Books by Petros Xanthopoulos
Data uncertainty is a concept closely related with most real life applications that involve data ... more Data uncertainty is a concept closely related with most real life applications that involve data collection and interpretation. Examples can be found in data acquired with biomedical instruments or other experimental techniques. Integration of robust optimization in the existing data mining techniques aim to create new algorithms resilient to error and noise.
This work encapsulates all the latest applications of robust optimization in data mining. This brief contains an overview of the rapidly growing field of robust data mining research field and presents the most well known machine learning algorithms, their robust counterpart formulations and algorithms for attacking these problems.
This brief will appeal to theoreticians and data miners working in this field.
This volume covers some of the topics that are related to the rapidly growing field of biomedica... more This volume covers some of the topics that are related to the rapidly growing field of biomedical informatics. In June 11–12, 2010 a workshop entitled ‘Optimization and Data Analysis in Biomedical Informatics’ was organized at The Fields Institute. Following this event, invited contributions were gathered based on the talks presented at the workshop, and additional invited chapters were solicited from leading experts. In this publication, the authors share their expertise in the form of state-of-the-art research and review chapters, bringing together researchers from different disciplines and emphasizing the value of mathematical methods in the areas of clinical sciences.
This work is targeted to applied mathematicians, computer scientists, industrial engineers, and clinical scientists who are interested in exploring emerging and fascinating interdisciplinary topics of research. It is designed to further stimulate and enhance fruitful collaborations between scientists from different disciplines.
The human brain is among the most complex systems known to mankind. Neuroscientists seek to under... more The human brain is among the most complex systems known to mankind. Neuroscientists seek to understand brain function through detailed analysis of neuronal excitability and synaptic transmission. Only in the last few years has it become feasible to capture simultaneous responses from a large enough number of neurons to empirically test the theories of human brain function computationally. This book is comprised of state-of-the-art experiments and computational techniques that provide new insights and improve our understanding of the human brain.
This volume includes contributions from diverse disciplines including electrical engineering, biomedical engineering, industrial engineering, and medicine, bridging a vital gap between the mathematical sciences and neuroscience research. Covering a wide range of research topics, this volume demonstrates how various methods from data mining, signal processing, optimization and cutting-edge medical techniques can be used to tackle the most challenging problems in modern neuroscience.
The results presented in this book are of great interest and value to scientists, graduate students, researchers and medical practitioners interested in the most recent developments in computational neuroscience.
Data Mining for Biomarker Discovery is designed to motivate collaboration and discussion among va... more Data Mining for Biomarker Discovery is designed to motivate collaboration and discussion among various disciplines and will be of interest to students and researchers in engineering, computer science, applied mathematics, medicine, and anyone interested in the interdisciplinary application of data mining techniques. Biomarker discovery is an important area of biomedical research that can lead to significant breakthroughs in disease analysis and targeted therapy. Moreover, the discovery and management of new biomarkers is a challenging and attractive problem in the emerging field of biomedical informatics.
This volume is a collection of state-of-the-art research from select participants of the “International Conference on Biomedical Data and Knowledge Mining: Towards Biomarker Discovery,” held July 7-9, 2010 in Chania, Greece. Contributions focus on biomarker data integration, information retrieval methods, and statistical machine learning techniques, all presented with new results, models, and algorithms.
Papers by Petros Xanthopoulos
Journal of Neuroengineering and Rehabilitation, 2008
In this primer, we give a review of the inverse problem for EEG source localization. This is inte... more In this primer, we give a review of the inverse problem for EEG source localization. This is intended for the researchers new in the field to get insight in the state-of-the-art techniques used to find approximate solutions of the brain sources giving rise to a scalp potential recording. Furthermore, a review of the performance results of the different techniques is provided to compare these different inverse solutions. The authors also include the results of a Monte-Carlo analysis which they performed to compare four non parametric algorithms and hence contribute to what is presently recorded in the literature. An extensive list of references to the work of other researchers is also provided. This paper starts off with a mathematical description of the inverse problem and proceeds to discuss the two main categories of methods which were developed to solve the EEG inverse problem, mainly the non parametric and parametric methods. The main difference between the two is to whether a fixed number of dipoles is assumed a priori or not. Various techniques falling within these categories are described including minimum norm estimates and their generalizations, LORETA, sLORETA, VARETA, S-MAP, ST-MAP, Backus-Gilbert, LAURA, Shrinking LORETA FOCUSS (SLF), SSLOFO and ALF for non parametric methods and beamforming techniques, BESA, subspace techniques such as MUSIC and methods derived from it, FINES, simulated annealing and computational intelligence algorithms for parametric methods. From a review of the performance of these techniques as documented in the literature, one could conclude that in most cases the LORETA solution gives satisfactory results. In situations involving clusters of dipoles, higher resolution algorithms such as MUSIC or FINES are however preferred. Imposing reliable biophysical and psychological constraints, as done by LAURA has given superior results. The Monte-Carlo analysis performed, comparing WMN, LORETA, sLORETA and SLF, for different noise levels and different simulated source depths has shown that for single source localization, regularized sLORETA gives the best solution in terms of both localization error and ghost sources. Furthermore the computationally intensive solution given by SLF was not found to give any additional benefits under such simulated conditions.
Parameter estimation for an assumed sleep EEG spindle model (AM-FM signal) is performed by using ... more Parameter estimation for an assumed sleep EEG spindle model (AM-FM signal) is performed by using four time-frequency analysis methods. Results from simulated as well as from real data are presented. In simulated data, the Hilbert Transform-based method has the lowest average percentage error but produces considerable signal distortion. The Complex Demodulation and the Matching Pursuit-based methods have error rates below 10%, but the Matching Pursuit-based method produces considerable signal distortion as well. The Wavelet Transform-based method has the poorest performance. In real data, all methods produce reasonable parameter values. However, the Hilbert Transform and the Matching Pursuitbased methods may not be applicable for sleep spindles shorter than about 0.8 sec. Matching Pursuit-based curve fitting is utilized as part of the parameter estimation process.
The reconstruction of the brain current sources from scalp electric recordings (Electroen‐cephalo... more The reconstruction of the brain current sources from scalp electric recordings (Electroen‐cephalogram) also known as the inverse source localization problem is a highly underdetermined problem in the field of computational neuroscience, and this problem still remains open. In this chapter we propose an alternative formulation for the inverse electroencephalography (EEG) problem based on optimization theory. For simulation purposes, a three shell realistic head model based on an averaged magnetic resonance ...
In this chapter a potential problem with application of the Granger‐causality based on the simple... more In this chapter a potential problem with application of the Granger‐causality based on the simple vector autoregressive (VAR) modeling to EEG data is investigated. Although some initial studies tested whether the data support the stationarity assumption of VAR, the stability of the estimated model is rarely (if ever) been verified. In fact, in cases when the stability condition is violated the process may exhibit a random walk like behavior or even be explosive. The problem is illustrated by an example.
Discrimination of different cell types is very important in many medical and biological applicati... more Discrimination of different cell types is very important in many medical and biological applications. Existing methodologies are based on cost inefficient technologies or tedious one-by-one empirical examination of the cells. Recently, Raman spectroscopy, a inexpensive and efficient method, has been employed for cell discrimination. Nevertheless, the traditional protocols for analyzing Raman spectra require preprocessing and peak fitting analysis which does not allow simultaneous examination of many spectra. In this paper we examine the applicability of supervised learning algorithms in the cell differentiation problem. Five different methods are presented and tested on two different datasets. Computational results show that machine learning algorithms can be employed in order to automate cell discrimination tasks.
This paper deals with the progress made in applications of quantum computing in control and optim... more This paper deals with the progress made in applications of quantum computing in control and optimization. It concentrates on applying the geometric technique in order to investigate a finite control problem of a two-level quantum system, resonance control of a three-level system, simulation of bilinear quantum control systems, and optimal control using the Bellman principle. We show that a quantum object described by a Schroedinger equation can be controlled in an optimal way by electromagnetic modes. We also demonstrate an application of these techniques and an algebra-geometric approach to the study of dynamic processes in nonlinear systems. The information processing by means of controlled quantum lattices is discussed: we present new mathematical models of classical (CL) and quantum-mechanical lattices (QML) and their application to information processing. system-theoretical results on the observability, controllability and minimal realizability theorems are formulated for cl. The cellular dynamaton (CD) based on quantum oscillators is presented. Cellular's quantum computational search procedure can provide the basis for implementing adaptive global optimization algorithms. A brief overview of the procedure is given and a framework called lattice adaptive search is set up. A method of Yatsenko and one introduced by the authors fit into this framework and are compared.
Many studies have documented abnormal horizontal and vertical eye movements in human neurodegener... more Many studies have documented abnormal horizontal and vertical eye movements in human neurodegenerative disease as well as during altered states of consciousness (including drowsiness and intoxication) in healthy adults. Eye movement measurement may play an important role measuring the progress of neurodegenerative diseases and state of alertness in healthy individuals. There are several techniques for measuring eye movement, Infrared detection technique (IR). Video-oculography (VOG), Scleral eye coil and EOG. Among those available recording techniques, EOG is a major source for monitoring the abnormal eye movement. In this real-time quantitative analysis study, the methods which can capture the characteristic of the eye movement were proposed to accurately categorize the state of neurodegenerative subjects. The EOG recordings were taken while 5 tested subjects were watching a short (>120 s) animation clip. In response to the animated clip the participants executed a number of eye movements, including vertical smooth pursued (SVP), horizontal smooth pursued (HVP) and random saccades (RS). Detection of abnormalities in ocular movement may improve our diagnosis and understanding a neurodegenerative disease and altered states of consciousness. A standard real-time quantitative analysis will improve detection and provide a better understanding of pathology in these disorders.
In this chapter, we consider an optimization technique for estimating the Lyapunov exponents from... more In this chapter, we consider an optimization technique for estimating the Lyapunov exponents from nonlinear chaotic systems. We then describe an algorithm for solving the optimization model and discuss the computational aspects of the proposed algorithm. To show the efficiency of the algorithm, we apply it to some well-known data sets. Numerical tests show that the algorithm is robust and quite effective, and its performance is comparable with that of other well-known algorithms.
The unified myoclonus rating scale (UMRS) has been utilized to assess the severity of myoclonus a... more The unified myoclonus rating scale (UMRS) has been utilized to assess the severity of myoclonus and the efficacy of antiepileptic drug (AED) treatment in patients with Unverricht–Lundborg disease (ULD). Electroencephalographic (EEG) recordings are normally used as a supplemental tool for the diagnosis of epilepsy disorders. In this study, mutual information and nonlinear interdependence measures were applied to the EEG recordings in an attempt to identify the effect of treatment on the coupling strength and directionality of mutual information and nonlinear interdependences between different brain cortical regions. Two 1-h EEG recordings were acquired from four ULD subjects; one prior and one after a minimum of 2 months treatment with an add-on AED. Subjects in this study were siblings of same parents and suffered from ULD for approximately 37 years. Our results indicated that the coupling strength was low between different brain cortical regions in the patients with disease of less severity. Adjunctive AED treatment was associated with significant decrease of the coupling strength in all subjects. The mutual information between different brain cortical regions was also reduced after treatment. These findings could provide a new insight for developing a novel surrogate outcome measure for patients with epilepsy when clinical tools or observations could potentially fail to detect a significant difference.
Cancer is one of the leading causes of death throughout the world. Advancements in early and impr... more Cancer is one of the leading causes of death throughout the world. Advancements in early and improved diagnosis could help prevent a significant number of these deaths. Raman spectroscopy is a vibrational spectroscopic technique which has received considerable attention recently with regards to applications in clinical oncology. Raman spectroscopy has the potential not only to improve diagnosis of cancer but also to advance the treatment of cancer.
Abstract. Many studies have documented abnormal horizontal and vertical eye movements in human ne... more Abstract. Many studies have documented abnormal horizontal and vertical eye movements in human neurodegenerative disease as well as during altered states of consciousness (including drowsiness and intoxication) in healthy adults. Eye movement measurement may play an important role measuring the progress of neurodegenerative diseases and state of alertness in healthy individuals.
Abstract This paper deals with the initial modeling of water salinity and its diffusion into the ... more Abstract This paper deals with the initial modeling of water salinity and its diffusion into the lakes during lock operation on the Panama Canal. A hybrid operational model was implemented using the AnyLogic software simulation environment. This was accomplished by generating an operational discrete-event simulation model and a continuous simulation model based on differential equations, whichmodeled the salinity diffusion in the lakes.
Abstract The time-varying microstructure of sleep EEG spindles may have clinical significance in ... more Abstract 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, the sleep spindle is modeled as an AM-FM signal in terms of six parameters, three quantifying the instantaneous envelope (IE) and three quantifying the instantaneous frequency (IF) of the spindle model. An application of such parameterization is proposed, in search of EEG-based biomarkers in dementia.
Discrimination of different cell types is very important in many medical and biological applicati... more Discrimination of different cell types is very important in many medical and biological applications. Existing methodologies are based on cost inefficient technologies or tedious one-by-one empirical examination of the cells. Recently, Raman spectroscopy, a inexpensive and efficient method, has been employed for cell discrimination. Nevertheless, the traditional protocols for analyzing Raman spectra require preprocessing and peak fitting analysis which does not allow simultaneous examination of many spectra.
abstract This paper deals with the progress made in applications of quantum computing in control ... more abstract This paper deals with the progress made in applications of quantum computing in control and optimization. It concentrates on applying the geometric technique in order to investigate a finite control problem of a two-level quantum system, resonance control of a three-level system, simulation of bilinear quantum control systems, and optimal control using the Bellman principle. We show that a quantum object described by a Schroedinger equation can be controlled in an optimal way by electromagnetic modes.
EEG spindle model (AM-FM signal) is performed by using four time-frequency analysis methods. Resu... more EEG spindle model (AM-FM signal) is performed by using four time-frequency analysis methods. Results from simulated as well as from real data are presented. In simulated data, the Hilbert Transform-based method has the lowest average percentage error but produces considerable signal distortion. The Complex Demodulation and the Matching Pursuit-based methods have error rates below 10%, but the Matching Pursuit-based method produces considerable signal distortion as well.
Uploads
Books by Petros Xanthopoulos
This work encapsulates all the latest applications of robust optimization in data mining. This brief contains an overview of the rapidly growing field of robust data mining research field and presents the most well known machine learning algorithms, their robust counterpart formulations and algorithms for attacking these problems.
This brief will appeal to theoreticians and data miners working in this field.
This work is targeted to applied mathematicians, computer scientists, industrial engineers, and clinical scientists who are interested in exploring emerging and fascinating interdisciplinary topics of research. It is designed to further stimulate and enhance fruitful collaborations between scientists from different disciplines.
This volume includes contributions from diverse disciplines including electrical engineering, biomedical engineering, industrial engineering, and medicine, bridging a vital gap between the mathematical sciences and neuroscience research. Covering a wide range of research topics, this volume demonstrates how various methods from data mining, signal processing, optimization and cutting-edge medical techniques can be used to tackle the most challenging problems in modern neuroscience.
The results presented in this book are of great interest and value to scientists, graduate students, researchers and medical practitioners interested in the most recent developments in computational neuroscience.
This volume is a collection of state-of-the-art research from select participants of the “International Conference on Biomedical Data and Knowledge Mining: Towards Biomarker Discovery,” held July 7-9, 2010 in Chania, Greece. Contributions focus on biomarker data integration, information retrieval methods, and statistical machine learning techniques, all presented with new results, models, and algorithms.
Papers by Petros Xanthopoulos
This work encapsulates all the latest applications of robust optimization in data mining. This brief contains an overview of the rapidly growing field of robust data mining research field and presents the most well known machine learning algorithms, their robust counterpart formulations and algorithms for attacking these problems.
This brief will appeal to theoreticians and data miners working in this field.
This work is targeted to applied mathematicians, computer scientists, industrial engineers, and clinical scientists who are interested in exploring emerging and fascinating interdisciplinary topics of research. It is designed to further stimulate and enhance fruitful collaborations between scientists from different disciplines.
This volume includes contributions from diverse disciplines including electrical engineering, biomedical engineering, industrial engineering, and medicine, bridging a vital gap between the mathematical sciences and neuroscience research. Covering a wide range of research topics, this volume demonstrates how various methods from data mining, signal processing, optimization and cutting-edge medical techniques can be used to tackle the most challenging problems in modern neuroscience.
The results presented in this book are of great interest and value to scientists, graduate students, researchers and medical practitioners interested in the most recent developments in computational neuroscience.
This volume is a collection of state-of-the-art research from select participants of the “International Conference on Biomedical Data and Knowledge Mining: Towards Biomarker Discovery,” held July 7-9, 2010 in Chania, Greece. Contributions focus on biomarker data integration, information retrieval methods, and statistical machine learning techniques, all presented with new results, models, and algorithms.
Methods and material: Raman spectra are collected for 84 samples of A549 cell line (human lung cancer epithelia cells) that has been exposed to toxins to simulate the necrotic and apoptotic death. The proposed data mining approach for the multiclass cell death discrimination problem uses a multiclass regular- ized generalized eigenvalue algorithm for classification (multiReGEC), together with a dimensionality reduction algorithm based on spectral clustering.
Results: The proposed algorithmic scheme can classify A549 lung cancer cells from three different classes (apoptotic death, necrotic death and control cells) with 97.78% ± 0.047 accuracy versus 92.22 ± 0.095 without the proposed feature selection preprocessing. The spectrum areas depicted by the algorithm corresponds to the ⟩C   O bond from the lipids and the lipid bilayer. This chemical structure undergoes different change of state based on cell death type. Further evidence of the validity of the technique is obtained through the successful classification of 7 cell spectra that undergo hyperthermic treatment. Conclusions: In this study we propose a fast and automated way of processing Raman spectra for cell death discrimination, using a feature selection algorithm that not only enhances the classification accuracy, but also gives more insight in the undergoing cell death process.