Research on Constructing Surrogate Model of Coronary Stent Mechanical Property

Abstract:

Coronary stents are small tube-like structures expanded into stenotic arteries to restore blood flow perfusion to the downstream tissues. In this paper, based on the finite element analysis (FEA), both the response surface model (RSM) and radial basis function neural network (RBFNN) are used to construct the surrogate model for stent recoil performance. Precision between these two surrogate models are compared. The results show that the surrogate model with RBFNN is more reliable and efficient.

• Periodical: Key Engineering Materials (Volume 464)
• Pages: 564-567
• DOI: https://doi.org/10.4028/www.scientific.net/KEM.464.564
• Online since: January 2011
• Authors: Xiang Shen
• Keywords: RBFNN, RSM, Stent, Surrogate Model

References

[1] F. Migliavacca, L. Petrini, V. Montanari, I. Quagliana, F. Auricchio and G. Dubini: Medical Engineering & Physics Vol. 27 (2005), p.13.

DOI: 10.1016/j.medengphy.2004.08.012

[2] F. Etave, G. Finet, M. Boivin, J.C. Boyer, G. Rioufol and G. Thollet: Journal of Biomechanics Vol. 34 (2001), p.1065.

DOI: 10.1016/s0021-9290(01)00026-4

[3] S.M. Clarke, J.H. Griebsch and T.W. Simpson: Journal of Mechanical Design Vol. 127 (2005), p.1077.

[4] X.G. Song, J.H. Jung, H.J. Son, J.H. Park, K.H. Lee and Y.C. Park: Computers and Mathematics with Applications, in press.

[5] W. You, Y.X. Liu, B.Z. Bai and H.S. Fang: Journal of Iron and Steel Research, International Vol. 15 (2008), p.87.

[6] J. Moody and C. Darken: Neural Computation Vol. 1 (1989), p.281.