In the context of this paper, we propose the use of ensemble models, which are allowed to change in their composition or weighting of base models over time and can thus adapt to highly dynamic pandemic or epidemic situations. In that regard, we also explored the use of secondary metadata - Google searches - to inform the ensemble model. We tested our approach using surveillance data from COVID-19, Influenza, and hospital syndromic surveillance of severe acute respiratory infections (SARI). In general, we found ensembles to be more robust than the individual models. Altogether we see our work as a contribution to enhance the preparedness for future pandemic situations.
Open AccessA dynamic ensemble model for short-term forecasting in pandemic situations
During the COVID-19 pandemic, many hospitals reached their capacity limits and could no longer guarantee treatment of all patients. At the same time, governments endeavored to take sensible measures to stop the spread of the virus while at the same time trying to keep the economy afloat. Many models extrapolating confirmed cases and hospitalization rate over short periods of time have been proposed, including several ones coming from the field of machine learning. However, the highly dynamic nature of the pandemic with rapidly introduced interventions and new circulating variants imposed non-trivial challenges for the generalizability of such models. In the context of this paper, we propose the use of ensemble models, which are allowed to change in their composition or weighting of base models over time and can thus adapt to highly dynamic pandemic or epidemic situations. In that regard, we also explored the use of secondary metadata - Google searches - to inform the ensemble model. We tested our approach using surveillance data from COVID-19, Influenza, and hospital syndromic surveillance of severe acute respiratory infections (SARI). In general, we found ensembles to be more robust than the individual models. Altogether we see our work as a contribution to enhance the preparedness for future pandemic situations.
- Open Access
- 16 October 2023
- Published by Wiley in CPT: Pharmacometrics & Systems Pharmacology
- Vol. 13 (1) , 41-53
Integrating machine learning with pharmacokinetic models: Benefits of scientific machine learning in adding neural networks components to existing PK models
Recently, the use of machine‐learning (ML) models for pharmacokinetic (PK) modeling has grown significantly. Although most of the current approaches use ML techniques as black boxes, there are only a few that have proposed interpretable architectures which integrate mechanistic knowledge. In this work, we use as the test case a one‐compartment PK model using a scientific machine learning (SciML) framework and consider learning an unknown absorption using neural networks, while simultaneously estimating other parameters of drug distribution and elimination. We generate simulated data with different sampling strategies to show that our model can accurately predict concentrations in extrapolation tasks, including new dosing regimens with different sparsity levels, and produce reliable forecasts even for new patients. By using a scenario of fitting PK data with complex absorption, we demonstrate that including known physiological structure into an SciML model allows us to obtain highly accurate predictions while preserving the interpretability of classical compartmental models.
- 16 October 2023
- Published by Wiley in CPT: Pharmacometrics & Systems Pharmacology
- Vol. 13 (1) , 41-53
Open AccessIntegrating machine learning with pharmacokinetic models: Benefits of scientific machine learning in adding neural networks components to existing PK models
Recently, the use of machine‐learning (ML) models for pharmacokinetic (PK) modeling has grown significantly. Although most of the current approaches use ML techniques as black boxes, there are only a few that have proposed interpretable architectures which integrate mechanistic knowledge. In this work, we use as the test case a one‐compartment PK model using a scientific machine learning (SciML) framework and consider learning an unknown absorption using neural networks, while simultaneously estimating other parameters of drug distribution and elimination. We generate simulated data with different sampling strategies to show that our model can accurately predict concentrations in extrapolation tasks, including new dosing regimens with different sparsity levels, and produce reliable forecasts even for new patients. By using a scenario of fitting PK data with complex absorption, we demonstrate that including known physiological structure into an SciML model allows us to obtain highly accurate predictions while preserving the interpretability of classical compartmental models.