Dr Qing Li is a Professor in School of Aerospace, Mechanical and Mechatronic Engineering at the University of Sydney (USYD), Australia. Qing is currently the Director of “Centre of Advanced Materials Technology (CAMT)” at USYD. Dr Li received his PhD from the University of Sydney in 2000; and he was an Australian Research Council (ARC) Australian Postdoctoral Research Fellow (2001-03) as well as an ARC Future Fellow (2013-17). Prof Li is interested in computational optimization, data science, biomechanics, tissue engineering, biofabrication, and additive manufacturing. Professor Li is the President of International Society of Structural and Multidisciplinary Optimization (ISSMO) (2023-27) and have been named a Clarivate analytics “Highly Cited Researcher” since 2020. Professor Li published over 400 research articles in leading international journals with an h-index of 98 (Google Scholar).

Topic title: Machine Learning Based Design for Biomanufacturing of Reconstructive Tissue Prosthesis

Abstract:

Tissue prostheses, such as bone scaffolds, have emerged as a promising solution for treatment of critical size bone defects, offering compelling advantages over conventional strategies [1]. One of the key functionalities of bone scaffolds is their ability to promote long-term bone growth [2]. To enhance this functionality, we present a novel time-dependent optimization framework to customize scaffolds with maximum bone growth outcomes over a long-term period. To improve the design efficiency, we leverage machine learning (ML) techniques within the proposed time-dependent design framework (Figure 1(a-b)). Specifically, two neural networks are integrated into a dynamic bone growth model, and another neural network is coupled with a genetic algorithm for design optimization. The sophisticated topological design is implemented using an additive manufacturing technique. To demonstrate the effectiveness of the proposed approach, we exemplified a sheep mandible model with a major bone defect here (Figure 1(c)). The optimized scaffolds were first additively manufactured with Polyetherketone (PEK). Then, we compare three optimized designs, namely uniform, laterally graded, and vertically graded scaffolds, against an empirical design. The results demonstrate significant improvement in promoting long-term bone growth inside the optimized scaffolds. This study provides a systematic design approach combining ML and time-dependent optimization for better clinical outcomes.