Page 154 - v11i4

International Journal of Bioprinting AI for sustainable bioprinting

Table 1. Summary of artificial intelligence-driven studies in sustainable bioprinting
Study Artificial intelligence algo- Bioink material Bioprinting Sustainability indicators
rithms process
Chen et al. 14 Decision tree, random forest, Composite hydrogel, Direct ink writing Sustainable bioink use; material efficiency;
deep learning polymer experimental productivity
Lee et al. 81 Multiple regression Collagen, hyaluronic Extrusion Process efficiency; experimental productivity
acid, fibrin
Nadernezhad and Random forest Hyaluronic acid- Extrusion Sustainable bioink use; process efficiency;
Groll 80 based hydrogel experimental productivity
Bone et al. 85 Hierarchical machine learning Alginate-based Extrusion Sustainable bioink use; process efficiency;
experimental productivity
Chen et al. 86 Image recognition, random Hydrogel scaffold Extrusion Material efficiency; process efficiency;
forest experimental productivity
Fu et al. 87 Support vector machine Pluronic F127 Extrusion Process efficiency; experimental productivity
Xu et al. 88 Ensemble (ridge regression, Gelatin methacrylate Stereolithography Process efficiency; experimental productivity
k-nearest neighbor, random
forest, neural network)
Zhang et al. 89 Support vector regressor, Alginate-based Extrusion Sustainable bioink use; process efficiency;
multilayer perceptron experimental productivity
regressors
Wu and Xu 91 Random forest, least absolute Inkjet-compatible Inkjet Sustainable bioink use; process efficiency;
shrinkage and selection polymers experimental productivity
operator, support vector
regressor, extreme boosting
gradient
Chen et al. 92 Open multimodal particle- Soft materials Direct ink writing Sustainable bioink use; material efficiency
based displays, boundary
element method
Zhao et al. 94 Vision + adaptive feedback Gelatin-based Extrusion Material efficiency; process efficiency
Zboinska et al. 98 Artificial intelligence-assisted Cellulose nanofibril- Extrusion Sustainable bioink use; material efficiency
toolpath alginate

that encompass a broad spectrum of materials, processes, facilitating adaptability to new materials and dynamic
and conditions. 16,18,110,111 printing environments. 74,115 These initiatives, combined
with interdisciplinary collaboration and technological
To overcome these challenges, the creation of open-
source platforms and collaborative databases is helpful. innovation, will help overcome current limitations, paving
the way for AI-driven sustainable bioprinting to achieve its
Such platforms would centralize diverse datasets, full potential.
reflecting the variety of bioinks, printing parameters,
and application-specific conditions, thus enabling 4.2. Roadmap for artificial intelligence in
more representative and comprehensive AI model sustainable bioprinting
training. 112,113 Establishing standardized sustainability In addition to addressing current challenges, a strategic
metrics such as material efficiency, energy consumption, roadmap for future development suggests several
and waste minimization should also be an integral part directions that could advance sustainability alongside
of these centralized datasets. Quantifying these aspects technological progress. In material discovery and
would provide clear, measurable insights for sustainable development, AI-assisted platforms can leverage hybrid
bioprinting, enabling data-driven decisions to optimize modeling approaches that merge AI-driven predictive
AI algorithms, improve resource efficiency, and minimize models with physics-based simulations to enhance
environmental impact. Collaborative efforts could accuracy, reduce experimentation, and minimize
114
streamline data collection, eliminate redundancy, and foster resource usage. 77,116,117 Integrating life cycle assessment
a cohesive research community. Furthermore, integrating indicators such as embodied energy, carbon footprint,
advanced AI techniques like transfer learning and domain and toxicity profiles alongside mechanical and biological
adaptation can enhance AI model generalizability, performance metrics would allow the design of truly

Volume 11 Issue 4 (2025) 146 doi: 10.36922/IJB025170164