Structured regression models for high-dimensional spatial spectroscopy data

DOI10.1214/17-EJS1301MaRDI QIDQ1676783zbMATH OpenWikidataFDO

Authors Arash A. Amini, Elizaveta Levina, Kerby A. Shedden

Publication date 10 November 2017

Published in Electronic Journal of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1311.0416, https://projecteuclid.org/euclid.ejs/1508918427

zbMATH Keywords

functional dataspatial dataspectroscopystructured regression

Mathematics Subject Classification ID

Nonparametric regression and quantile regression (62G08) Inference from stochastic processes and spectral analysis (62M15) Inference from stochastic processes and prediction (62M20) Inference from spatial processes (62M30) Applications of statistics to biology and medical sciences; meta analysis (62P10)

Abstract: Modeling and analysis of spectroscopy data is an active area of research with applications to chemistry and biology. This paper focuses on analyzing Raman spectra obtained from a bone fracture healing experiment, although the functional regression model for predicting a scalar response from high-dimensional tensors can be applied to any spectroscopy data. The regression model is built on a sparse functional representation of the spectra, and accommodates multiple spatial dimensions. We apply our models to the task of predicting bone-mineral-density (BMD), an important indicator of fracture healing, from Raman spectra, in both the in vivo and ex vivo settings of the bone fracture healing experiment. To illustrate the general applicability of the method, we also use it to predict lipoprotein concentrations from spectra obtained by nuclear magnetic resonance (NMR) spectroscopy.

Review text

Processing and analysis of microscopy data is a very active area of research with numerous applications in various scientific disciplines. The authors present a functional model for analysing high-dimensional spectra for the purpose of noise reduction and prediction. They consider the effectiveness of the approach in three settings: in vivo and ex vivo Raman data from a fracture healing experiment, and using NMR data in a lipoprotein concentration study.

Recommendations

• Multivariate functional response regression, with application to fluorescence spectroscopy in a cervical pre-cancer study
• Bayesian multiple response kernel regression model for high dimensional data and its practical applications in near infrared spectroscopy
• MICAI 2004: Advances in Artificial Intelligence
• Enhanced predictions of wood properties using hybrid models of PCR and PLS with high-dimensional NIR spectral data
• Editorial for the special issue on high-dimensional and functional data analysis

Cited in

(9)

• Multivariate functional response regression, with application to fluorescence spectroscopy in a cervical pre-cancer study
• A novel preprocessing algorithm for three-way HPLC data
• Bayesian multiple response kernel regression model for high dimensional data and its practical applications in near infrared spectroscopy
• Parsimonious Bayesian factor analysis for modelling latent structures in spectroscopy data
• New insights into evaluation of regression models through a decomposition of the prediction errors: application to near-infrared spectral data
• MICAI 2004: Advances in Artificial Intelligence
• Functional statistical techniques applied to vine leaf water content determination
• Enhanced predictions of wood properties using hybrid models of PCR and PLS with high-dimensional NIR spectral data
• Noise reduction for enhanced component identification in multi-dimensional biomolecular NMR studies

This page was built for publication: Structured regression models for high-dimensional spatial spectroscopy data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1676783)