Artificial Intelligence Driven Optimisation of Scaffold Design and Production in Tissue Engineering: A Critical Review
Ozieme, A.D.
Department of Biomedical Engineering, University of Ibadan, Nigeria and Department of Chemistry, Covenant University, Ota, Nigeria.
Adeleye, A.A. *
Department of Biomedical Engineering, University of Ibadan, Nigeria.
Ajide, O.O.
Department of Biomedical Engineering, University of Ibadan, Nigeria and Department of Mechanical Engineering, University of Ibadan, Nigeria.
*Author to whom correspondence should be addressed.
Abstract
The synergy of Artificial Intelligence (AI) technology and Tissue Engineering (TE) is creating a new frontier in regenerative medicine. Conventional design of scaffolds for enhancing cell attachment, growth and differentiation is well reported to be achieved through experimental methods. The AI is an emerging technique offering a powerful computational approach that can be deployed to analyse nonlinear and other complex interactions between different scaffold parameters. The foregoing makes it possible to predict and improve designs more efficiently. In this work, attempt is made to critically review the use of AI from traditional machine learning models to advanced deep learning systems for improving scaffold structure, composition and biological performance. Critical attention is made to the roles of supervised and unsupervised learning in predictive modelling, the use of generative algorithms for structural design and the application of evolutionary methods for optimising multiple design objectives. The synergy between AI, computational biomechanics, finite element modelling and Additive Manufacturing (AM) is examined, showing how innovations such as digital twin technology and automatic bioprinting have emerged. The review also explores how AI is deployed for material selection, surface property prediction and biodegradability evaluation in composite materials. Current challenges such as limited data, difficulty in model interpretation and ethical considerations are well discussed along with future directions that include physics-informed networks and combined AI systems. This review highlights progress in AI-driven scaffold optimization, emphasizing improved accuracy, consistency, and personalization in regenerative medicine.
Keywords: Scaffold optimization, AI systems, machine learning models, deep learning systems, regenerative medicine