Paul Dan Cristea

Wavelet image compression - the quadtree coding approach

Perfect reconstruction, quality scalability and region-of-interest coding are basic features needed for the image compression schemes used in telemedicine applications. This paper proposes a new wavelet-based embedded compression technique that efficiently exploits the intraband dependencies and uses a quadtree-based approach to encode the significance maps. The algorithm produces a losslessly compressed embedded data stream, supports quality scalability and permits region-of-interest coding. Moreover, experimental results obtained on various images show that the proposed algorithm provides competitive lossless/lossy compression results. The proposed technique is well-suited for telemedicine applications that require fast interactive handling of large image sets over networks with limited and/or variable bandwidth.

Large scale features in DNA genomic signals

Abstract Complex representations of the nucleotides, codons and amino acids derived from the projection of the Genetic Code Tetrahedron on adequately oriented planes are presented. By converting the sequences of nucleotides and polypeptides into digital genomic signals, this approach offers the possibility of using signal processing methods for the analysis of genomic information. New tools for genomic signal analysis are introduced at the nucleotide, codon and amino acid levels, in a multiresolution approach. It is shown that some important features of nucleotide sequences can be revealed using these signal representations. The paper reports the existence of large scale and global trends of DNA genomic signals in both eukaryotes and prokaryotes, reflecting an almost constant second order nucleotide statistics along DNA strands even at the points where the first order nucleotide statistics show marked changes, as it is the case in prokaryotes.

WAVELET-BASED LOSSLESS COMPRESSION SCHEME WITH PROGRESSIVE TRANSMISSION CAPABILITY

Lossless image compression with progressive transmission capabilities plays a key role in measurement applications, requiring quantitative analysis and involving large sets of images. This work proposes a wavelet‐based compression scheme that is able to operate in the lossless mode. The quantization module implements a new technique for the coding of the wavelet coefficients that is more effective than the classical zerotree coding. The experimental results obtained on a set of multimodal medical images show that the proposed algorithm outperforms the embedded zerotree coder combined with the integer wavelet transform by 0.28 bpp, the set‐partitioning coder by 0.1 bpp, and the lossless JPEG coder by 0.6 bpp. The scheme produces a losslessly compressed embedded data stream; hence, it supports progressive refinement of the decompressed images. Therefore, it is a good candidate for telematics applications requiring fast user interaction with the image data, retaining the option of lossless transmission and archiving of the images.

Wavelet-based lossless compression of coronary angiographic images

The final diagnosis in coronary angiography has to be performed on a large set of original images. Therefore, lossless compression schemes play a key role in medical database management and telediagnosis applications. This paper proposes a wavelet-based compression scheme that is able to operate in the lossless mode. The quantization module implements a new way of coding of the wavelet coefficients that is more effective than the classical zerotree coding. The experimental results obtained on a set of 20 angiograms show that the algorithm outperforms the embedded zerotree coder, combined with the integer wavelet transform, by 0.38 bpp, the set partitioning coder by 0.21 bpp, and the lossless JPEG coder by 0.71 bpp. The scheme is a good candidate for radiological applications such as teleradiology and picture archiving and communications systems (PACSs).

Genetic signal representation and analysis

An original tetrahedral representation of the Genetic Code (GC), that better catches its structure, degeneracy and evolution trends, is defined. The possibility to reduce the dimensionality of the description by the projection of the GC tetrahedron on an adequately oriented plane is also considered, leading to complex representations of the GC. On these bases, optimal symbolic-to-digital mappings of the linear, one-dimensional and one-directional strands of nucleic acids into real or complex genetic signals are derived at nucleotide, codon and amino acid levels. By converting the sequences of nucleotides and polypeptides into digital genetic signals, this approach opens the possibility to use a large variety of signal processing methods for their processing and analysis. It is also shown that some essential features of nucleotide sequences can be better extracted using this representation. Some preliminary results in the comparative analysis of the statistical properties of intragenic vs. intergenic genetic signals are also presented. The use of Independent Component Analysis (ICA) to search for control sequences in the intergenic DNA, i.e., the part of the genome that does not encode proteins, is suggested.

Signal Representation and Processing of Nucleotide Sequences

Sets of related signals can be represented by separating their joint variation and showing the individual signal offsets with respect to this reference. An example is the genomic signal analysis of pathogen variability. The conversion of symbolic nucleotide sequences to genomic signals allows to use signal processing methods to analyze genomic data. This approach reveals striking regularities in the distribution of nucleotides and pair of nucleotides along the sequences, in both prokaryotes and eukaryotes. Genomic signals can also be used for sequence prediction, similarly to time series prediction. The methodology is also adequate for studying the development of pathogen multiple resistance to drugs.

Study of HIV Variability based on Genomic Signal Analysis of Protease and Reverse Transcriptase Genes

The conversion of genomic sequences into digital genomic signals offers the possibility to use signal processing methods for the analysis of genomic information. The study of genomic signals reveals local and global features of chromosomes that would be difficult to identify by using only the symbolic representation used in genomic data bases. The paper presents a study of the HIV protease (PR) and reverse transcriptase (RT) genes by combining standard nucleotide sequence analysis with IT techniques based on the genomic signal approach. Cumulated and unwrapped phases of genomic signals are analyzed to characterize the variability of clade F HIV-1 strains isolated in Romania

Genomic signal analysis of pathogen variability

The paper presents results in the study of pathogen variability by using genomic signals. The conversion of symbolic nucleotide sequences into digital signals offers the possibility to apply signal processing methods to the analysis of genomic data. The method is particularly well suited to characterize small size genomic sequences, such as those found in viruses and bacteria, being a promising tool in tracking the variability of pathogens, especially in the context of developing drug resistance. The paper is based on data downloaded from GenBank [32], and comprises results on the variability of the eight segments of the influenza type A, subtype H5N1, virus genome, and of the Hemagglutinin (HA) gene, for the H1, H2, H3, H4, H5 and H16 types. Data from human and avian virus isolates are used.

Multiresolutional distributed filtering: a novel technique that reduces the amount of data required in high resolution electrocardiography

Abstract High resolution ECG analysis is widely accepted as the best non-invasive technique for the assessment of ventricular tachycardia risk in post-myocardial infarction patients. However, the standard analysis approaches involve an extensive averaging procedure which requires long data records, accompanied by the consequent efforts for storage and transmission. This paper outlines an algorithm for multiresolutional distributed filtering, that can significantly reduce the necessary amount of data. The proposed filtering method comprises three basic steps: the dyadic wavelet transform computation, the shrinkage of the wavelet coefficients using adaptive Bayesian rules, and the reconstruction of the denoised signal through the inverse wavelet transform. The performance evaluation using controlled simulation experiments revealed that the present technique could accelerate the noise reduction, preserving the diagnostic value of the signals.

Genomic signals of chromosomes and of concatenated reoriented coding regions

Symbolic nucleotide sequences are converted into digital genomic signals by using a complex representation derived from a tetrahedral vector representation of nucleotides. The study of complex genomic signals using signal processing methods reveals large scale features of chromosomes that would be difficult to grasp by using the statistical and pattern matching methods for the analysis of symbolic genomic sequences. On the other hand, in the context of operating with a large volume of data at various resolutions and visualizing the results to make them available to humans, the problem of data representability becomes critical. A novel mathematical description of data representability, based on the data scattering ratio on a pixel is defined and is applied for several typical cases of standard signals and for genomic signals. It is shown that the variation of genomic data along nucleotide sequences, specifically the cumulated and unwrapped phase, can be visualized adequately as simple graphic lines for low and large scales, while for medium scales (thousands to tens of thousands of base pairs) the statistical descriptions have to be used.