Physics‐informed neural networks guided modelling and multiobjective optimization of a mAb production process DOI

Md Nasre Alam,

Anurag Anurag,

Neelesh Gangwar

et al.

The Canadian Journal of Chemical Engineering, Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 1, 2024

Abstract In this paper, we aim to correlate various process and product quality attributes of a mammalian cell culture with parameters. To achieve this, employed physics‐informed neural networks that solve the governing ordinary differential equations comprising independent variables (inputs‐ time, flow rates, volume) dependent (outputs‐ viable density, dead glucose concentration, lactate monoclonal antibody concentration). The proposed model surpasses prediction accuracy capabilities other commonly used modelling approaches, such as multilayer perceptron model. It has higher R ‐squared ( 2 ), lower root mean square error, absolute error than for all output (viable viability, Furthermore, incorporate Bayesian optimization study maximize density concentration. Single objective weighted sum multiobjective were carried out concentration in separate (single optimization) combined (multiobjective forms. An increment 13.01% 18.57% respectively, projected under single optimization, 46.32% 67.86%, compared base case. This highlights potential networks‐based upstream processing cell‐based antibodies biopharmaceutical operations.

Language: Английский

A Bayesian regularization neural network procedure to solve the language learning system DOI
Zulqurnain Sabir,

Samir Khansa,

Ghida Baltaji

et al.

Knowledge-Based Systems, Journal Year: 2025, Volume and Issue: unknown, P. 112997 - 112997

Published: Jan. 1, 2025

Language: Английский

Citations

5
A novel radial basis neural network for the Zika virus spreading model DOI
Zulqurnain Sabir,

Tino Bou Rada,

Zeinab Kassem

et al.

Computational Biology and Chemistry, Journal Year: 2024, Volume and Issue: 112, P. 108162 - 108162

Published: July 26, 2024

Language: Английский

Citations

10
An adaptive algorithm for numerically solving fractional partial differential equations using Hermite wavelet artificial neural networks DOI

Amina Ali,

Norazak Senu, Nadihah Wahi

et al.

Communications in Nonlinear Science and Numerical Simulation, Journal Year: 2024, Volume and Issue: 137, P. 108121 - 108121

Published: June 1, 2024

Language: Английский

Citations

5
Optimizing trigger timing in minimal ovarian stimulation for In Vitro fertilization using machine learning models with random search hyperparameter tuning DOI
Nayeli Areli Pérez-Padilla,

Rodolfo Garcia-Sánchez,

Omar Avalos

et al.

Computers in Biology and Medicine, Journal Year: 2024, Volume and Issue: 179, P. 108856 - 108856

Published: July 24, 2024

Language: Английский

Citations

4
On QSPR analysis of glaucoma drugs using machine learning with XGBoost and regression models DOI

Lina Huang,

Khawlah Alhulwah,

Muhammad Farhan Hanif

et al.

Computers in Biology and Medicine, Journal Year: 2025, Volume and Issue: 187, P. 109731 - 109731

Published: Jan. 28, 2025

Language: Английский

Citations

0
Disruptive Attacks on Artificial Neural Networks: A Systematic Review of Attack Techniques, Detection Methods, and Protection Strategies DOI Creative Commons
Talal Bonny, Talal Bonny, Maher Alrahhal

et al.

Intelligent Systems with Applications, Journal Year: 2025, Volume and Issue: unknown, P. 200529 - 200529

Published: April 1, 2025

Language: Английский

Citations

0
Design and experimental implementation of skewed stator structured PM generator using Multi-Objective Genetic Algorithm for gearless direct driven wind system DOI Creative Commons
İsmail Topaloğlu, Enes Bektaş, Mahmud Esad Yiǧit

et al.

Ain Shams Engineering Journal, Journal Year: 2024, Volume and Issue: unknown, P. 103067 - 103067

Published: Sept. 1, 2024

Language: Английский

Citations

1
Towards key genes identification for breast cancer survival risk with neural network models DOI Creative Commons
Gang Liu,

Xiao Yang,

Nan Li

et al.

Computational Biology and Chemistry, Journal Year: 2024, Volume and Issue: 112, P. 108143 - 108143

Published: Aug. 2, 2024

Breast cancer, one common malignant tumor all over the world, has a considerably high rate of recurrence, which endangers health and life patients. While more data have been available, how to leverage gene expression predict survival risk cancer patients identify key genes become hot topic for research. Therefore, in this work, we investigate clinical breast patients, specifically novel framework is proposed focusing on classification identification task. We firstly combine differential univariate Cox regression analysis achieve dimensional reduction data. The median time subsequently as threshold learning model based neural network trained classify Innovatively, activation region visualization technology selected tool, 20 related further analyze function these STRING database. It critical learn that, genetic biomarkers identified paper may possess value following treatment according literature findings. Importantly, Our work accomplishes objective proposing targeted approach enhancing therapeutic strategies through advanced computational techniques analysis.

Language: Английский

Citations

0
Physics‐informed neural networks guided modelling and multiobjective optimization of a mAb production process DOI

Md Nasre Alam,

Anurag Anurag,

Neelesh Gangwar

et al.

The Canadian Journal of Chemical Engineering, Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 1, 2024

Abstract In this paper, we aim to correlate various process and product quality attributes of a mammalian cell culture with parameters. To achieve this, employed physics‐informed neural networks that solve the governing ordinary differential equations comprising independent variables (inputs‐ time, flow rates, volume) dependent (outputs‐ viable density, dead glucose concentration, lactate monoclonal antibody concentration). The proposed model surpasses prediction accuracy capabilities other commonly used modelling approaches, such as multilayer perceptron model. It has higher R ‐squared ( 2 ), lower root mean square error, absolute error than for all output (viable viability, Furthermore, incorporate Bayesian optimization study maximize density concentration. Single objective weighted sum multiobjective were carried out concentration in separate (single optimization) combined (multiobjective forms. An increment 13.01% 18.57% respectively, projected under single optimization, 46.32% 67.86%, compared base case. This highlights potential networks‐based upstream processing cell‐based antibodies biopharmaceutical operations.

Language: Английский

Citations

0