{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T10:08:22Z","timestamp":1790071702768,"version":"4.0.1"},"reference-count":48,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T00:00:00Z","timestamp":1790035200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Pharmacol."],"abstract":"<jats:p>\n                    Therapeutic drug monitoring (TDM) starts from a single measured concentration, which population pharmacokinetics (PopPK) with Bayesian maximum\n                    <jats:italic>a posteriori<\/jats:italic>\n                    (MAPB) estimation converts into a revised dose. Since roughly 2021, two research lines have been converging. Time-series deep learning models read longitudinal electronic health record (EHR) data and predict concentrations directly, led by long short-term memory (LSTM) networks, neural ordinary differential equations (neural-ODEs), and attention-based architectures. Physics-informed neural networks (PINNs), scientific machine learning (SciML), and related mechanism-aware approaches embed pharmacokinetic knowledge into neural networks, chasing accuracy without giving up interpretability. We survey both strands from 2019 to 2026, covering foundational papers, representative applications, and the open-source ecosystem. We organize the mechanism-learning relationship into three levels: loose coupling feeds pharmacokinetic-derived features into a machine learning model; medium coupling builds mechanistic structure into the architecture; tight coupling writes the mechanism into the loss function as differentiable constraints. We compare the three across predictive accuracy, extrapolation, interpretability, data needs, computational cost, and clinical acceptability. A study-level evidence map and a critical appraisal of the primary statistics underpin a nuanced picture: mechanism constraints consistently improve extrapolation and identifiability and keep the physics consistent, but in-distribution accuracy gains over well-tuned PopPK + MAPB are modest, and several headline results rest on preprints or synthetic-data validation. We also locate each study on an evidence hierarchy separating concentration prediction from dose recommendation and clinical benefit. Open problems include data sparsity, identifiability, uncertainty quantification, multi-drug generalization, regulatory acceptance, and foundation models. We close with a roadmap toward mechanism-informed, clinically deployable precision-dosing systems. Throughout, pediatric antiepileptic drug (AED) TDM, above all lacosamide, lamotrigine, and oxcarbazepine, serves as the guiding clinical scenario; narrow therapeutic windows, developmentally heterogeneous sampling, and multi-drug comorbidity make it the clearest proving ground for the coupling framework, and we map the evidence gaps for these drugs.\n                  <\/jats:p>","DOI":"10.3389\/fphar.2026.1963219","type":"journal-article","created":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T09:31:46Z","timestamp":1790069506000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["From time-series deep learning to physics-constrained population pharmacokinetics: AI-driven prediction of drug concentrations and exposure for model-informed dosing"],"prefix":"10.3389","volume":"17","author":[{"given":"Yichao","family":"Xu","sequence":"first","affiliation":[{"name":"Center of Clinical Pharmacology, The Second Affiliated Hospital of Zhejiang University, School of Medicine","place":["Hangzhou, Zhejiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zourong","family":"Ruan","sequence":"additional","affiliation":[{"name":"Center of Clinical Pharmacology, The Second Affiliated Hospital of Zhejiang University, School of Medicine","place":["Hangzhou, Zhejiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bo","family":"Jiang","sequence":"additional","affiliation":[{"name":"Center of Clinical Pharmacology, The Second Affiliated Hospital of Zhejiang University, School of Medicine","place":["Hangzhou, Zhejiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,22]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"303","DOI":"10.1146\/annurev.pharmtox.48.113006.094708","article-title":"Mechanism-based concepts of size and maturity in pharmacokinetics","volume":"48","author":"Anderson","year":"2008","journal-title":"Annu. Rev. Pharmacol. Toxicol."},{"key":"B2","doi-asserted-by":"publisher","first-page":"737","DOI":"10.1136\/archdischild-2013-303720","article-title":"Understanding dosing: children are small adults, neonates are immature children","volume":"98","author":"Anderson","year":"2013","journal-title":"Arch. Dis. Child."},{"key":"B3","volume-title":"GitHub","year":"2026"},{"key":"B4","volume-title":"Physics-Informed Neural Networks for Chemotherapy Pharmacokinetics: Benchmarking the Clinical Estimator and Exposing Parameter Identifiability","author":"Bisht","year":"2026"},{"key":"B5","doi-asserted-by":"publisher","first-page":"6571","DOI":"10.48550\/arXiv.1806.07366","article-title":"Neural ordinary differential equations","volume":"31","author":"Chen","year":"2018","journal-title":"Adv. Neural Inf. Process Syst."},{"key":"B6","doi-asserted-by":"publisher","first-page":"1333","DOI":"10.1007\/s00228-025-03874-y","article-title":"Dosing prediction of valproic acid in pediatric patients with epilepsy: population pharmacokinetic model or machine learning model?","volume":"81","author":"Chen","year":"","journal-title":"Eur. J. Clin. Pharmacol."},{"key":"B7","doi-asserted-by":"publisher","first-page":"108","DOI":"10.3870\/j.issn.1004-0781.2025.01.014","article-title":"External validation and precision drug use of the pharmacokinetic model of lacosamide in foreign pediatric populations","volume":"44","author":"Chen","year":"","journal-title":"Her. Med."},{"key":"B8","doi-asserted-by":"publisher","first-page":"1898","DOI":"10.6039\/j.issn.1001-0408.2026.14.17","article-title":"Prediction model of tacrolimus blood concentration based on population pharmacokinetics and machine learning","volume":"37","author":"Chen","year":"2026","journal-title":"China Pharm."},{"key":"B9","doi-asserted-by":"publisher","first-page":"749","DOI":"10.1007\/s40262-017-0594-5","article-title":"A physiologically based pharmacokinetic model for pregnant women to predict the pharmacokinetics of drugs metabolized via several enzymatic pathways","volume":"57","author":"Dallmann","year":"2018","journal-title":"Clin. Pharmacokinet."},{"key":"B10","doi-asserted-by":"publisher","first-page":"1859","DOI":"10.3390\/pharmaceutics15071859","article-title":"Model-informed precision dosing software tools for dosage regimen individualization: a scoping review","volume":"15","author":"Del Valle-Moreno","year":"2023","journal-title":"Pharmaceutics"},{"key":"B11","volume-title":"Reflection Paper on the Use of Artificial Intelligence (AI) in the Medicinal Product Lifecycle","year":"2024"},{"key":"B12","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1007\/s40262-012-0020-y","article-title":"Benchmarking therapeutic drug monitoring software: a review of available computer tools","volume":"52","author":"Fuchs","year":"2013","journal-title":"Clin. Pharmacokinet."},{"key":"B13","doi-asserted-by":"publisher","first-page":"e70062","DOI":"10.1002\/phar.70062","article-title":"Nephrocast-V: a deep learning model for the prediction of vancomycin trough concentration using electronic health record data","volume":"46","author":"Ghanbari","year":"2026","journal-title":"Pharmacotherapy"},{"key":"B14","volume-title":"Pharmaconode: Pharmacokinetic Modeling Using nNeural ODEs","author":"Giacometti","year":"2026"},{"key":"B15","doi-asserted-by":"publisher","first-page":"602","DOI":"10.3390\/e27060602","article-title":"Leveraging neural ODEs for population pharmacokinetics of dalbavancin in sparse clinical data","volume":"27","author":"Giacometti","year":"2025","journal-title":"Entropy"},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.21203\/rs.3.rs-9308510\/v1","article-title":"Sim2Real-PK framework for precision dosing: integrating pharmacokinetic mechanisms and deep learning to decode clinical heterogeneity from sparse data","author":"Han","year":"2026","journal-title":"Res. Square"},{"key":"B17","doi-asserted-by":"publisher","first-page":"1417","DOI":"10.1038\/aps.2012.118","article-title":"Population pharmacokinetics of lamotrigine in Chinese children with epilepsy","volume":"33","author":"He","year":"2012","journal-title":"Acta Pharmacol. Sin."},{"key":"B18","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41597-022-01899-x","article-title":"MIMIC-IV, a freely accessible electronic health record dataset","volume":"10","author":"Johnson","year":"2023","journal-title":"Sci. Data"},{"key":"B19","doi-asserted-by":"publisher","first-page":"620","DOI":"10.3389\/fphar.2020.00620","article-title":"Software tools for model-informed precision dosing: how well do they satisfy the needs?","volume":"11","author":"Kantasiripitak","year":"2020","journal-title":"Front. Pharmacol."},{"key":"B20","doi-asserted-by":"publisher","first-page":"422","DOI":"10.1038\/s42254-021-00314-5","article-title":"Physics-informed machine learning","volume":"3","author":"Karniadakis","year":"2021","journal-title":"Nat. Rev. Phys."},{"key":"B21","doi-asserted-by":"publisher","first-page":"1157","DOI":"10.1056\/NEJMra035092","article-title":"Developmental pharmacology\u2014drug disposition, action, and therapy in infants and children","volume":"349","author":"Kearns","year":"2003","journal-title":"N. Engl. J. Med."},{"key":"B22","doi-asserted-by":"publisher","first-page":"1139","DOI":"10.3390\/pharmaceutics15041139","article-title":"Deep learning methods applied to drug concentration prediction of olanzapine","volume":"15","author":"Khusial","year":"2023","journal-title":"Pharmaceutics"},{"key":"B23","doi-asserted-by":"publisher","first-page":"878","DOI":"10.1002\/psp4.12473","article-title":"Open systems pharmacology community\u2014an open access, open source, open science approach to modeling and simulation in pharmaceutical sciences","volume":"8","author":"Lippert","year":"2019","journal-title":"CPT Pharmacometrics Syst. Pharmacol."},{"key":"B24","doi-asserted-by":"publisher","first-page":"4021","DOI":"10.2147\/DDDT.S495050","article-title":"A real-time plasma concentration prediction model for voriconazole in elderly patients via machine learning combined with population pharmacokinetics","volume":"19","author":"Liu","year":"","journal-title":"Drug Des. Devel Ther."},{"key":"B25","volume-title":"Dynamic Graph Neural Networks for Physiological Based Pharmacokinetic Modeling: a Novel Data Driven Approach to Drug Concentration Prediction","author":"Liu","year":""},{"key":"B26","volume-title":"Physiologically Informed Deep Learning: a Multi-Scale Framework for Next-Generation PBPK Modeling","author":"Liu","year":"2026"},{"key":"B27","doi-asserted-by":"publisher","first-page":"102804","DOI":"10.1016\/j.isci.2021.102804","article-title":"Neural-ODE for pharmacokinetics modeling and its advantage to alternative machine learning models in predicting new dosing regimens","volume":"24","author":"Lu","year":"","journal-title":"iScience"},{"key":"B28","doi-asserted-by":"publisher","first-page":"696","DOI":"10.1038\/s42256-021-00357-4","article-title":"Deep learning prediction of patient response time course from early data via neural-pharmacokinetic\/pharmacodynamic modelling","volume":"3","author":"Lu","year":"","journal-title":"Nat. Mach. Intell."},{"key":"B29","doi-asserted-by":"publisher","first-page":"1482","DOI":"10.1111\/bcp.14801","article-title":"Machine learning in pharmacometrics: opportunities and challenges","volume":"88","author":"McComb","year":"2022","journal-title":"Br. J. Clin. Pharmacol."},{"key":"B30","volume-title":"GitHub","year":"2026"},{"key":"B31","doi-asserted-by":"publisher","first-page":"1134","DOI":"10.1016\/j.automatica.2006.12.024","article-title":"Nonparametric identification of population models via Gaussian processes","volume":"43","author":"Neve","year":"2007","journal-title":"Automatica"},{"key":"B32","volume-title":"Vancomycin_Dose_Predictions","author":"O'Callaghan","year":"2026"},{"key":"B33","doi-asserted-by":"publisher","first-page":"571","DOI":"10.1002\/cpt.1642","article-title":"Evaluation of CYP2C19 genotype-guided voriconazole prophylaxis after allogeneic hematopoietic cell transplant","volume":"107","author":"Patel","year":"2020","journal-title":"Clin. Pharmacol. Ther."},{"key":"B34","doi-asserted-by":"publisher","first-page":"706","DOI":"10.5414\/CP202078","article-title":"Population pharmacokinetics of oxcarbazepine active metabolite in Chinese children with epilepsy","volume":"52","author":"Peng","year":"2014","journal-title":"Int. J. Clin. Pharmacol. Ther."},{"key":"B35","doi-asserted-by":"publisher","first-page":"2130","DOI":"10.1212\/01.wnl.0000289511.20864.2a","article-title":"Lamotrigine in pregnancy: clearance, therapeutic drug monitoring, and seizure frequency","volume":"70","author":"Pennell","year":"2008","journal-title":"Neurology"},{"key":"B36","doi-asserted-by":"publisher","first-page":"106897","DOI":"10.1016\/j.eplepsyres.2022.106897","article-title":"Effect of UGT1A4, UGT2B7, UGT2B15, UGT2B17 and ABC1B polymorphisms on lamotrigine metabolism in Danish patients","volume":"183","author":"Petrenaite","year":"2022","journal-title":"Epilepsy Res."},{"key":"B37","doi-asserted-by":"publisher","first-page":"686","DOI":"10.1016\/j.jcp.2018.10.045","article-title":"Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations","volume":"378","author":"Raissi","year":"2019","journal-title":"J. Comput. Phys."},{"key":"B38","volume-title":"Gaussian Processes for Machine Learning","author":"Rasmussen","year":"2006"},{"key":"B39","doi-asserted-by":"publisher","first-page":"2695","DOI":"10.1111\/bcp.13392","article-title":"Population pharmacokinetics of oxcarbazepine and its monohydroxy derivative in epileptic children","volume":"83","author":"Rodrigues","year":"2017","journal-title":"Br. J. Clin. Pharmacol."},{"key":"B40","volume-title":"GitHub","year":"2026"},{"key":"B41","doi-asserted-by":"publisher","first-page":"1344","DOI":"10.1002\/jps.2600711211","article-title":"Bayesian individualization of pharmacokinetics: simple implementation and comparison with non-Bayesian methods","volume":"71","author":"Sheiner","year":"1982","journal-title":"J. Pharm. Sci."},{"key":"B42","doi-asserted-by":"publisher","first-page":"294","DOI":"10.1002\/cpt1979263294","article-title":"Forecasting individual pharmacokinetics","volume":"26","author":"Sheiner","year":"1979","journal-title":"Clin. Pharmacol. Ther."},{"key":"B43","volume-title":"GitHub","year":"2026"},{"key":"B44","doi-asserted-by":"publisher","first-page":"11","DOI":"10.1007\/s10928-026-10019-w","article-title":"A physics-informed neural network approach for estimating population-level pharmacokinetic parameters from aggregated concentration data","volume":"53","author":"Tsiros","year":"2026","journal-title":"J. Pharmacokinet. Pharmacodyn."},{"key":"B45","article-title":"Considerations for the use of artificial intelligence to support regulatory decision-making for drug and biological products","volume-title":"Draft Guidance for Industry","year":"2025"},{"key":"B46","doi-asserted-by":"publisher","first-page":"41","DOI":"10.1002\/psp4.13054","article-title":"Integrating machine learning with pharmacokinetic models: benefits of scientific machine learning in adding neural networks components to existing PK models","volume":"13","author":"Valderrama","year":"2024","journal-title":"CPT Pharmacometrics Syst. Pharmacol."},{"key":"B47","doi-asserted-by":"publisher","first-page":"178","DOI":"10.1002\/jcph.1340","article-title":"Population pharmacokinetics of adjunctive lacosamide in pediatric patients with epilepsy","volume":"59","author":"Winkler","year":"2019","journal-title":"J. Clin. Pharmacol."},{"key":"B48","doi-asserted-by":"publisher","first-page":"859","DOI":"10.13286\/j.1001-5213.2025.08.02","article-title":"Prediction of AUC24h of vancomycin by population pharmacokinetics combined with XGBoost algorithms","volume":"45","author":"Zhang","year":"2025","journal-title":"Chin. J. Hosp. Pharm."}],"container-title":["Frontiers in Pharmacology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fphar.2026.1963219\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T09:31:56Z","timestamp":1790069516000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fphar.2026.1963219\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,22]]},"references-count":48,"alternative-id":["10.3389\/fphar.2026.1963219"],"URL":"https:\/\/doi.org\/10.3389\/fphar.2026.1963219","relation":{},"ISSN":["1663-9812"],"issn-type":[{"value":"1663-9812","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,22]]},"article-number":"1963219"}}