Integrating AI and statistical modelling for enhanced microalgae growth in 3D Bioprinted polymeric Scaffolds-a hybrid approach
The optimization of microalgal growth within three-dimensional (3D) scaffolds presents a promising avenue for enhancing biomass production for biofuel, pharmaceutical, and biotechnological applications. This study employs Artificial Neural Networks (ANN) and Response Surface Methodology (RSM) as com...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-08-01
|
| Series: | Chemical Engineering Journal Advances |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666821125001188 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The optimization of microalgal growth within three-dimensional (3D) scaffolds presents a promising avenue for enhancing biomass production for biofuel, pharmaceutical, and biotechnological applications. This study employs Artificial Neural Networks (ANN) and Response Surface Methodology (RSM) as computational approaches to model and optimize key parameters influencing microalgal proliferation inside 3D scaffolds. ANN, a machine learning AI tool, is leveraged to capture complex, nonlinear relationships between variables, while RSM provides a structured statistical framework to fine-tune process parameters through regression modeling and experimental design. The study develops an ANN framework to predict the impact of bioprinting parameters such as cell concentration (in 106 number), sodium alginate (as SA%), crosslinker CaCl2 (in mM) and medial volume (in ml) on cell proliferation inside the hydrogel matrix. The evaluation of these strategies through comparison demonstrates their capabilities for precise predictions along with optimized operation efficiency while delivering practical results for enhancing microalgal biomass production. The findings reveal that ANN outperforms traditional RSM in capturing intricate dependencies, whereas RSM remains valuable for identifying optimal conditions with reduced computational complexity. The findings show that ANN outperforms RSM in identifying intricate patterns by offering superior generalization and prediction accuracy across nonlinear data. Due of its versatility, ANN is a more reliable tool for accurate process optimization. The research demonstrates that combined computational approaches have great potential for improving scaffold-dependent microalgae cultivation, paving the way for scalable and efficient bioprocesses. |
|---|---|
| ISSN: | 2666-8211 |