Application of artificial neural network modelling to identify severely ill patients whose aminoglycoside concentrations are likely to fall below therapeutic concentrations

Yamamura, Shigeo; Takehira, Rieko; Kawada, Keiko; Nishizawa, Kenji; Katayama, Shirou; Hirano, Masaaki; Momose, Yasunori

doi:10.1046/j.0269-4727.2003.00514.x

Cited by 16 publications

(8 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The flexible function of the ANN models allows the successful application of these models also to pharmacokinetic problems, as can be illustrated by the following examples. 1) ANN models can be employed for the prediction of peak and trough plasma levels of drugs, on the basis of observed plasma levels and parameters related to patients' conditions, as exemplified for aminoglycoside antibiotics (arbekacin sulfate and amikacin sulfate) given to patients with severe diseases (Yamamura et al 1998, 2003a & 2003b). In the given studies, the predictive performance of the ANN modeling was shown to be superior to that of a multiple linear regression analysis.…”

Section: Methods Based On Concept Of Artificial Neural Networkmentioning

confidence: 99%

New Mathematical Methods in Pharmacokinetic Modeling

Ďurišová

Dedik

2005

Basic Clin Pharmacol Toxicol

View full text Add to dashboard Cite

In recent years, several new methods for the mathematical modeling have gradually emerged in pharmacokinetics, and the development of pharmacokinetic models based on these methods has become one of the most rapidly growing and exciting application-oriented sub-disciplines of the mathematical modeling. The goals of our MiniReview are twofold: i) to briefly outline fundamental ideas of some new modeling methods that have not been widely utilized in pharmacokinetics as yet, i.e. the methods based on the following concepts: linear time-invariant dynamic system, artificialneural-network, fuzzy-logic, and fractal; ii) to arouse the interest of pharmacological, toxicological, and pharmaceutical scientists in the given methods, by sketching some application examples which indicate the good performance and perspective of these methods in solving pharmacokinetic problems.It often occurs in science that tools, knowledge, or techniques developed in one field enable advances in other fields. This applies also to methods for building mathematical models, i.e. tools that have been developed predominantly in the field of mathematics, computer science, and system engineering. In contrast to animal or disease models commonly used in pharmacology, mathematical models are mathematical structures which, on abstracting from reality, allow formulations in mathematical terms of essential assumptions believed to underlie particular actual-world problems. The mathematical modeling relies on: i) the sound understanding and appreciation of the problem under study; ii) the realistic mathematical representation of the important phenomena; iii) the finding of useful solutions; iv) the interpretation of the mathematical results, the model-based prediction or simulation, and so on. All that is required for the practical use of mathematical models is to understand these models properly and then to apply them to study phenomena of the interest in various fields of science, including pharmacokinetics. In the latter field, mathematical models have historically played a vital role, and methods for the development models such as deterministic linear compartment models, physiologically-based models, and population models have become quite common. Along with this, other modeling methods have gradually emerged in pharmacokinetics over recent years, and the Author for correspondence: Mária Ď urišová, Institute of Experimental Pharmacology, Slovak Academy of Sciences, Dubravska cesta 9, SK-841 04 Bratislava 4, Slovak Republic (fax π42 12 5477 5928, e-mail maria.durisova/savba.sk).development of pharmacokinetic models has become one of the most rapidly growing and exciting applicationoriented sub-disciplines of the mathematical modeling.The goals of this MiniReview are twofold: i) to briefly outline fundamental ideas of some mathematical methods that have not been widely utilized in pharmacokinetics so far, i.e. the methods based on the following concepts: linear time-invariant dynamic system, artificial-neural-network, fuzzy-logic, and fractal; ...

show abstract

Section: Methods Based On Concept Of Artificial Neural Networkmentioning

confidence: 99%

New Mathematical Methods in Pharmacokinetic Modeling

Ďurišová

Dedik

2005

Basic Clin Pharmacol Toxicol

View full text Add to dashboard Cite

show abstract

“…Several new methods for mathematical modelling have emerged in pharmacokinetics and have shown good performance in solving pharmacokinetic problems [6–8]. One of them, artificial neural network (ANN), is a powerful empirical pattern‐recognition and mapping tool for approximation of complex nonlinear relationships.…”

Section: Introductionmentioning

confidence: 99%

Population pharmacokinetic and pharmacodynamic models of remifentanil in healthy volunteers using artificial neural network analysis

Kang

Poynton

Kim

et al. 2007

Brit J Clinical Pharma

View full text Add to dashboard Cite

AimsAn ordinary sigmoid Emax model could not predict overshoot of electroencephalographic approximate entropy (ApEn) during recovery from remifentanil effect in our previous study. The aim of this study was to evaluate the ability of an artificial neural network (ANN) to predict ApEn overshoot and to evaluate the predictive performance of the pharmacokinetic model, and pharmacodynamic models of ANN with respect to data used. MethodsUsing a reduced number of ApEn instances (n = 1581) to make NONMEM modelling feasible and complete ApEn data (n = 24 509), the presence of overshoot was assessed. A total of 1077 measured remifentanil concentrations and ApEn data, and a total of 24 509 predicted concentrations and ApEn data were used in the pharmacodynamic model A and B of ANN, respectively. The testing subset of model B (n = 7352) was used to evaluate the ability of ANN to predict overshoot of ApEn. Mean squared error (MSE) was calculated to evaluate the predictive performance of the ANN models. ResultsWith complete ApEn data, ApEn overshoot was observed in 66.7% of subjects, but only in 37% with a reduced number of ApEn instances. The ANN model B predicted 77.8% of ApEn overshoot. MSE (95% confidence interval) was 57.1 (3.22, 71.03) for the pharmacokinetic model, 0.148 (0.004, 0.007) for model A and 0.0018 (0.0017, 0.0019) for model B. ConclusionsThe reduced ApEn instances interfered with the approximation of true electroencephalographic response. ANN predicted 77.8% of ApEn overshoot. The predictive performance of model B was significantly better than that of model A.

show abstract

“…In terms of the use of AI in clinical pharmacology, a series of studies were published in the late 1990s and early 2000s investigating the usage of AI, in particular, neural network for predicting pharmacokinetic concentrations of mainly antibiotics and immunosuppressants, to guide medication dosage based on patient characteristics. [1][2][3][4][5][6][7][8] In more recent years, AI is experiencing a resurgence in drug development due to the availability of increased computing power, as the strength of AI over traditional analytical methodologies is mainly its ability to deal with large, not easily interpretable data, from which multiple comparisons and interactions can be made from the number of possible covariates in clinical studies. The ultimate aim of AI approaches is model prediction, and AI algorithms have the ability to incorporate many important but collinear variables simultaneously.…”

mentioning

confidence: 99%

“…The major success of artificial intelligence (AI) in medicine so far has been in the augmentation of disease diagnosis based on imaging data, and AI is increasingly applied to improve prediction in drug development, such as candidate selection and analysis of genomics data. In terms of the use of AI in clinical pharmacology, a series of studies were published in the late 1990s and early 2000s investigating the usage of AI, in particular, neural network for predicting pharmacokinetic concentrations of mainly antibiotics and immunosuppressants, to guide medication dosage based on patient characteristics 1–8 . In more recent years, AI is experiencing a resurgence in drug development due to the availability of increased computing power, as the strength of AI over traditional analytical methodologies is mainly its ability to deal with large, not easily interpretable data, from which multiple comparisons and interactions can be made from the number of possible covariates in clinical studies.…”

mentioning

confidence: 99%

Application of Machine Learning for Tumor Growth Inhibition – Overall Survival Modeling Platform

Chan

Zhou

Wang

et al. 2020

CPT Pharmacom & Syst Pharma

View full text Add to dashboard Cite

Machine learning (ML) was used to leverage tumor growth inhibition (TGI) metrics to characterize the relationship with overall survival (OS) as a novel approach and to compare with traditional TGI‐OS modeling methods. Historical dataset from a phase III non‐small cell lung cancer study (OAK, atezolizumab vs. docetaxel, N = 668) was used. ML methods support the validity of TGI metrics in predicting OS. With lasso, the best model with TGI metrics outperforms the best model without TGI metrics. Boosting was the best linear ML method for this dataset with reduced estimation bias and lowest Brier score, suggesting better prediction accuracy. Random forest did not outperform linear ML methods despite hyperparameter optimization. Kernel machine was marginally the best nonlinear ML method for this dataset and uncovered nonlinear and interaction effects. Nonlinear ML may improve prediction by capturing nonlinear effects and covariate interactions, but its predictive performance and value need further evaluation with larger datasets.

show abstract

Application of artificial neural network modelling to identify severely ill patients whose aminoglycoside concentrations are likely to fall below therapeutic concentrations

Cited by 16 publications

References 19 publications

New Mathematical Methods in Pharmacokinetic Modeling

New Mathematical Methods in Pharmacokinetic Modeling

Population pharmacokinetic and pharmacodynamic models of remifentanil in healthy volunteers using artificial neural network analysis

Application of Machine Learning for Tumor Growth Inhibition – Overall Survival Modeling Platform

Contact Info

Product

Resources

About