Surrogate optimization of a lattice foot orthotic

Surrogate optimisation strategies for intraocular lens formula constant optimisation

“…The formula prediction error PE was defined as the difference between the formula predicted refraction and the SEQ from the postoperative follow‐up examination. For the surrogate formula constant optimisation method (Cruz et al., 2022; Forrester et al., 2008; Forrester & Keane, 2009; Gorissen et al., 2010; Johnson et al., 2019; Moeini et al., 2023; Wang et al., 2014; Zhang et al., 2022), the root mean squared PE (rmsPE) was used as an objective function to be minimised globally within the boundaries (Boyd & Vandenberghe, 2004; Mezura‐Montes & Coello Coello, 2011; Zhang et al., 2022). An internal function code was programmed to calculate the rmsPE for the input parameters (AL, CCT, ACD, LT and Rmean/Kmean/Rmean K ) a function of the formula constants A/pACD/SF/a0, a1, a2/C, H, R.…”

“…Surrogate model based optimisation is a very modern technique of active learning used in engineering and advanced statistics (Cruz et al., 2022; Forrester et al., 2008; Forrester & Keane, 2009; Gorissen et al., 2010; Johnson et al., 2019; Moeini et al., 2023; Wang et al., 2014; Zhang et al., 2022), and machine learning to find optimal design parameter combinations in terms of minimising (or maximising) an objective function globally over a restricted parameter space (Mezura‐Montes & Coello Coello, 2011). The parameter space has to be restricted using boundaries, and in most cases box constraints are used.…”

“…For a detailed explanation of the working principle of surrogate optimisation, we refer to the literature (Cruz et al., 2022; Forrester et al., 2008; Forrester & Keane, 2009; Gorissen et al., 2010; Johnson et al., 2019; Moeini et al., 2023; Wang et al., 2014; Zhang et al., 2022). In general, surrogate optimisation is always based on a surrogate model (Figure 1).…”

“…Note: AL refers to the axial length, ACD to the external phakic anterior chamber depth measured from the corneal epithelium to the front apex of the crystalline lens, LT to the central thickness of the crystalline lens, Rmean (mean of the flat and steep meridians) to the mean radius of curvature and Kmean (mean of the flat and steep meridians each converted to dioptric power using a keratometer index n K = 1.3375) for the corneal front surface, PIOL to the refractive power of the intraocular lens implant and SEQ to the spherical equivalent power achieved 5-12 weeks after cataract surgery. Forrester & Keane, 2009;Gorissen et al, 2010;Johnson et al, 2019;Moeini et al, 2023;Wang et al, 2014;Zhang et al, 2022), and machine learning to find optimal design parameter combinations in terms of minimising (or maximising) an objective function globally over a restricted parameter space (Mezura-Montes & Coello Coello, 2011). The parameter space has to be restricted using boundaries, and in most cases box constraints are used.…”