Person:

Vidakovic, Brani

Permanent Link

https://hdl.handle.net/1853/72209

Associated Organization(s)

Organizational Unit

H. Milton Stewart School of Industrial and Systems Engineering

Organizational Unit

Wallace H. Coulter Department of Biomedical Engineering

Full item page

Publication Search Results

Now showing 1 - 10 of 35

Multiscale approach to functional data analysis with applications in monitoring

(Georgia Institute of Technology, 2009-06-18) Vidakovic, Brani
Collaborative research: Analysis of functional and high-dimensional data with applications

(Georgia Institute of Technology, 2008-07-01) Vidakovic, Brani
Wavelet-based 3-D Multifractal Spectrum with Applications in Breast MRI Images

(Georgia Institute of Technology, 2008) Derado, Gordana ; Lee, Kichun ; Nicolis, Orietta ; Bowman, F. DuBois ; Newell, Mary ; Ruggeri, Fabrizio ; Vidakovic, Brani

Breast cancer is the second leading cause of death in women in the United States. Breast Magnetic Resonance Imaging (BMRI) is an emerging tool in breast cancer diagnostics and research, and it is becoming routine in clinical practice. Recently, the American Cancer Society (ACS) recommended that women at very high risk of developing breast cancer have annual BMRI exams, in addition to annual mammograms, to increase the likelihood of early detection. (Saslow et al. [20]). Many medical images demonstrate a certain degree of self-similarity over a range of scales. The multifractal spectrum (MFS) summarizes possibly variable degrees of scaling in one dimensional signals and has been widely used in fractal analysis. In this work, we develop a generalization of MFS to three dimensions and use dynamics of the scaling as discriminatory descriptors for the classification of BMRI images to benign and malignant. Methodology we propose was tested using breast MRI images for four anonymous subjects (two cancer, and two cancer-free cases). The dataset consists of BMRI scans obtained on a 1.5T GE Signa MR (with VIBRANT) scanner at Emory University. We demonstrate that meaningful descriptors show potential for classifying inference.
Testing Equality of Stationary Autocovariances

(Georgia Institute of Technology, 2007-08-09) Lund, Robert ; Bassily, Hany ; Vidakovic, Brani

This paper studies tests for assessing whether two stationary and independent time series have the same dynamics, specifically, whether the autocovariances of both series coincide at all lags. Several frequency domain statistics previously proposed for this purpose are reviewed. A time domain statistic is then developed and investigated. The performance of these statistics are compared. As the previous literature on this topic resides almost exclusively within the spectral domain, it is perhaps surprising that the time domain test outperforms the frequency domain tests. Multivariate versions of the results are then investigated. The methods are applied in the analysis of temperatures and precipitations from two towns in the state of Georgia. Our interest here is driven by the need to identify a good climatological reference series for a given station. Efforts are made to keep the exposition rudimentary and expository.
Wavelet-based 2D Multifractal Spectrum with Applications in Analysis of Digital Mammography Images

(Georgia Institute of Technology, 2007-07-26) Ramírez, Pepa ; Vidakovic, Brani

Breast cancer is the second leading cause of death in women in the United States and at present, mammography is the only proven method that can detect minimal breast cancer. On the other hand, many medical images demonstrate a certain degree of self-similarity over a range of scales. The Multifractal spectrum (MFS) summarizes possibly variable degrees of scaling in one dimensional signals and has been widely used in fractal analysis. In this work, we develop a generalization to two dimensions of MFS and use dynamics of the scaling as discriminatory descriptors to do classification of mammographic images to benign and malignant. Methodology we propose was tested using images from the University of South Florida Digital Database for Screening Mammography (DDSM) (Heat et al. [8]).
On Bayesian estimation of multinomial probabilities under incomplete experimental information

(Georgia Institute of Technology, 2007-05-28) Ramírez, Pepa ; Vidakovic, Brani

In this note, we discuss Bayesian estimation of multinomial probabilities associated with a finite alphabet A, under incomplete experimental information. Two types of prior information are considered: (i) number of letters needed to see a particular pattern for the first time, and (ii) the fact that for two fixed words one appeared before the other.
Self-similarity in NMR spectra: an application in assessing the level of cysteine

(Georgia Institute of Technology, 2007-01-15) Jung, Yoon Young ; Park, Youngja ; Jones, Dean P. ; Ziegler, Thomas R. ; Vidakovic, Brani

High resolution of NMR spectroscopic data of biosamples are a rich source of information on the metabolic response to physiological variation or pathological events. There are many advantages of NMR techniques such as the sample preparation is fast, simple and non-invasive. Statistical analysis of NMR spectra usually focuses on differential expression of large resonance intensity corresponding to abundant metabolites and involves several data preprocessing steps. In this paper we estimate functional components of spectra and test their significance using multiscale techniques. We also explore scaling in NMR spectra and use the systematic variability of scaling descriptors to predict the level of cysteine, an important precursor of glutathione, a control antioxidant in human body. This is motivated by high cost (in time and resources) of traditional methods for assessing cysteine level by high performance liquid chromatograph (HPLC).
The Effectiveness of Sealants in Managing Carious Lesions

(Georgia Institute of Technology, 2007) Griffin, S. O. ; Oong, E. ; Kohn, W. ; Vidakovic, Brani ; Gooch, B. F. ; Bader, J. ; Clarkson, J. ; Fontana, M. R. ; Meyer, D. M. ; Rozier, R. G. ; Weintraub, J. A. ; Zero, D. T.

A barrier to providing sealants is concern about inadvertently sealing over caries. The objective of this meta-analysis was to examine whether sealants are effective in preventing caries progression. Methods: Our search of electronic databases for comparative studies examining caries progression in sealed permanent teeth located 1905 unique records. We ordered 311 articles that met the inclusion criteria. We used a random-effects model to estimate percentage reduction in caries progression in sealed carious teeth compared to not-sealed carious teeth. Findings: Six studies including 4 randomized-controlled trials (RCT) were used in the analysis (1090 surfaces, 840 teeth, and 384 persons). The median annual percentage of non-cavitated lesions progressing was 2.6% for sealed and 12.6% for unsealed carious teeth. The summary prevented fraction for RCT was 71.3% (95%CI: 52.8%-82.5%; no observed heterogeneity). Conclusions: Sealing non-cavitated caries in permanent teeth reduces caries progression by over 70% up to 5 years after placement.
Wavelet Image Interpolation (WII): A Wavelet-Based Approach to Enhancement of Digital Mammography Images

(Georgia Institute of Technology, 2007) Derado, Gordana ; Bowman, F. DuBois ; Patel, Rajan ; Newell, Mary ; Vidakovic, Brani

Cancer detection using mammography focuses on characteristics of tiny microcalcifications, including the number, size, and spatial arrangement of microcalcification clusters as well as morphological features of individual microcalcifications. We developed state-of-the-art wavelet-based methods to enhance the resolution of microcalcifications visible in digital mammograms, thereby improving the specificity of breast cancer diagnoses. In our research, we develop, refine, and evaluate a Wavelet Image Interpolation (WII) procedure and create accompanying software to implement it. WII involves the application of an inverse wavelet transformation to a coarse or degraded image and constructed detail coefficients to produce an enhanced higher resolution image. The construction of detail coefficients is supervised by the observed image and innate regular scaling assessed by a statistical model. Methodology we propose was tested by an experienced radiologist in a blind study using 40 images from the University of South Florida Digital Database for Screening Mammography (DDSM) (Heat et. al. [10]).
Bayesian Spherical Wavelet Shrinkage: Applications to Shape Analysis

(Georgia Institute of Technology, 2007) Le Faucheur, Xavier ; Vidakovic, Brani ; Tannenbaum, Allen R.

Multiscale analysis has become a very useful tool in image processing and computer vision. Our work is motivated by the need to efficiently represent 3D shapes that exhibit a spherical topology. This note presents a wavelet based model for shape denoising and data compression. The 3D shape signal is first encoded using biorthogonal spherical wavelet functions defined a 3D triangulated mesh. We propose a Bayesian thresholding model for this type of second generation wavelet in order to eliminate wavelet coefficients that are considered as noise. This way, we are able to reduce dimension without losing significant information by estimating a noiseless version of our shape.