Liviu Goraş

Technical University of Iasi, Electronics, Telecommunications and Information Technology, Faculty Member

Followers

Following

Co-authors

Public Views

Interests

Uploads

Papers by Liviu Goraş

A Study on Dictionary Selection in Compressive Sensing for ECG Signals Compression and Classification

Biosensors, 2022

The paper proposes a comparative analysis of the projection matrices and dictionaries used for co... more The paper proposes a comparative analysis of the projection matrices and dictionaries used for compressive sensing (CS) of electrocardiographic signals (ECG), highlighting the compromises between the complexity of preprocessing and the accuracy of reconstruction. Starting from the basic notions of CS theory, this paper proposes the construction of dictionaries (constructed directly by cardiac patterns with R-waves, centered or not-centered) specific to the application and the results of their testing. Several types of projection matrices are also analyzed and discussed. The reconstructed signals are analyzed quantitatively and qualitatively by standard distortion measures and by the classification of the reconstructed signals. We used a k-nearest neighbors (KNN) classifier to evaluate the reconstructed models. The KNN module was trained with the models from the mega-dictionary used in the classification block and tested with the models reconstructed with class-specific dictionaries....

Download

Image Analysis Techniques For Cultural Heritage Restoration Methods Evaluation

Publication in the conference proceedings of EUSIPCO, Bucharest, Romania, 2012

Download

Ecg Compressed Sensing Based On Classification In Compressed Space And Specified Dictionaries

Publication in the conference proceedings of EUSIPCO, Barcelona, Spain, 2011

Download

The Analysis of the Specific Dictionaries for Compressive Sensing of EEG Signals

In this paper the possibility of the electroencephalogram (EEG) compressed sensing based on speci... more In this paper the possibility of the electroencephalogram (EEG) compressed sensing based on specific dictionaries is presented. Three types of dictionaries (wavelet, temporal EEG signal specific, and channel specific) are analyzed and the results are expressed through quantitative measures (distortion) and by qualitative measures (stimulus classification rate of the Brain Computer Interface (BC)I paradigm Spelling). KeywordsEEG; Compressed sensing; BCI; P300.

Download

On the Classification of ECG and EEG Signals with Various Degrees of Dimensionality Reduction

Biosensors, 2021

Classification performances for some classes of electrocardiographic (ECG) and electroencephalogr... more Classification performances for some classes of electrocardiographic (ECG) and electroencephalographic (EEG) signals processed to dimensionality reduction with different degrees are investigated. Results got with various classification methods are given and discussed. So far we investigated three techniques for reducing dimensionality: Laplacian eigenmaps (LE), locality preserving projections (LPP) and compressed sensing (CS). The first two methods are related to manifold learning while the third addresses signal acquisition and reconstruction from random projections under the supposition of signal sparsity. Our aim is to evaluate the benefits and drawbacks of various methods and to find to what extent they can be considered remarkable. The assessment of the effect of dimensionality decrease was made by considering the classification rates for the processed biosignals in the new spaces. Besides, the classification accuracies of the initial input data were evaluated with respect to t...

Download

On a compression algorithm for ECG signals

2005 13th European Signal Processing Conference, Sep 1, 2005

The paper presents a new algorithm for ECG signal compression based on local extreme extraction, ... more The paper presents a new algorithm for ECG signal compression based on local extreme extraction, adaptive hysteretic filtering and LZW coding. Basically the method consists in smoothing the ECG signal with a Savitzky-Golay filter, extraction of the local minima and maxima, a hysteretic filtering and LZW coding. The reconstruction of the ECG signal is done by cubic interpolation. The algorithm is robust with respect to noise, has a rather small computational complexity and provides good compression ratios with excellent reconstruction quality. The results of the algorithm compare favourably to other algorithms like AZTEC, SAPA, TP, CORTES, SPHIT, JPEG2000 and wavelet-based ECG compressions.

Results on ECG Compressed Sensing using Specific Dictionaries and its Validation

The paper presents a new method and results regarding the compressed sensing (CS) and classificat... more The paper presents a new method and results regarding the compressed sensing (CS) and classification of ECG waveforms using a general dictionary as well as specific dictionaries built using normal and pathological cardiac patterns. The proposed method has been validated by computation of the distortion errors between the original and the reconstructed signals and by the classification ratio of the reconstructed signals obtained with the k-nearest neighbors (KNN) algorithm.

Preprocessing Method for Improving Ecg Signal Classification and Compression Validation

World Scientific Series on Nonlinear Science Series B, 2010

Download

On the CNN template design for Gabor-type filters based on Pade approximation

SCS 2003. International Symposium on Signals, Circuits and Systems. Proceedings (Cat. No.03EX720), 2003

Gabor filters are widely used in various image processing and computer-vision applications. Being... more Gabor filters are widely used in various image processing and computer-vision applications. Being computationally intensive, analog implementation using Cellular Neural Networks (CNN) can be an attractive solution. In this communication is presented a method for CNN template design of Gabor like filters, based on Pad6 approximation of Gaussian filters. The errors of approximation are evaluated for various neighborhood radii.

Phase influence on mode competition in Turing pattern formation

Proceedings of the 2000 6th IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA 2000) (Cat. No.00TH8509), 2000

ABSTRACT Within linear theory, Turing patterns in CNNs can be viewed as the consequence of a comp... more ABSTRACT Within linear theory, Turing patterns in CNNs can be viewed as the consequence of a competition between unstable spatial modes. The aim of the paper is to show that the final pattern might depend on the relative position (phase) of the competing modes-a result that cannot be explained using the mode decoupling linear theory

Accurate temperature estimation in large DMOS transistors using a semi-empirical thermal conductivity model

ISSCS 2011 - International Symposium on Signals, Circuits and Systems, 2011

ABSTRACT A method for improving the temperature estimation accuracy in large DMOS transistors bas... more ABSTRACT A method for improving the temperature estimation accuracy in large DMOS transistors based on a semi-empirical fitting model for the Si thermal conductivity is presented. The model is used in the analytical solution of the 3D transient heat equation. The performances of the method are further improved by taking into account the power metallisation using a simple RC thermal model. Measurement results show that using this approach, not only the peak temperature, but also the temperature distribution across the device are accurately estimated.

Two-dimensional filter design based on frequency mapping of analog prototype filters

2011 20th European Conference on Circuit Theory and Design (ECCTD), 2011

ABSTRACT In this work an analytical design method for two types of 2D recursive filters is propos... more ABSTRACT In this work an analytical design method for two types of 2D recursive filters is proposed. The design starts from a given analog prototype filter with specified parameters and a transfer function which is factorized into elementary (biquad) functions. Applying adequate frequency transformations to these functions, the desired 2D filters are directly obtained in a factorized form, preserving the selectivity properties of their 1D counterparts. The types of 2D filters approached here are narrow orientation-selective filters with a specified orientation and maximally-flat low-pass filters with circular symmetry. Design examples starting from imposed specifications and simulation results are provided.

Scalor based electrically controlled impedances

ISSCS 2011 - International Symposium on Signals, Circuits and Systems, 2011

ABSTRACT Several results regarding the CMOS implementation of electrically controlled impedances ... more

Aspects concerning cycle-to-cycle jitter in a comparator based relaxation oscillator

CAS 2012 (International Semiconductor Conference), 2012

ABSTRACT In this paper several formulas for cycle-to-cycle jitter in a simple comparator based re... more

On nonlinear distortions analysis of CMOS circuits

2014 International Semiconductor Conference (CAS), 2014

The aim of this communication is to present some aspects regarding nonlinear distortion analysis ... more The aim of this communication is to present some aspects regarding nonlinear distortion analysis in circuits implemented in CMOS technology. Basically, the investigations envisage the shape of the total harmonic distortion (THD) dependence on the amplitude and frequency of the input signals. It is apparent that the above function gives interesting insight of the nonlinear behavior of the studied circuits and it is worth considering for further investigations.

Nonhomogeneous CNN’s and their use for texture classification

2008 9th Symposium on Neural Network Applications in Electrical Engineering, 2008

Considerations concerning the dynamics of nonhomogeneous CNNpsilas are made and the possibility o... more

On the design of a class of CNN's for ECG classification

2011 20th European Conference on Circuit Theory and Design (ECCTD), 2011

ABSTRACT The paper discusses the possibility of using the dynamics of a class of Cellular Neural ... more ABSTRACT The paper discusses the possibility of using the dynamics of a class of Cellular Neural Networks (CNN&#39;s) for electrocardiogram (ECG) signals classification. The main idea is that of segmentation and transformation of the temporal signal into a 1D spatial one which is further processed by means of a bank of linear spatial filters using a parallel architecture of CNN type. A major advantage of the proposed solution is the independence of the filters spatial frequency characteristics on the number of samples of the ECG pattern, which allows dealing very easily with the heart rate variability. The principle of the proposed architecture is briefly discussed and the design of a bank of spatial filters for ECG classification is presented. Transistor level simulation and considerations regarding the architecture reconfiguration are given as well.

OTA based piecewise linear conductance

2012 10th International Symposium on Electronics and Telecommunications, 2012

ABSTRACT In this paper we present an operational transconductance amplifier based structure that ... more

Spatio-temporal dynamics in Cellular Neural Networks with nonideal OTA's

CAS 2011 Proceedings (2011 International Semiconductor Conference), 2011

Page 1. 405 SPATIO-TEMPORAL DYNAMICS IN CELLULAR NEURAL NETWORKS WITH NONIDEAL OTA'S Liviu G... more

Nonideal behavior of a class of Cellular Neural Networks

CAS 2012 (International Semiconductor Conference), 2012

ABSTRACT The aim of this paper is to report several new results regarding the nonideal behavior o... more

Liviu Goraş

Uploads

Papers by Liviu Goraş

Log In