Correction for ambiguous solutions in factor analysis using a penalized least squares objective

@article{Sitek2002CorrectionFA,
  title={Correction for ambiguous solutions in factor analysis using a penalized least squares objective},
  author={Arkadiusz Sitek and Grant T. Gullberg and Ronald H. Huesman},
  journal={IEEE Transactions on Medical Imaging},
  year={2002},
  volume={21},
  pages={216-225}
}
Factor analysis is a powerful tool used for the analysis of dynamic studies. One of the major drawbacks of factor analysis of dynamic structures (FADS) is that the solution is not mathematically unique when only nonnegativity constraints are used to determine factors and factor coefficients. In this paper, a method to correct for ambiguous FADS solutions has been developed. A nonambiguous solution (to within certain scaling factors) is obtained by constructing and minimizing a new objective… 

Figures and Tables from this paper

The application of maximum likelihood factor analysis (MLFA) with uniqueness constraints on dynamic cardiac microPET data.
TLDR
A maximum likelihood framework for factor analysis (MLFA) of dynamic images was developed and an efficient iterative algorithm was developed to solve the maximum likelihood problem with non-negativity constraint embedded inside the iteration loop.
Generalized five-dimensional dynamic and spectral factor analysis.
TLDR
5D-GFA is a novel and promising approach in dynamic SPECT imaging that allows accurate and precise dynamic analysis while compensating for Compton scatter and cross-talk and can also be used in other modalities.
Replace-approximation method for ambiguous solutions in factor analysis of ultrasonic hepatic perfusion
TLDR
This paper proposes a new method of replace-approximation based on apex-seeking for ambiguous FADS solutions and shows that the technique outperformed two variants in comparison of region of interest measurements from phantom data.
Image-derived input function with factor analysis and a-priori information
TLDR
A factor-analysis-based IDIF for [11C](R)-rolipram brain PET studies that relies on a single blood sample and population data can be used for accurate quantification of Logan-VT values.
Robust Unmixing of Dynamic Sequences Using Regions of Interest
TLDR
An unmixing method, based on an objective function minimization and integrating classic regions of interest (ROIs) methods in a different and robust manner is proposed, which promotes consistent solutions regarding ROIs while relaxing the model outside ROIs.
R-clustering technique for initialization of factor analysis of dynamic PET images
TLDR
The developed method is applied to analyze 11C-PIB dynamic brain PET images acquired to study β-amyloid deposition in brains of Alzheimer patients and uses a concept of representative clustering, when only a small number of the analyzed time series are used for clustering.
Clustering-Initiated Factor Analysis Application for Tissue Classification in Dynamic Brain Positron Emission Tomography
TLDR
Clustering-initiated factor analysis can be an accurate and convenient tool for measuring specific binding tissue concentration and for analyzing tracer kinetics from dynamic images for a variety of PET tracers.
Reconstruction of 4-D Dynamic SPECT Images From Inconsistent Projections Using a Spline Initialized FADS Algorithm (SIFADS)
TLDR
An algorithm of extracting voxel-by-voxel time activity curves directly from inconsistent projections applied in dynamic cardiac SPECT to solve the inverse problem of dynamic SPECT with slow gantry rotation is proposed.
Quantitative Evaluation of Two-Factor Analysis Applied to Hepatic Perfusion Study Using Contrast-enhanced Ultrasound
TLDR
The results indicate that two-factor analysis has the potential to perform quantitative analysis of hepatic perfusion, which would be helpful to the differential diagnosis of FLLs.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 19 REFERENCES
Factor analysis of dynamic function studies using a priori physiological information.
TLDR
A procedure called IBFADS (information-based factor analysis in dynamic structures) is described which incorporates the IM (intersection method) technique previously developed into FADS, in order to reduce the error in the estimation of the correct model.
THE USE OF SET THEORY AND CLUSTER ANALYSIS TO INVESTIGATE THE CONSTRAINT PROBLEM IN FACTOR ANALYSIS IN DYNAMIC STRUCTURES (FADS)
In recent years factor analysis has played an increasingly important role in the analysis of nuclear medicine dynamic studies. Schmidlin (1979) demonstrated how the technique could be applied to a
Factor analysis with a priori knowledge--application in dynamic cardiac SPECT.
TLDR
Two factor analysis of dynamic structures (FADS) methods for the extraction of time-activity curves (TACs) from cardiac dynamic SPECT data sequences were investigated and indicate that the factor analysis techniques can be used for semi-automatic estimation of activity curves derived from cardiacynamic SPECT images.
Factor analysis of dynamic structures in dynamic SPECT imaging using maximum entropy
TLDR
It is shown that the FADS result is improved by using the Maximum Entropy Principle, and the inclusion of an entropy term is a useful method for obtaining more accurate extractions of TACs from dynamic SPECT data.
The effect of apex-finding errors on factor images obtained from factor analysis and oblique transformation.
  • A. Houston
  • Mathematics
    Physics in medicine and biology
  • 1984
TLDR
The effects of errors in apex-finding on physiological factors from a factor analysis of dynamic studies are investigated and it is shown that, apart from scaling, the factor image obtained from a given apex is independent of the position of that apex.
The Importance of Constraints in Factor Analysis of Dynamic Studies
A dynamic radionuclide study provides a description of the way the administrated tracer behaves in the several separate structures in the body. The usual aim of dynamic study analysis is to attempt
A method for recovering physiological components from dynamic radionuclide images using the maximum entropy principle: a numerical investigation
TLDR
The results demonstrate that the method can recover the component time-activity curves and the corresponding images of physiological components from a sequence of dynamic radionuclide images.
Imaging radiotracer model parameters in PET: a mixture analysis approach
TLDR
Two methodologies for fitting radiotracer models on a pixel-wise basis to PET data are considered and the results obtained by mixture analysis are found to have substantially improved mean square error performance characteristics.
Target apex-seeking in factor analysis of medical image sequences.
TLDR
A new target apex-seeking method which integrates physical or physiological knowledge in this stage is described, which concerns some of the fundamental functions and reacts on the determination of all the factors.
The use of principal components in the quantitative analysis of gamma camera dynamic studies.
  • D. Barber
  • Biology
    Physics in medicine and biology
  • 1980
TLDR
Using a principal components analysis of the elemental curves representing the change of activity with time in each pixel, it is possible to extract the fundamental functional changes of activity which underly the observed variation of activity.
...
1
2
...