Anne Sauve - Academia.edu (original) (raw)
Related Authors
Università degli Studi del Piemonte Orientale "Amedeo Avogadro"
Uploads
Papers by Anne Sauve
We propose several preprocessing steps to be used before biomarker clustering or classifying for ... more We propose several preprocessing steps to be used before biomarker clustering or classifying for high-throughput Mass Spectrometry (MS) data. These preprocessing steps for the mass spectra are multiple alignment of technical replicates, baseline correction and normalization along the mass/charge axis. While the benefits from baseline correction and alignment seem obvious we studied more carefully the benefit from normalizing using some human prostate cancer SELDI TOF MS data (obtained from the Virginia Prostate Center Tissue and body Fluid Bank and approved by the Eastern Virginia Medical School). We show on these data that our global normalization by scaling helps in distinguishing between different cancer groups as well as between cancer and non-cancer groups. We used the Between to Within sum of squares ratio introduced by Fisher as well as visual inspection to illustrate the improvement brought by the normalization.
2009 IEEE Nuclear Science Symposium Conference Record (NSS/MIC), 2009
The goal of this research is to quantitatively compare fatty acid metabolism in the hearts of Wis... more The goal of this research is to quantitatively compare fatty acid metabolism in the hearts of Wistar-Kyoto (WKY) normal rats and spontaneously hypertensive rats (SHR) as a function of age, and thereby track physiological changes associated with the onset and progression of heart failure in the SHR model. The fatty acid analog, 123 I-labeled BMIPP, was used in longitudinal dynamic pinhole SPECT imaging studies performed on two WKY normal rats and two SHRs every seven months for 21 months. In previous work, we addressed issues associated with reconstructing dynamic data acquired with a slowly rotating camera. In this work, we address quantitative effects of limited spatial resolution that result in underestimation of metabolic rate from compartmental models. In particular, blurring of activity between the left ventricular blood pool and surrounding myocardial tissue decreases contrast between blood input and tissue uptake time-activity curves (TACs). Standard compartmental modeling straightforwardly accounts for spillover of blood activity into tissue volumes. However, accounting for spillover of tissue activity into blood volumes is more problematic. Because of tissue spillover, there is no reconstructed voxel that contains a pure blood TAC. Thus, we developed a method to jointly estimate the pure blood input along with compartmental model parameters from B-spline TACs reconstructed directly from dynamic SPECT projection data for 11 studies. Tissue spillover correction improved the contrast between blood input and myocardial uptake curves for all studies and visually improved the fit of the compartmental model for some studies. Estimates of metabolic rate of 123 I-labeled BMIPP increased by an average of 72±45% across all 11 studies, compared to estimates obtained without spillover correction. Thus, the tissue spillover correction method resulted in improved quantitative dynamic imaging of fatty acid metabolism in the rat heart, even with slow camera rotation.
Computation of the Cramer-Rao bound (CRB) on estimator variance requires the inverse or the pseud... more Computation of the Cramer-Rao bound (CRB) on estimator variance requires the inverse or the pseudo-inverse Fisher information matrix (FIM). Direct matrix inversion can be computationally intractable when the number of unknown parameters is large. In this correspondence, we compare several iterative methods for approximating the CRB using matrix splitting and preconditioned conjugate gradient algorithms. For a large class of inverse problems, we show that nonmonotone Gauss-Seidel and preconditioned conjugate gradient algorithms require significantly fewer flops for convergence than monotone "bound preserving" algorithms.
We propose several preprocessing steps to be used before biomarker clustering or classifying for ... more We propose several preprocessing steps to be used before biomarker clustering or classifying for high-throughput Mass Spectrometry (MS) data. These preprocessing steps for the mass spectra are multiple alignment of technical replicates, baseline correction and normalization along the mass/charge axis. While the benefits from baseline correction and alignment seem obvious we studied more carefully the benefit from normalizing using some human prostate cancer SELDI TOF MS data (obtained from the Virginia Prostate Center Tissue and body Fluid Bank and approved by the Eastern Virginia Medical School). We show on these data that our global normalization by scaling helps in distinguishing between different cancer groups as well as between cancer and non-cancer groups. We used the Between to Within sum of squares ratio introduced by Fisher as well as visual inspection to illustrate the improvement brought by the normalization.
2009 IEEE Nuclear Science Symposium Conference Record (NSS/MIC), 2009
The goal of this research is to quantitatively compare fatty acid metabolism in the hearts of Wis... more The goal of this research is to quantitatively compare fatty acid metabolism in the hearts of Wistar-Kyoto (WKY) normal rats and spontaneously hypertensive rats (SHR) as a function of age, and thereby track physiological changes associated with the onset and progression of heart failure in the SHR model. The fatty acid analog, 123 I-labeled BMIPP, was used in longitudinal dynamic pinhole SPECT imaging studies performed on two WKY normal rats and two SHRs every seven months for 21 months. In previous work, we addressed issues associated with reconstructing dynamic data acquired with a slowly rotating camera. In this work, we address quantitative effects of limited spatial resolution that result in underestimation of metabolic rate from compartmental models. In particular, blurring of activity between the left ventricular blood pool and surrounding myocardial tissue decreases contrast between blood input and tissue uptake time-activity curves (TACs). Standard compartmental modeling straightforwardly accounts for spillover of blood activity into tissue volumes. However, accounting for spillover of tissue activity into blood volumes is more problematic. Because of tissue spillover, there is no reconstructed voxel that contains a pure blood TAC. Thus, we developed a method to jointly estimate the pure blood input along with compartmental model parameters from B-spline TACs reconstructed directly from dynamic SPECT projection data for 11 studies. Tissue spillover correction improved the contrast between blood input and myocardial uptake curves for all studies and visually improved the fit of the compartmental model for some studies. Estimates of metabolic rate of 123 I-labeled BMIPP increased by an average of 72±45% across all 11 studies, compared to estimates obtained without spillover correction. Thus, the tissue spillover correction method resulted in improved quantitative dynamic imaging of fatty acid metabolism in the rat heart, even with slow camera rotation.
Computation of the Cramer-Rao bound (CRB) on estimator variance requires the inverse or the pseud... more Computation of the Cramer-Rao bound (CRB) on estimator variance requires the inverse or the pseudo-inverse Fisher information matrix (FIM). Direct matrix inversion can be computationally intractable when the number of unknown parameters is large. In this correspondence, we compare several iterative methods for approximating the CRB using matrix splitting and preconditioned conjugate gradient algorithms. For a large class of inverse problems, we show that nonmonotone Gauss-Seidel and preconditioned conjugate gradient algorithms require significantly fewer flops for convergence than monotone "bound preserving" algorithms.