Neural Network-based rheological analysis and modeling of hybrid nanoblood flow across a stenosed artery

Abstract Blood, being a complex fluid, exhibits characteristics such as yield stress, variable viscosity, shear thinning, viscoelasticity, etc., which can be captured by utilizing different non-Newtonian models. This study incorporates two renowned non-Newtonian models, namely Casson and Carreau-Yasuda rheologies, to assess the velocity and thermal properties of tetra hybrid nano blood flow via a cosine-shaped stenosed artery subject to MHD effects, Cattaneo-Christov thermal flux, radiative heat, Joule heating, and the consequence of non-uniform heat source/sink. The governing equations replicating both models are described as nonlinear PDEs and are subsequently transformed into a collection of ODEs by defining an appropriate set of similarity variables. The obtained equations are further evaluated numerically using the bvp5c MATLAB solver to obtain the solutions for velocity and temperature profiles of both models simultaneously. In addition to that, two feed-forward-oriented data-driven Artificial Neural Networks (ANN) are built to estimate the heat transfer rate and arterial wall friction index for both tetra hybrid Casson and Carreau-Yasuda fluid models. A great level of agreement among the ANN data-driven estimations and the numerical results was confirmed by the remarkably small MSE of orders 10 −6 , and R 2 values ascending to 1. Further, a repeated random-split test was carried out to ensure the robustness of the constructed ANN, and a Feature-Importance analysis was conducted to assess the level of sensitivity of the outcomes to the parameters. The feature-importance analysis revealed that for the Casson ANN model, β (mean importance of 136.05 %) and θw (133.60 %), and for the Carreau-Yasuda ANN model, M (162.48 %) and θw (137.65 %) are highly potent for outcomes of wall friction and Nusselt number, respectively. Under the considered circumstances, the tetra hybrid Casson fluid reports elevated velocity and heat transfer rates, whereas the tetra hybrid Carreau-Yasuda fluid reports enhanced temperature and wall frictional values, in a comparative framework. The key findings illustrate a significant difference between the models, emphasizing the importance of effective rheological modelling to precisely capture blood characteristics and offer a theoretical foundation for further research relevant to potential biomedical applications involving the understanding of combinational nano blood transport in stenosed arteries. Also, the numerical and ANN framework renders a potential computational tool for forthcoming biomedical heat transfer and blood flow investigations.

Authors

Institutions

Publication Details

Journal
Chemical Product and Process Modeling
Published
2026-11-12
DOI
https://doi.org/10.1515/cppm-2026-0200
Primary Topic
Nanofluid Flow and Heat Transfer
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Neural Network-based rheological analysis and modeling of hybrid nanoblood flow across a stenosed artery

Naramgari Sandeep, Ahlada Ravula
Chemical Product and Process Modeling
Nanofluid Flow and Heat Transfer
article

Neural Network-based rheological analysis and modeling of hybrid nanoblood flow across a stenosed artery

Naramgari Sandeep, Ahlada Ravula
article en

Abstract

Abstract Blood, being a complex fluid, exhibits characteristics such as yield stress, variable viscosity, shear thinning, viscoelasticity, etc., which can be captured by utilizing different non-Newtonian models. This study incorporates two renowned non-Newtonian models, namely Casson and Carreau-Yasuda rheologies, to assess the velocity and thermal properties of tetra hybrid nano blood flow via a cosine-shaped stenosed artery subject to MHD effects, Cattaneo-Christov thermal flux, radiative heat, Joule heating, and the consequence of non-uniform heat source/sink. The governing equations replicating both models are described as nonlinear PDEs and are subsequently transformed into a collection of ODEs by defining an appropriate set of similarity variables. The obtained equations are further evaluated numerically using the bvp5c MATLAB solver to obtain the solutions for velocity and temperature profiles of both models simultaneously. In addition to that, two feed-forward-oriented data-driven Artificial Neural Networks (ANN) are built to estimate the heat transfer rate and arterial wall friction index for both tetra hybrid Casson and Carreau-Yasuda fluid models. A great level of agreement among the ANN data-driven estimations and the numerical results was confirmed by the remarkably small MSE of orders 10 −6 , and R 2 values ascending to 1. Further, a repeated random-split test was carried out to ensure the robustness of the constructed ANN, and a Feature-Importance analysis was conducted to assess the level of sensitivity of the outcomes to the parameters. The feature-importance analysis revealed that for the Casson ANN model, β (mean importance of 136.05 %) and θw (133.60 %), and for the Carreau-Yasuda ANN model, M (162.48 %) and θw (137.65 %) are highly potent for outcomes of wall friction and Nusselt number, respectively. Under the considered circumstances, the tetra hybrid Casson fluid reports elevated velocity and heat transfer rates, whereas the tetra hybrid Carreau-Yasuda fluid reports enhanced temperature and wall frictional values, in a comparative framework. The key findings illustrate a significant difference between the models, emphasizing the importance of effective rheological modelling to precisely capture blood characteristics and offer a theoretical foundation for further research relevant to potential biomedical applications involving the understanding of combinational nano blood transport in stenosed arteries. Also, the numerical and ANN framework renders a potential computational tool for forthcoming biomedical heat transfer and blood flow investigations.

Chemical Product and Process Modeling
Central University of Karnataka (IN)
Nanofluid Flow and Heat Transfer
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon