Pasar al contenido principal

Functional data analysis

IJCI-2017-34038

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal de Promoción del Talento y su Empleabilidad en I+D+i
Subprograma Subprograma Estatal de Incorporación
Convocatoria Juan de la Cierva Incorporación
Año convocatoria 2017
Plan Estatal 2017-2020 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSIDAD CARLOS III DE MADRID

Resultados relacionados

Resultados totales (Incluyendo duplicados): 5
Encontrada(s) 1 página(s)

e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publicación Artículo preliminar (WorkingPaper)

Quantile regression : a penalization approach

  • Méndez Civieta, Álvaro
  • Aguilera Morillo, María del Carmen
  • Lillo Rodríguez, Rosa Elvira
Sparse group LASSO (SGL) is a penalization technique used in regression problems where the covariates have a natural grouped structure and provides solutions that are both between and within group sparse. In this paper the SGL is introduced to the quantile regression (QR) framework, and a more flexible version, the adaptive sparse group LASSO (ASGL), is proposed. This proposal adds weights to the penalization improving prediction accuracy. Usually, adaptive weights are taken as a function of the original non-penalized solution model. This approach is only feasible in the n > p framework. In this work, a solution that allows using adaptive weights in high-dimensional scenarios is proposed. The benefits of this proposal are studied both in synthetic and real datasets., In this research we have made use of Uranus, a supercomputer cluster located at University Carlos III of Madrid and funded jointly by EU-FEDER funds and by the Spanish Government via the National Projects No. UNC313-4E- 2361, No. ENE2009-12213- C03-03, No. ENE2012-33219 and No. ENE2015- 68265-P. This research was partially supported by research grants and Project ECO2015-66593-P from Ministerio de Economía, Industria y Competitividad, Project MTM2017-88708-P from Ministerio de Economía y Competi- tividad, FEDER funds and Project IJCI-2017-34038 from Agencia Estatal de Investigación, Ministerio de Ciencia, Innovación y Universidades.
e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publicación Artículo científico (JournalArticle)

Advanced Statistical Techniques for Noninvasive Hyperglycemic States Detection in Mice Using Millimeter-Wave Spectroscopy

  • Moreno Oyervides, Aldo Luis
  • Aguilera Morillo, María del Carmen
  • Larcher Laguzzi, Fernando
  • Krozer, Viktor
  • Acedo Gallardo, Pablo
In this article, we discuss the use of advanced statistical techniques (functional data analysis) in millimeter-wave (mm-wave) spectroscopy for biomedical applications. We employ a W-band transmit-receive unit with a reference channel to acquire spectral data. The choice of the W-band is based on a tradeoff between penetration through the skin providing an upper bound for the frequencies and spectral content across the band. The data obtained are processed using functional principal component logit regression (FPCLoR), which enables to obtain a predictive model for sustained hyperglycemia, typically associated with diabetes. The predictions are based on the transmission data from noninvasive mm-wave spectrometer at W-band. We show that there exists a frequency range most suitable for identification, classification, and prediction of sustained hyperglycemia when evaluating the functional parameter of the functional logit model (β). This allows for the optimization of the spectroscopic instrument in the aim to obtain a compact and potential low-cost noninvasive instrument for hyperglycemia assessment. Furthermore, we also demonstrate that the statistical tools alleviate the problem of calibration, which is a serious obstacle in similar measurements at terahertz and IR frequencies.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Early, Non-Invasive Sensing of Sustained Hyperglycemia in Mice Using Millimeter-Wave Spectroscopy

  • Moreno-Oyervides, Aldo
  • Martín-Mateos, Pedro
  • Ulisse, Giacomo
  • Arriba, María C.
  • Durban, María
  • Del Rio, Marcela
  • Larcher, Fernando
  • Krozer, Viktor
  • Acedo, Pablo
  • Aguilera-Morillo, M. Carmen|||0000-0003-1027-9773
[EN] Diabetes is a very complex condition affecting millions of people around the world. Its occurrence, always accompanied by sustained hyperglycemia, leads to many medical complications that can be greatly mitigated when the disease is treated in its earliest stage. In this paper, a novel sensing approach for the early non-invasive detection and monitoring of sustained hyperglycemia is presented. The sensing principle is based on millimeter-wave transmission spectroscopy through the skin and subsequent statistical analysis of the amplitude data. A classifier based on functional principal components for sustained hyperglycemia prediction was validated on a sample of twelve mice, correctly classifying the condition in diabetic mice. Using the same classifier, sixteen mice with drug-induced diabetes were studied for two weeks. The proposed sensing approach was capable of assessing the glycemic states at different stages of induced diabetes, providing a clear transition from normoglycemia to hyperglycemia typically associated with diabetes. This is believed to be the first presentation of such evolution studies using non-invasive sensing. The results obtained indicate that gradual glycemic changes associated with diabetes can be accurately detected by non-invasively sensing the metabolism using a millimeter-wave spectral sensor, with an observed temporal resolution of around four days. This unprecedented detection speed and its non-invasive character could open new opportunities for the continuous control and monitoring of diabetics and the evaluation of response to treatments (including new therapies), enabling a much more appropriate control of the condition., This research was supported in part by the Spanish Ministry of Economy and Competitiveness through the project TEC2017-86271-R and the Instituto de Salud Carlos III (Spain) through the grant DTS17/00135. Thanks to the Consejo Nacional de Ciencia y Tecnología de México (CONACYT) for financially supporting the doctoral education of Aldo M-O (PhD. Grant). Viktor Krozer is thankful for partial financial support in the frame of the Chairs of Excellence program of the Universidad Carlos III, Madrid, Spain. This research was partially supported by Project MTM2017-88708-P from Ministerio de Economía y Competitividad, FEDER funds, Project IJCI-2017-34038 from Agencia Estatal de Investigación, Ministerio de Ciencia, Innovación y Universidades and the Spanish Ministry of Economy and Competitiveness MTM2014-52184-P.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Linear-Phase-Type probability modelling of functional PCA with applications to resistive memories

  • Ruiz-Castro, Juan E.
  • Acal, Christian
  • Aguilera, Ana M.
  • Roldán, Juan B.
  • Aguilera-Morillo, M. Carmen|||0000-0003-1027-9773
[EN] Functional principal component analysis (FPCA) based on Karhunen-Loeve (K-L) expansion allows to describe the stochastic evolution of the main characteristics associated to multiple systems and devices. Identifying the probability distribution of the principal component scores is fundamental to characterize the whole process. The aim of this work is to consider a family of statistical distributions that could be accurately adjusted to a previous transformation. Then, a new class of distributions, the linear-phase-type, is introduced to model the principal components. This class is studied in detail in order to prove, through the K-L expansion, that certain linear transformations of the process at each time point are phase-type distributed. This way, the one-dimensional distributions of the process are in the same linear-phase-type class. Finally, an application to model the reset process associated with resistive memories is developed and explained. (C) 2020 Published by Elsevier B.V. on behalf of International Association for Mathematics and Computers in Simulation (IMACS)., We would like to thank F. Campabadal and M.B. Gonzalez from the IMB-CNM (CSIC) in Barcelona for fabricating and providing the experimental measurements of the devices employed here. We acknowledge the support of the Spanish Ministry of Science, Innovation and Universities under projects TEC2017-84321-C4-3-R, MTM201788708-P, IJCI-2017-34038 (also supported by the FEDER, Spain program) and the PhD grant, Spain (FPU18/01779) awarded to Christian Acal. This work has made use of the Spanish ICTS Network MICRONANOFABS
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Adaptive sparse group LASSO in quantile regression

  • Mendez-Civieta, Alvaro
  • Lillo, Rosa E.
  • Aguilera-Morillo, M. Carmen|||0000-0003-1027-9773
[EN] This paper studies the introduction of sparse group LASSO (SGL) to the quantile regression framework. Additionally, a more flexible version, an adaptive SGL is proposed based on the adaptive idea, this is, the usage of adaptive weights in the penalization. Adaptive estimators are usually focused on the study of the oracle property under asymptotic and double asymptotic frameworks. A key step on the demonstration of this property is to consider adaptive weights based on a initial root n-consistent estimator. In practice this implies the usage of a non penalized estimator that limits the adaptive solutions to low dimensional scenarios. In this work, several solutions, based on dimension reduction techniques PCA and PLS, are studied for the calculation of these weights in high dimensional frameworks. The benefits of this proposal are studied both in synthetic and real datasets., We appreciate the work of the referees that has contributed to substantially improve the scientific contributions of this work. In this research we have made use of Uranus, a supercomputer cluster located at University Carlos III of Madrid and funded jointly by EU-FEDER funds and by the Spanish Government via the National Projects No. UNC313-4E-2361, No. ENE2009-12213- C03-03, No. ENE2012-33219 and No. ENE2015-68265-P. This research was partially supported by research grants and Project ECO2015-66593-P from Ministerio de Economia, Industria y Competitividad, Project MTM2017-88708-P from Ministerio de Economia y Competitividad, FEDER funds and Project IJCI-2017-34038 from Agencia Estatal de Investigacion, Ministerio de Ciencia, Innovacion y Universidades.