International Journal of
Advanced Multidisciplinary Application

Volume 3, Issue 7

• Author(s)Meera S. Chandrasekaran
• AffiliationDepartment of Biomedical Engineering, Sandeep Foundation, Nashik
• Page No.71-76
• Volume, Issue & YearVolume 3, Issue 7, July 2026
• Published On2026/07/07
• JournalInternational Journal of Advanced Multidisciplinary Application (IJAMA)
• ISSN No.3048-9350

Abstract

Scaffold-based bone tissue engineering requires simultaneous satisfaction of competing mechanical and biological design objectives: sufficient compressive modulus to bear physiological load, and sufficient pore interconnectivity and permeability to support vascularisation, nutrient transport, and osteoblast infiltration. Fixed-geometry unit-cell scaffolds (diamond, gyroid) typically optimise one objective at the expense of the other, since increasing strut thickness to raise modulus simultaneously reduces pore size and permeability. This study develops a machine learning (ML) surrogate-assisted multi-objective optimisation framework that jointly predicts compressive modulus and permeability from scaffold design variables, enabling generative design of spatially graded scaffold architectures that resolve this trade-off.
A gradient-boosted regression surrogate model was trained on 3,600 finite element analysis (FEA) compression simulations and 1,800 computational fluid dynamics (CFD) permeability simulations spanning pore diameter, strut thickness, porosity gradient, and unit-cell type. The validated surrogate was embedded within a genetic algorithm-driven generative design loop to identify a Pareto-optimal graded scaffold architecture, which was subsequently fabricated in medical-grade polycaprolactone (PCL) via fused deposition modelling and benchmarked against fixed diamond and gyroid scaffolds through mechanical compression testing, micro-CT pore architecture analysis, and 6-week MC3T3-E1 osteoblast culture assessing proliferation, alkaline phosphatase activity, and calcium mineralisation.
The ML-optimised graded scaffold achieved a compressive modulus of 312 MPa at 70% mean porosity, within the native cancellous bone range, while maintaining a CFD-predicted permeability of 7.8×10⁻⁹ m² - 38% higher than the gyroid scaffold and 64% higher than the diamond scaffold at matched porosity. The ML-optimised scaffold supported significantly greater osteoblast proliferation (9.2-fold increase by day 21 versus 5.9-fold for diamond) and calcium deposition (1.18 mg/scaffold at 6 weeks versus 0.74 mg for diamond), a 59.5% improvement attributed to the graded pore architecture's closer match to the ML-identified optimal pore diameter window of 400-600 µm. The permeability surrogate achieved R² = 0.95 against held-out CFD validation data, with pore diameter and strut thickness identified as the dominant design drivers via SHAP analysis.

Keywords: bone tissue engineering, scaffold design, machine learning, additive manufacturing, permeability, compressive modulus, osteogenic differentiation, polycaprolactone, generative design, mechanobiology

References

1. [1] Bose, S., Vahabzadeh, S., & Bandyopadhyay, A. (2013). Bone tissue engineering using 3D printing. Materials Today, 16(12), 496-504.
2. [2] Cheng, A., Humayun, A., Cohen, D. J., et al. (2014). Additively manufactured 3D porous Ti-6Al-4V constructs mimic trabecular bone structure and regulate osteoblast proliferation, differentiation and local factor production. Biofabrication, 6(4), 045007.
3. [3] Egan, P. F., Gonella, V. C., Engensperger, M., et al. (2017). Computationally designed lattices with tuned properties for tissue engineering using 3D printing. PLOS ONE, 12(8), e0182902.
4. [4] Hollister, S. J. (2005). Porous scaffold design for tissue engineering. Nature Materials, 4(7), 518-524.
5. [5] Karageorgiou, V., & Kaplan, D. (2005). Porosity of 3D biomaterial scaffolds and osteogenesis. Biomaterials, 26(27), 5474-5491.
6. [6] Lundberg, S. M., & Lee, S. I. (2017). A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems, 30, 4765-4774.
7. [7] Melchels, F. P. W., Tonnarelli, B., Olivares, A. L., et al. (2011). The influence of the scaffold design on the distribution of adhering cells after perfusion cell seeding. Biomaterials, 32(11), 2878-2884.
8. [8] Murphy, C. M., Haugh, M. G., & O'Brien, F. J. (2010). The effect of mean pore size on cell attachment, proliferation and migration in collagen-glycosaminoglycan scaffolds. Biomaterials, 31(3), 461-466.
9. [9] Naghieh, S., & Chen, X. (2021). Engineering parameters and machine learning in scaffold design and fabrication. Materials Today Bio, 12, 100143.
10. [10] Olszta, M. J., Cheng, X., Jee, S. S., et al. (2007). Bone structure and formation: A new perspective. Materials Science and Engineering R, 58(3-5), 77-116.
11. [11] Sanz-Herrera, J. A., Garcia-Aznar, J. M., & Doblare, M. (2009). A mathematical approach to bone tissue engineering. Philosophical Transactions of the Royal Society A, 367(1895), 2055-2078.
12. [12] Wang, X., Xu, S., Zhou, S., et al. (2016). Topological design and additive manufacturing of porous metals for bone scaffolds and orthopaedic implants: A review. Biomaterials, 83, 127-141.
13. [13] Zadpoor, A. A. (2019). Mechanical performance of additively manufactured meta-biomaterials. Acta Biomaterialia, 85, 41-59.

Explore Our Related Journals

Looking for the right journal for your next manuscript? Explore our international peer-reviewed journals covering engineering, management, computer science, artificial intelligence and multidisciplinary research.