[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117075-en":3,"doc-seo-117075-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},117075,4810365810221,"Aurora","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Population pharmacokinetic model selection assisted by machine learning","A fit-for-purpose structural and statistical model is the prerequisite for population pharmacometric model development. The study examines how supervised machine learning can support this complex, computationally intensive task, comparing the classical pharmacometric workflow with genetic algorithms and neural networks. Using simulated pharmacokinetic data, genetic algorithm performance is evaluated via log-likelihood-based fitness, while neural networks are trained with mean squared error or binary cross-entropy loss. Results show accurate selection driven by statistical rules, with neural network classification delivering the highest accuracy and substantial computational gains for large datasets and complex models.","Journal of Pharmacokinetics and Pharmacodynamics [https://doi.org/10.1007/s10928-021-09793-6](https://doi.org/10.1007/s10928-021-09793-6)  \nPopulation pharmacokinetic model selection assisted by machine learning  \nEmeric Sibieude1,2 • Akash Khandelwal3 • Pascal Girard2 • Jan S. Hesthaven4 • Nadia Terranova2   \nReceived: 1 February 2021/Accepted: 17 October 2021  \n􀀂 The Author(s) 2021  \nAbstract  \nA fit-for-purpose structural and statistical model is the first major requirement in population pharmacometric model development. In this manuscript we discuss how this complex and computationally intensive task could benefit from supervised machine learning algorithms. We compared the classical pharmacometric approach with two machine learning methods, genetic algorithm and neural networks, in different scenarios based on simulated pharmacokinetic data. Genetic algorithm performance was assessed using a fitness function based on log-likelihood, whilst neural networks were trained using mean square error or binary cross-entropy loss. Machine learning provided a selection based only on statistical rules and achieved accurate selection. The minimization process of genetic algorithm was successful at allowing the algorithm to select plausible models. Neural network classification tasks achieved the most accurate results. Neural network regression tasks were less precise than neural network classification and genetic algorithm methods. The computational gain obtained by using machine learning was substantial, especially in the case of neural networks. We demonstrated that machine learning methods can greatly increase the efficiency of pharmacokinetic population model selection in case of large datasets or complex models requiring long run-times. Our results suggest that machine learning approaches can achieve a first fast selection of models which can be followed by more conventional pharmacometric approaches.  \nKeywords Deep learning 􀀂 Genetic algorithm 􀀂 Model-informed drug discovery and development 􀀂 Neural network 􀀂 Pharmacometrics 􀀂 Population PK/PD  \nIntroduction  \nModel-informed drug discovery and development (MID3) is a process which applies quantitative modeling to preclinical and clinical data to accelerate and optimize drug development [1] . MID3 plays a key role at each stage of drug development by quantifying the risk–benefit ratio of the treatment in the general population and in sub-  \n& Nadia Terranova [nadia.terranova@merckgroup.com](nadia.terranova@merckgroup.com)  \n1 School of Basic Sciences, EPFL, Lausanne, Switzerland  \n2 Merck Institute for Pharmacometrics (an affiliate of Merck KGaA, Darmstadt, Germany), Lausanne, Switzerland  \n3 Merck KGaA, Darmstadt, Germany  \n4 Chair of Computational Mathematics and Simulation Science (MCSS), Ecole Polytechnique F´ed´erale de Lausanne (EPFL), Lausanne, Switzerland  \npopulations, therefore increasing confidence in decisionmaking and reducing development costs [2] .  \nMID3 has a large range of applications, including characterizing the drug concentration-pharmacodynamic (PD) response relationships [3], explaining drug variability by identifying clinically relevant factors which impact on desired outcomes [4], and predicting the consequences of formulation changes on drug performance [5] .  \nAmong the techniques available in MID3, population modeling is a tool which describes the relationships between patient´s physiological characteristics and model parameters governing drug concentrations, or drug response and their distribution across a population [6] . Population pharmacokinetic (PK) and PD models are used to describe relationships between a dependent variable (e.g., concentration or response) and an independent variable (e.g., time) . These models are also used to investigate sources of variability [7] . Population models favor statistical (nonlinear) mixed effect modeling techniques. This  \n1 3  \nmethodology allows the development of models containing both fixed and random effects.  \nNonl","cbCainI4915vPcIz","https://ap.wps.com/l/cbCainI4915vPcIz","pdf",1613675,1,14,"English","en",105,"# Abstract\n# Introduction\n## Model-informed drug discovery and development (MID3)\n## Population modeling and population PK/PD\n## Nonlinear mixed effects modeling and estimation tools\n## Need for model selection and opportunities for ML/DL\n## Genetic algorithms for model selection\n## Neural networks for model selection","[{\"question\":\"Why is fit-for-purpose structural and statistical modeling important in population pharmacometrics?\",\"answer\":\"It is the first major requirement for developing population pharmacometric models, because the model must be appropriate for the data and the selection process.\"},{\"question\":\"How were genetic algorithms and neural networks evaluated for model selection?\",\"answer\":\"Genetic algorithms were assessed using a fitness function based on log-likelihood, while neural networks were trained using mean squared error or binary cross-entropy loss.\"},{\"question\":\"What do the results indicate about the computational and accuracy benefits of machine learning?\",\"answer\":\"Machine learning enables statistically rule-based selection with accurate results, with substantial computational gains, especially for neural networks, and best performance from neural network classification tasks.\"}]","Population pharmacokinetic model selection assisted by machine learning | PDF",1785673571,35,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"population-pharmacokinetic-model-selection-assisted-by-machine-learning","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/population-pharmacokinetic-model-selection-assisted-by-machine-learning/117075/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-02",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is fit-for-purpose structural and statistical modeling important in population pharmacometrics?","Question",{"text":75,"@type":76},"It is the first major requirement for developing population pharmacometric models, because the model must be appropriate for the data and the selection process.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How were genetic algorithms and neural networks evaluated for model selection?",{"text":80,"@type":76},"Genetic algorithms were assessed using a fitness function based on log-likelihood, while neural networks were trained using mean squared error or binary cross-entropy loss.",{"name":82,"@type":73,"acceptedAnswer":83},"What do the results indicate about the computational and accuracy benefits of machine learning?",{"text":84,"@type":76},"Machine learning enables statistically rule-based selection with accurate results, with substantial computational gains, especially for neural networks, and best performance from neural network classification tasks.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]