Machine Learning Model for Predicting CVD Risk on NHANES Data.

Cardiovascular disease (CVD) is a major health problem throughout the world. It is the leading cause of morbidity and mortality and also causes considerable economic burden to society. The early symptoms related to previous observations and abnormal events, which can be subjectively acquired by self-assessment of individuals, bear significant clinical relevance and are regularly preserved in the patient's health record. The aim of our study is to develop a machine learning model based on selected CVD-related information encompassed in NHANES data in order to assess CVD risk. This model can be used as a screening tool, as well as a retrospective reference in association with current clinical data in order to improve CVD assessment. In this form it is planned to be used for mass screening and evaluation of young adults entering their army service. The experimental results are promising in that the proposed model can effectively complement and support the CVD prediction for the timely alertness and control of cardiovascular problems aiming to prevent the occurrence of serious cardiac events.

Integrating Microarray Data and GRNs.

With the completion of the Human Genome Project and the emergence of high-throughput technologies, a vast amount of molecular and biological data are being produced. Two of the most important and significant data sources come from microarray gene-expression experiments and respective databanks (e,g., Gene Expression Omnibus-GEO (http://www.ncbi.nlm.nih.gov/geo)), and from molecular pathways and Gene Regulatory Networks (GRNs) stored and curated in public (e.g., Kyoto Encyclopedia of Genes and Genomes-KEGG (http://www.genome.jp/kegg/pathway.html), Reactome (http://www.reactome.org/ReactomeGWT/entrypoint.html)) as well as in commercial repositories (e.g., Ingenuity IPA (http://www.ingenuity.com/products/ipa)). The association of these two sources aims to give new insight in disease understanding and reveal new molecular targets in the treatment of specific phenotypes.Three major research lines and respective efforts that try to utilize and combine data from both of these sources could be identified, namely: (1) de novo reconstruction of GRNs, (2) identification of Gene-signatures, and (3) identification of differentially expressed GRN functional paths (i.e., sub-GRN paths that distinguish between different phenotypes). In this chapter, we give an overview of the existing methods that support the different types of gene-expression and GRN integration with a focus on methodologies that aim to identify phenotype-discriminant GRNs or subnetworks, and we also present our methodology.

Biclustering strategies for genetic marker selection in gynecologic tumor cell lines.

Over the past few decades great interest has been focused on cell lines derived from tumors, because of their usability as models to understand the biology of cancer. At the same time, advanced technologies such as DNA-microarrays have been broadly used to study the expression level of thousands of genes in primary tumors or cancer cell lines in a single experiment. Results from microarray analysis approaches have provided valuable insights into the underlying biology and proven useful for tumor classification, prognostication and prediction. Our approach utilizes biclustering methods for the discovery of genes with coherent expression across a subset of conditions (cell lines of a tumor type). More specifically, we present a novel modification on Cheng & Church's algorithm that searches for differences across the studied conditions, but also enforces consistent intensity characteristics of each cluster within each condition. The application of this approach on a gynecologic panel of cell lines succeeds to derive discriminant groups of compact bi-clusters across four types of tumor cell lines. In this form, the proposed approach is proven efficient for the derivation of tumor-specific markers.

Extracting reliable gene expression signatures through Stable Bootstrap Validation.

Identification of candidate genes responsible for specific phenotypes, such as cancer, has been a major challenge in the field of bioinformatics. Given a DNA Microarray dataset, traditional feature selection methods produce lists of candidate genes which vary significantly under variations of the training data. That instability hinders the validity of research findings and raises doubts about the reliability of such methods. In this study, we propose a framework for the extraction of stable genomic signatures. The proposed methodology enforces stability at the validation step, independent of the feature selection and classification methods used. The statistical significance of the selected gene set is also assessed. The results of this study demonstrate the importance of stability issues in genomic signatures, beyond their prediction capabilities.

Temporal and spatial patterns of gene profiles during chondrogenic differentiation.

Clustering analysis based on temporal profile of genes may provide new insights in particular biological processes or conditions. We report such an integrative clustering analysis which is based on the expression patterns but is also influenced by temporal changes. The proposed platform is illustrated with a temporal gene expression dataset comprised of pellet culture-conditioned human primary chondrocytes and human bone marrow-derived mesenchymal stem cells (MSCs). We derived three clusters in each cell type and compared the content of these classes in terms of temporal changes. We further considered the induced biological processes and the gene-interaction networks formed within each cluster and discuss their biological significance. Our proposed methodology provides a consistent tool that facilitates both the statistical and biological validation of temporal profiles through spatial gene network profiles.

Nonparametric network design and analysis of disease genes in oral cancer progression.

Biological networks in living organisms can be seen as the ultimate means of understanding the underlying mechanisms in complex diseases, such as oral cancer. During the last decade, many algorithms based on high-throughput genomic data have been developed to unravel the complexity of gene network construction and their progression in time. However, the small size of samples compared to the number of observed genes makes the inference of the network structure quite challenging. In this study, we propose a framework for constructing and analyzing gene networks from sparse experimental temporal data and investigate its potential in oral cancer. We use two network models based on partial correlations and kernel density estimation, in order to capture the genetic interactions. Using this network construction framework on real clinical data of the tissue and blood at different time stages, we identified common disease-related structures that may decipher the association between disease state and biological processes in oral cancer. Our study emphasizes an altered MET (hepatocyte growth factor receptor) network during oral cancer progression. In addition, we demonstrate that the functional changes of gene interactions during oral cancer progression might be particularly useful for patient categorization at the time of diagnosis and/or at follow-up periods.

Identification of altered MET network in oral cancer progression based on nonparametric network design.

Oral cancer is characterized by multiple genetic events such as alterations of a number of oncogenes and tumour suppressor genes. The aim of this study is to identify genes and their functional interactions that may play a crucial role on a specific disease-state, especially during oral cancer progression. We examine gene interaction networks on blood genomic data, obtained from twenty three oral cancer patients at four different time stages. We generate the gene-gene networks from sparse experimental temporal data using two methods, Partial Correlations and Kernel Density Estimation, in order to capture genetic interactions. The network study reveals an altered MET (hepatocyte growth factor receptor) network during oral cancer progression, which is further analyzed in relation to other studies.

On the relevance of glycolysis process on brain gliomas.

The proposed analysis considers aspects of both statistical and biological validation of the glycolysis effect on brain gliomas, at both genomic and metabolic level. In particular, two independent datasets are analyzed in parallel, one engaging genomic (Microarray Expression) data and the other metabolomic (Magnetic Resonance Spectroscopy Imaging) data. The aim of this study is twofold. First to show that, apart from the already studied genes (markers), other genes such as those involved in the human cell glycolysis significantly contribute in gliomas discrimination. Second, to demonstrate how the glycolysis process can open new ways towards the design of patient-specific therapeutic protocols. The results of our analysis demonstrate that the combination of genes participating in the glycolytic process (ALDOA, ALDOC, ENO2, GAPDH, HK2, LDHA, LDHB, MDH1, PDHB, PFKM, PGI, PGK1, PGM1 and PKLR) with the already known tumor suppressors (PTEN, Rb, TP53), oncogenes (CDK4, EGFR, PDGF) and HIF-1, enhance the discrimination of low versus high-grade gliomas providing high prediction ability in a cross-validated framework. Following these results and supported by the biological effect of glycolytic genes on cancer cells, we address the study of glycolysis for the development of new treatment protocols.

Strengths and weaknesses of 1.5T and 3T MRS data in brain glioma classification.

Although magnetic resonance spectroscopy (MRS) methods of 1.5Tesla (T) and 3T have been widely applied during the last decade for noninvasive diagnostic purposes, only a few studies have been reported on the value of the information extracted in brain cancer discrimination. The purpose of this study is threefold. First, to show that the diagnostic value of the information extracted from two different MRS scanners of 1.5T and 3T is significantly influenced in terms of brain gliomas discrimination. Second, to statistically evaluate the discriminative potential of publicly known metabolic ratio markers, obtained from these two types of scanners in classifying low-, intermediate-, and high-grade gliomas. Finally, to examine the diagnostic value of new metabolic ratios in the discrimination of complex glioma cases where the diagnosis is both challenging and critical. Our analysis has shown that although the information extracted from 3T MRS scanner is expected to provide better brain gliomas discrimination; some factors like the features selected, the pulse-sequence parameters, and the spectroscopic data acquisition methods can influence the discrimination efficiency. Finally, it is shown that apart from the bibliographical known, new metabolic ratio features such as N-acetyl aspartate/ S, Choline/ S, Creatine/ S , and myo-Inositol/ S play significant role in gliomas grade discrimination.

Intertrial coherence and causal interaction among independent EEG components.

Over the past few years there has been an increased interest in studying the underlying neural mechanism of attention and cognitive brain activity. This paper aims towards identifying and analyzing distinct responses in an auditory working memory paradigm, as independent components with variable latency, frequency and phase characteristics. The event-related nature of components (either phase or non-phase-locked) over multiple trials is thoroughly examined through intertrial coherence measures. Furthermore, the functional coupling of independent components is investigated through the concept of partial directed coherence depicted as a directed graph. Using these tools, the paper compares issues of activation, connectivity and directionality in the synchronization maps of two populations, of control and Alzheimer's subjects. The results on real data from an oddball experiment verify and further enhance the findings of previous studies and illustrate the potential of the proposed analysis framework.

Waveform analysis of non-oscillatory independent components in single-trial auditory event-related activity in healthy subjects and Alzheimer's disease patients.

The objective was to characterize the non-oscillatory independent components (ICs) of the auditory event-related potential (ERP) waveform of an oddball task for normal and newly diagnosed Alzheimer's disease (AD) subjects, and to seek biomarkers for AD. Single trial ERP waveforms were analysed using independent components analysis (ICA) and k-means clustering. Two stages of clustering depended upon the magnitudes and latencies, and the scalp topographies of the non-oscillatory back-projected ICs (BICs) at electrode Cz. The electrical current dipole sources of the BICs were located using Low Resolution Electromagnetic Tomography (LORETA). Generally 3-10 BICs, of different latencies and polarities, occurred in each trial. Each peak was associated with positive and negative BICs. The trial-to-trial variations in their relative numbers and magnitudes may explain the variations in the averaged ERP reported, and the delay in the averaged P300 for AD patients. The BIC latencies, topographies and electrical current density maximum locations varied from trial-to-trial. Voltage foci in the BIC topographies identify the BIC source locations. Since statistical differences were found between the BICs in healthy and AD subjects, the method might provide reliable biomarkers for AD, if these findings are reproduced in a larger study, independently of other factors influencing the comparison of the two populations. The method can extract artefact- and EEG-free single trial ERP waveforms, offers improved ERP averages by selecting the trials on the basis of their BICs, and is applicable to other evoked potentials, conditions and diseases.

Revealing the metabolic profile of brain tumors for diagnosis purposes.

The metabolic behavior of complex brain tumors, like Gliomas and Meningiomas, with respect to their type and grade was investigated in this paper. Towards this direction the smallest set of the most representative metabolic markers for each brain tumor type was identified, using ratios of peak areas of well established metabolites, from (1)H-MRSI (Proton Magnetic Resonance Spectroscopy Imaging) data of 24 patients and 4 healthy volunteers. A feature selection method that embeds Fisher's filter criterion into a wrapper selection scheme was applied; Support Vector Machine (SVM) and Least Squares-SVM (LS-SVM) classifiers were used to evaluate the ratio markers classification significance. The area under the Receiver Operating Characteristic curve (AUROC) was adopted to evaluate the classification significance. It is found that the NAA/CHO, CHO/S, MI/S ratios can be used to discriminate Gliomas and Meningiomas from Healthy tissue with AUROC greater than 0.98. Ratios CHO/S, CRE/S, MI/S, LAC/CRE, ALA/CRE, ALA/S and LIPS/CRE can identify type and grade differences in Gliomas giving AUROC greater than 0.98 apart from the scheme of Gliomas grade II vs grade III where 0.84 was recorded due to high heterogeneity. Finally NAA/CRE, NAA/S, CHO/S, MI/S and ALA/S manage to discriminate Gliomas from Meningiomas providing AUROC exceeding 0.90.

Time-frequency analysis methods to quantify the time-varying microstructure of sleep EEG spindles: possibility for dementia biomarkers?

The time-varying microstructure of sleep EEG spindles may have clinical significance in dementia studies and can be quantified with a number of techniques. In this paper, real and simulated sleep spindles were regarded as AM/FM signals modeled by six parameters that define the instantaneous envelope (IE) and instantaneous frequency (IF) waveforms for a sleep spindle. These parameters were estimated using four different methods, namely the Hilbert transform (HT), complex demodulation (CD), matching pursuit (MP) and wavelet transform (WT). The average error in estimating these parameters was lowest for HT, higher but still less than 10% for CD and MP, and highest (greater than 10%) for WT. The signal distortion induced by the use of a given method was greatest in the case of HT and MP. These two techniques would necessitate the removal of about 0.4s from the spindle data, which is an important limitation for the case of spindles with duration less than 1s. Although the CD method may lead to a higher error than HT and MP, it requires a removal of only about 0.23s of data. An application of this sleep spindle parameterization via the CD method is proposed, in search of efficient EEG-based biomarkers in dementia. Preliminary results indicate that the proposed parameterization may be promising, since it can quantify specific differences in IE and IF characteristics between sleep spindles from dementia subjects and those from aged controls.

Potential dementia biomarkers based on the time-varying microstructure of sleep EEG spindles.

The time-varying microstructure of sleep EEG spindles may have clinical significance in dementia studies. In this work, the sleep spindle is modeled as an AM-FM signal and parameterized in terms of six parameters, three quantifying the instantaneous envelope (IE) and three quantifying the instantaneous frequency (IF) of the spindle model. The IE and IF waveforms of sleep spindles from patients with dementia and normal controls were estimated using the time-frequency technique of Complex Demodulation (CD). Sinusoidal curve-fitting using a matching pursuit (MP) approach was applied to the IE and IF waveforms for the estimation of the six model parameters. Specific differences were found in sleep spindle instantaneous frequency dynamics between spindles from dementia subjects and spindles from controls.

Bioprofiling over grid for eHealthcare.

A trend in modern medicine is towards individualization of healthcare and, potentially, grid computing can play an important role in this by allowing sharing of resources and expertise to improve the quality of care. In this paper, we present a new test bed, the BIOPATTERN Grid, which aims to fulfil this role in the long term. The main objectives in this paper are 1) to report the development of the BIOPATTERN Grid, for biopattern analysis and bioprofiling in support of individualization of healthcare. The BIOPATTERN Grid is designed to facilitate secure and seamless sharing of geographically distributed bioprofile databases and to support the analysis of bioprofiles to combat major diseases such as brain diseases and cancer within a major EU project, BIOPATTERN (www.biopattern.org); 2) to illustrate how the BIOPATTERN Grid could be used for biopattern analysis and bioprofiling for early detection of dementia and for brain injury assessment on an individual basis. We highlight important issues that would arise from the mobility of citizens in the EU, such as those associated with access to medical data, ethical and security; and 3) to describe two grid services which aim to integrate BIOPATTERN Grid with existing grid projects on crawling service and remote data acquisition which is necessary to underpin the use of the test bed for biopattern analysis and bioprofiling.

Modeling the time-varying microstructure of simulated sleep EEG spindles using time-frequency analysis methods.

The time-varying microstructure of sleep spindles may have clinical significance and can be quantified and modeled with a number of techniques. In this paper, sleep spindles were regarded as AM-FM signals modeled by six parameters. The instantaneous envelope (IE) and instantaneous frequency (IF) waveforms were estimated using four different methods, namely Hilbert Transform (HT), Complex Demodulation (CD), Wavelet Transform (WT) and Matching Pursuit (MP). The six model parameters were subsequently estimated from the IE and IF waveforms. The average error, taking into account the error for each model parameter, was lowest for HT, higher but still less than 10% for CD and MP, and highest (greater than 10%) for WT, for three different spindle model examples. The amount of distortion induced by the use of a given method is also important; distortion was the greatest (0.4 sec) in the case of HT. Therefore, in the case of real spindles, one could utilize CD and MP and, if the spindle duration is more than 1 sec, HT as well.

An automated approach for analyzing D-periods in collagen fibril images.

This paper considers an approach for analyzing fibrillar collagen structures based on fundamental concepts of pattern recognition. It focuses on the quantitative comparison between collagen structural data (electron-optical data) and chemical data. Theoretical models in the form of sequence-generated histograms are used as reference for extracting and analyzing the structural unit in images from collagen fibrils. In this respect, collagen provides a valuable model system for studying the chemical basis of ultrastructure, as well as detecting the alterations in collagen fibril structure produced by a disorder. Application examples are presented and the results are compared with biochemical studies.

Analysis of fibrous proteins from electron microscopy images.

This paper considers an approach for analyzing fibrillar collagen structures from electron microscopy images. It enables the quantitative comparison between collagen structural data (electron-optical data) and chemical data. The particular objectives of the paper are to model the electron microscopy images according to the periodic structure of collagen, provide methods for extracting periodic features directly from the experimental data and propose schemes for comparing these features with the theoretical amino-acid distributions of the examined collagen tissue. Theoretical models in the form of sequence-generated histograms are used as reference for extracting and analyzing the structural unit in images from collagen fibrils. In this respect, collagen provides a valuable model system for studying the chemical basis of ultra-structure and the mechanisms of various treatments on a protein, as well as detecting the alterations in collagen fibril structure produced by a disorder. The algorithms developed in this study can be applied to any fibrous protein, provided that its amino acid sequences and structural properties are known. Several application examples are presented. The algorithmic results are compared with clinical studies as to verify the applicability and potential of the proposed methodology.

High-Order Neural Network structure selection for function approximation applications using Genetic Algorithms.

Neural network literature for function approximation is by now sufficiently rich. In its complete form, the problem entails both parametric (i.e., weights determination) and structural learning (i.e., structure selection). The majority of works deal with parametric uncertainty assuming knowledge of the appropriate neural structure. In this paper we present an algorithmic approach to determine the structure of High Order Neural Networks (HONNs), to solve function approximation problems. The method is based on a Genetic Algorithm (GA) and is equipped with a stable update law to guarantee parametric learning. Simulation results on an illustrative example highlight the performance and give some insight of the proposed approach.

A hybrid neural network/genetic algorithm approach to optimizing feature extraction for signal classification.

In this paper, a hybrid neural network/genetic algorithm technique is presented, aiming at designing a feature extractor that leads to highly separable classes in the feature space. The application upon which the system is built, is the identification of the state of human peripheral vascular tissue (i.e., normal, fibrous and calcified). The system is further tested on the classification of spectra measured from the cell nucleii in blood samples in order to distinguish normal cells from those affected by Acute Lymphoblastic Leukemia. As advantages of the proposed technique we may encounter the algorithmic nature of the design procedure, the optimized classification results and the fact that the system performance is less dependent on the classifier type to be used.