2011
Leiva-Murillo, Jose M; Artés-Rodríguez, Antonio; Baca-García, Enrique
Visualization and Prediction of Disease Interactions with Continuous-Time Hidden Markov Models Proceedings Article
En: NIPS 2011 Workshop on Personalized Medicine., Sierra Nevada, 2011.
Resumen | Enlaces | BibTeX | Etiquetas: Computational, Information-Theoretic Learning with Statistics, Theory & Algorithms
@inproceedings{Leiva-Murillo2011,
title = {Visualization and Prediction of Disease Interactions with Continuous-Time Hidden Markov Models},
author = {Jose M Leiva-Murillo and Antonio Art\'{e}s-Rodr\'{i}guez and Enrique Baca-Garc\'{i}a},
url = {http://eprints.pascal-network.org/archive/00009110/},
year = {2011},
date = {2011-01-01},
booktitle = {NIPS 2011 Workshop on Personalized Medicine.},
address = {Sierra Nevada},
abstract = {This paper describes a method for discovering disease relationships and the evolution of diseases from medical records. The method makes use of continuous-time Markov chain models that overcome some drawbacks of the more widely used discrete-time chain models. The model addresses uncertainty in the diagnoses, possible diagnosis errors and the existence of multiple alternative diagnoses in the records. A set of experiments, performed on a dataset of psychiatric medical records, shows the capability of the model to visualize maps of comorbidity and causal interactions among diseases as well as to perform predictions of future evolution of diseases.},
keywords = {Computational, Information-Theoretic Learning with Statistics, Theory \& Algorithms},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper describes a method for discovering disease relationships and the evolution of diseases from medical records. The method makes use of continuous-time Markov chain models that overcome some drawbacks of the more widely used discrete-time chain models. The model addresses uncertainty in the diagnoses, possible diagnosis errors and the existence of multiple alternative diagnoses in the records. A set of experiments, performed on a dataset of psychiatric medical records, shows the capability of the model to visualize maps of comorbidity and causal interactions among diseases as well as to perform predictions of future evolution of diseases.
2009
Perez-Cruz, Fernando; Rodrigues, Miguel R D; Verdu, Sergio
Optimal Precoding for Multiple-Input Multiple-Output Gaussian Channels Proceedings Article
En: Seminar PIIRS, Princeton, 2009.
Resumen | Enlaces | BibTeX | Etiquetas: Theory & Algorithms
@inproceedings{Perez-Cruz2009a,
title = {Optimal Precoding for Multiple-Input Multiple-Output Gaussian Channels},
author = {Fernando Perez-Cruz and Miguel R D Rodrigues and Sergio Verdu},
url = {http://eprints.pascal-network.org/archive/00006754/},
year = {2009},
date = {2009-01-01},
booktitle = {Seminar PIIRS},
address = {Princeton},
abstract = {We investigate the linear precoding and power allocation policies that maximize the mutual information for general multiple-input multiple-output (MIMO) Gaussian channels with arbitrary input distributions, by capitalizing on the relationship between mutual information and minimum mean-square error. The optimal linear precoder satisfies a fixed-point equation as a function of the channel and the input constellation. For nonGaussian inputs, a nondiagonal precoding matrix in general increases the information transmission rate, even for parallel noninteracting channels. Whenever precoding is precluded, the optimal power allocation policy also satisfies a fixed-point equation; we put forth a generalization of the mercury/waterfilling algorithm, previously proposed for parallel noninterfering channels, in which the mercury level accounts not only for the nonGaussian input distributions, but also for the interference among inputs.},
keywords = {Theory \& Algorithms},
pubstate = {published},
tppubtype = {inproceedings}
}
We investigate the linear precoding and power allocation policies that maximize the mutual information for general multiple-input multiple-output (MIMO) Gaussian channels with arbitrary input distributions, by capitalizing on the relationship between mutual information and minimum mean-square error. The optimal linear precoder satisfies a fixed-point equation as a function of the channel and the input constellation. For nonGaussian inputs, a nondiagonal precoding matrix in general increases the information transmission rate, even for parallel noninteracting channels. Whenever precoding is precluded, the optimal power allocation policy also satisfies a fixed-point equation; we put forth a generalization of the mercury/waterfilling algorithm, previously proposed for parallel noninterfering channels, in which the mercury level accounts not only for the nonGaussian input distributions, but also for the interference among inputs.