A Clinical Application of Ensemble ICA to the Quantification of Myocardial Blood Flow in Dynamic $$ H^{{15}}_{2} O $$ PET

Authors:
Byeong Il Lee;Jae Sung Lee;Dong Soo Lee;Won Jun Kang;Jong Jin Lee;Seungjin Choi
Affiliations:
Department of Nuclear Medicine, Seoul National University College of Medicine, Seoul, Korea;Department of Nuclear Medicine, Seoul National University College of Medicine, Seoul, Korea;Department of Nuclear Medicine, Seoul National University College of Medicine, Seoul, Korea;Department of Nuclear Medicine, Seoul National University College of Medicine, Seoul, Korea;Department of Nuclear Medicine, Seoul National University College of Medicine, Seoul, Korea;Department of Computer Science, Pohang University of Science and Technology, Pohang, Korea 790-784
Venue:
Journal of VLSI Signal Processing Systems
Year:
2007

Citing 2
Cited 0

Adaptive Blind Signal and Image Processing: Learning Algorithms and Applications

Adaptive Blind Signal and Image Processing: Learning Algorithms and Applications
Myocardial blood flow quantification in dynamic PET: an ensemble ICA approach

ICANN'05 Proceedings of the 15th international conference on Artificial neural networks: formal models and their applications - Volume Part II

Quantified Score

Hi-index	0.00

Visualization

Abstract

Ensemble independent component analysis (ICA) is a Bayesian multivariate data analysis method which allows various prior distributions for parameters and latent variables, leading to flexible data fitting. In this paper we apply ensemble ICA with a rectified Gaussian prior to dynamic $$ H^{{15}}_{2} O $$ positron emission tomography (PET) image data, emphasizing its clinical usefulness by showing that major cardiac components are successfully extracted in an unsupervised manner and myocardial blood flow can be estimated in 15 among 20 patients. Detailed experiments and results are illustrated.