Skip to Main content Skip to Navigation
Conference papers

Reconstruction d'un spectre RMN 2D par maximum d'entropie

Abstract : Nuclear magnetic resonance (NMR) is a modern spectroscopy tool for the study of biological structures. This article deals with the reconstruction of T1 - T2 correlation spectrum from 2D NMR measures. The ill-posed character of this inverse problem and the large size of data constitute the main difficulties of the reconstruction. The Maximum Entropy method is used for estimation and two optimization algorithms are compared, the first is based on the classical Skilling and Bryan method and the second uses the nonlinear conjugate gradient method. In order to reduce computational cost, we exploit the specific structure of the observation model to avoid the approximations recently proposed in Vankataramanan et coll.. Furthermore, an original line search strategy suitable for the entropy function is used to ensure the decrease of the criterion at each iteration. These proposed algorithms are tested on a synthetic example and their applicability is illustrated through real NMR data processing.
Document type :
Conference papers
Complete list of metadata
Contributor : Migration Irstea Publications Connect in order to contact the contributor
Submitted on : Friday, May 15, 2020 - 3:41:42 PM
Last modification on : Tuesday, November 17, 2020 - 1:34:04 PM


  • HAL Id : hal-02591991, version 1
  • IRSTEA : PUB00026554



E. Chouzenoux, Saïd Moussaoui, J. Idier, François Mariette. Reconstruction d'un spectre RMN 2D par maximum d'entropie. 22ème colloque Gretsi 2009, Sep 2009, Dijon, France. pp.4. ⟨hal-02591991⟩



Record views