Deformable mirror model for open-loop adaptive optics using multivariate adaptive regression splines

Guzmán, Dani; Juez, Francisco Javier de Cos; Lasheras, Fernando Sánchez; Myers, R; Young, Laura K.

doi:10.1364/oe.18.006492

Cited by 36 publications

(21 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The main purpose of the MARS model is to predict the values of continuous dependent variables from a set of independent explanatory variables. MARS does not require any a priori assumptions about the underlying functional relationship between dependent and independent variables [42]. Models were trained using as input variables the reduced subsets obtained in step 1 for each of the output variables.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Predictors of Posttreatment Drinking Outcomes in Patients with Alcohol Dependence

Flórez¹,

Sáiz²,

García‐Portilla³

et al. 2014

Eur Addict Res

View full text Add to dashboard Cite

Aim: This cohort study examined how predictors of alcohol dependence treatment outcomes work together over time by comparing pretreatment and posttreatment predictors. Methods: A sample of 274 alcohol-dependent patients was recruited and assessed at baseline, 6 months after treatment initiation (end of the active intervention phase), and 18 months after treatment initiation (end of the 12-month research follow-up phase). At each assessment point, the participants completed a battery of standardized tests [European Addiction Severity Index (EuropASI), Obsessive Compulsive Drinking Scale (OCDS), Alcohol Timeline Followback (TLFB), Fagerström, and International Personality Disorder Examination (IPDE)] that measured symptom severity and consequences; biological markers of alcohol consumption were also tested at each assessment point. A sequential strategy with univariate and multivariate analyses was used to identify how pretreatment and posttreatment predictors influence outcomes up to 1 year after treatment. Results: Pretreatment variables had less predictive power than posttreatment ones. OCDS scores and biological markers of alcohol consumption were the most significant variables for the prediction of posttreatment outcomes. Prior pharmacotherapy treatment and relapse prevention interventions were also associated with posttreatment outcomes. Conclusions: The findings highlight the positive impact of pharmacotherapy during the first 6 months after treatment initiation and of relapse prevention during the first year after treatment and how posttreatment predictors are more important than pretreatment predictors.

show abstract

Section: Methodsmentioning

confidence: 99%

“…The MARS model of a dependent variable y → with M basis functions (terms) can be written as follows [42]:…”

Section: Methodsmentioning

confidence: 99%

Predictors of Posttreatment Drinking Outcomes in Patients with Alcohol Dependence

Flórez¹,

Sáiz²,

García‐Portilla³

et al. 2014

Eur Addict Res

View full text Add to dashboard Cite

show abstract

“…For wide-field AO systems the error introduced by the atmospheric turbulence can be measured in different directions and computer tomography techniques can be used to calculate the aberration in the direction of the object of interest. This allows for the real-time compensation of the astronomical image with the deformable mirrors [3]. …”

Section: Introductionmentioning

confidence: 99%

“…CARMEN is a MOAO reconstructor created at the University of Oviedo. It was initially developed using regression techniques such as multivariate adaptive regression splines (MARS) [12,13], with promising results [3,14]. However, the use of machine learning techniques such as Supporting Vector Machines (SVM) [15,16,17,18] or Artificial Neural Networks (ANN) have proven very successful in different fields [19,20], which led us to create a solution based on an ANN [21].…”

Section: Introductionmentioning

confidence: 99%

Comparative Study of Neural Network Frameworks for the Next Generation of Adaptive Optics Systems

González-Gutiérrez

Santos

Martínez‐Zarzuela

et al. 2017

Sensors

View full text Add to dashboard Cite

Many of the next generation of adaptive optics systems on large and extremely large telescopes require tomographic techniques in order to correct for atmospheric turbulence over a large field of view. Multi-object adaptive optics is one such technique. In this paper, different implementations of a tomographic reconstructor based on a machine learning architecture named “CARMEN” are presented. Basic concepts of adaptive optics are introduced first, with a short explanation of three different control systems used on real telescopes and the sensors utilised. The operation of the reconstructor, along with the three neural network frameworks used, and the developed CUDA code are detailed. Changes to the size of the reconstructor influence the training and execution time of the neural network. The native CUDA code turns out to be the best choice for all the systems, although some of the other frameworks offer good performance under certain circumstances.

show abstract

“…To improve the fidelity, the main residual source, high spatial frequency deformation of a DM from actuator printthrough [12,13], should be cut off. In this paper, we introduce a DM concept that has a potential of generating certain high fidelity optical modes with less actuators than conventional DMs.…”

Section: Introductionmentioning

confidence: 99%

Concept and modeling analysis of a high fidelity multimode deformable mirror

Zhou

Wang

et al. 2015

Appl. Opt.

View full text Add to dashboard Cite

Conventional deformable mirrors (DM) cannot meet the requirement of aberration controlling for advanced lithography tools. This paper illustrates an approach using the property that deformation of a thin plate is similar to optical modes to realize a high fidelity multimode deformable mirror whose deformation has characteristics of optical aberration modes. The way to arrange actuators is also examined. In this paper, a 36-actuator deformable mirror is taken as an example to generate low-order Zernike modes. The result shows that this DM generates the fourth fringe Zernike mode (Z4) defocus, and primary aberration Z5-Z8 with an error less than 0.5%, generates the fifth-order aberration Z10-Z14, and generates the seventh-order aberration Z17-Z20 with an error less than 1.1%. The high fidelity replication of the Zernike mode indicates that the DM satisfies the demand of controlling aberrations corresponding to the first 20 Zernike modes in an advanced lithography tool.

show abstract

Deformable mirror model for open-loop adaptive optics using multivariate adaptive regression splines

Cited by 36 publications

References 18 publications

Predictors of Posttreatment Drinking Outcomes in Patients with Alcohol Dependence

Predictors of Posttreatment Drinking Outcomes in Patients with Alcohol Dependence

Comparative Study of Neural Network Frameworks for the Next Generation of Adaptive Optics Systems

Concept and modeling analysis of a high fidelity multimode deformable mirror

Contact Info

Product

Resources

About