1,721,086 research outputs found
Arrhythmia Detection by Data Fusion of ECG Scalograms and Phasograms
Full text linkThe automatic detection of arrhythmia is of primary importance due to the huge number of victims caused worldwide by cardiovascular diseases. To this aim, several deep learning approaches have been recently proposed to automatically classify heartbeats in a small number of classes. Most of these approaches use convolutional neural networks (CNNs), exploiting some bi-dimensional representation of the ECG signal, such as spectrograms, scalograms, or similar. However, by adopting such representations, state-of-the-art approaches usually rely on the magnitude information alone, while the important phase information is often neglected. Motivated by these considerations, the focus of this paper is aimed at investigating the effect of fusing the magnitude and phase of the continuous wavelet transform (CWT), known as the scalogram and phasogram, respectively. Scalograms and phasograms are fused in a simple CNN-based architecture by using several fusion strategies, which fuse the information in the input layer, some intermediate layers, or in the output layer. Numerical results evaluated on the PhysioNet MIT-BIH Arrhythmia database show the effectiveness of the proposed ideas. Although a simple architecture is used, their competitiveness is high compared to other state-of-the-art approaches, by obtaining an overall accuracy of about 98.5% and sensitivity and specificity of 98.5% and 95.6%, respectively
Flexible methods for blind separation of complex signals
Full text linkOne of the main matter in Blind Source Separation (BSS) performed with a neural network approach is the choice of the nonlinear activation function (AF). In fact if the shape of the activation function is chosen as the cumulative density function (c.d.f.) of the original source the problem is solved.
For this scope in this thesis a flexible approach is introduced and the shape of the
activation functions is changed during the learning process using the so-called “spline functions”.
The problem is complicated in the case of separation of complex sources where there is the problem of the dichotomy between analyticity and boundedness of the complex activation functions. The problem is solved introducing the “splitting function” model as activation function. The “splitting function” is a couple of “spline function” which wind off the real and the imaginary part of the complex activation function, each of one depending from the real and imaginary variable.
A more realistic model is the “generalized splitting function”, which is formed by a couple of two bi-dimensional functions (surfaces), one for the real and one for
the imaginary part of the complex function, each depending by both the real and imaginary part of the complex variable.
Unfortunately the linear environment is unrealistic in many practical applications.
In this way there is the need of extending BSS problem in the nonlinear environment: in this case both the activation function than the nonlinear distorting function are realized by the “splitting function” made of “spline function”.
The complex and instantaneous separation in linear and nonlinear environment allow us to perform a complex-valued extension of the well-known INFOMAX algorithm in several practical situations, such as convolutive mixtures, fMRI signal analysis and bandpass signal transmission.
In addition advanced characteristics on the proposed approach are introduced and deeply described. First of all it is shows as splines are universal nonlinear functions for BSS problem: they are able to perform separation in anyway. Then it is analyzed as the “splitting solution” allows the algorithm to obtain a phase recovery:
usually there is a phase ambiguity. Finally a Cramér-Rao lower bound for ICA is discussed.
Several experimental results, tested by different objective indexes, show the
effectiveness of the proposed approaches
Introduzione all'audio real-time: Basi teoriche e prime applicazioni
No full textQuesto libro, nato dall'esperienza maturata dagli autori, si pone l'obiettivo di introdurre il mondo delle tecnologie, degli strumenti e delle soluzioni tecniche per la realizzazione di applicazioni audio in tempo reale. Lo strumento ideale per ottenere un risultato di questo tipo è costituito dalla libreria software open-source PortAudio, orientata appunto all'audio streaming, con cui è possibile sviluppare applicativi anche piuttosto d'effetto, con una conoscenza basilare nell'ambito della programmazione. L'approccio pragmatico, orientato cioè all'esposizione della specifica tecnologia, sarà arricchito da nozioni di Digital Signal Processing estremamente utili nell'implementazione di algoritmi per l'elaborazione del segnale audio. La trattazione sarà arricchita altresì da alcuni esempi applicativi: verranno presentati esempi di applicazione di streaming e di alcuni algoritmi di elaborazione del segnale, come ad esempio alcuni tra i più noti effetti audio. Verrà infine riservata una sezione alla gestione dei diversi formati di file audio e una panoramica sul protocollo MIDI.
La trattazione dei vari argomenti avviene in maniera graduale. Si parte dalle nozioni elementari sui fondamenti dei sistemi audio e poi, attraverso esempi pratici, il lettore arriva passo dopo passo alla comprensione di argomenti non elementari. Il codice di tutti gli esempi proposti nel testo è reso disponibile online per il download
An Empirical and Semi Blind Algorithm for Resolving Overlapped Peaks in Chromatography: Application to the Analysis of Environmental Samples
No full textIn this paper we describe a new algorithm for enhancing the resolution (i.e. deconvolution) of chromatographic peaks strongly overlapped among them, often observed in the analysis of complex environmental samples. The main characteristic of this algorithm does not require an “a priori” knowledge of the statistic moments (i.e. width, retention time and height) of the peak to be quantified so that it is considered an empirical and semi-blind (ESB) algorithm. The efficiency of the ESB algorithm has been verified for synthetic overlapped peaks, for the gas chromatographic (GC) analysis of hydrocarbons from marine sediments and for the planar chromatographic analysis (TLC) of carbohydrates in marine organic matter samples. In all the examined cases, standard errors lower than 25% and quadratic (R2) correlation coefficients higher than 0.85 were obtained, showing the comparability of the ESB algorithm with other deconvolution methods
Flexible estimation of probability and cumulative density functions
No full textA novel, simple and effective algorithm for the estimation of the probability density function and cumulative density function is presented. The algorithm is based on an information maximisation approach. The nonlinear function involved in the algorithm is adaptively modi?ed during learning and is implemented by using a spline function
Fog-supported delay-constrained energy-saving live migration of VMs over multiPath TCP/IP 5G connections
Full text linkThe incoming era of the fifth-generation fog computing-supported radio access networks (shortly, 5G FOGRANs) aims at exploiting computing/networking resource virtualization, in order to augment the limited resources of wireless devices through the seamless live migration of virtual machines (VMs) toward nearby fog data centers. For this purpose, the bandwidths of the multiple wireless network interface cards of the wireless devices may be aggregated under the control of the emerging MultiPathTCP (MPTCP) protocol. However, due to the fading and mobility-induced phenomena, the energy consumptions of the current state-of-the-art VM migration techniques may still offset their expected benefits. Motivated by these considerations, in this paper, we analytically characterize and implement in software and numerically test the optimal minimum-energy settable-complexity bandwidth manager (SCBM) for the live migration of VMs over 5G FOGRAN MPTCP connections. The key features of the proposed SCBM are that: 1) its implementation complexity is settable on-line on the basis of the target energy consumption versus implementation complexity tradeoff; 2) it minimizes the network energy consumed by the wireless device for sustaining the migration process under hard constraints on the tolerated migration times and downtimes; and 3) by leveraging a suitably designed adaptive mechanism, it is capable to quickly react to (possibly, unpredicted) fading and/or mobility-induced abrupt changes of the wireless environment without requiring forecasting. The actual effectiveness of the proposed SCBM is supported by extensive energy versus delay performance comparisons that cover: 1) a number of heterogeneous 3G/4G/WiFi FOGRAN scenarios; 2) synthetic and real-world workloads; and, 3) MPTCP and wireless connections
Twinned Residual Auto-Encoder (TRAE)-A new DL architecture for denoising super-resolution and task-aware feature learning from COVID-19 CT images
No full textThe detection of the COronaVIrus Disease 2019 (COVID-19) from Computed Tomography (CT) scans has become a very important task in modern medical diagnosis. Unfortunately, typical resolutions of state-of-the-art CT scans are still not adequate for reliable and accurate automatic detection of COVID-19 disease. Motivated by this consideration, in this paper, we propose a novel architecture that jointly affords the Single-Image Super-Resolution (SISR) and the reliable classification problems from Low Resolution (LR) and noisy CT scans. Specifically, the proposed architecture is based on a couple of Twinned Residual Auto-Encoders (TRAE), which exploits the feature vectors and the SR images recovered by a Master AE for performing transfer learning and then improves the training of a "twinned" Follower AE. In addition, we also develop a Task-Aware (TA) version of the basic TRAE architecture, namely the TA-TRAE, which further utilizes the set of feature vectors generated by the Follower AE for the joint training of an additional auxiliary classifier, so to perform automated medical diagnosis on the basis of the available LR input images without human support. Experimental results and comparisons with a number of state-of-the-art CNN/GAN/CycleGAN benchmark SISR architectures, performed by considering ×2 , ×4 , and ×8 super-resolution (i.e., upscaling) factors, support the effectiveness of the proposed TRAE/TA-TRAE architectures. In particular, the detection accuracy attained by the proposed architectures outperforms the corresponding ones of the implemented CNN, GAN and CycleGAN baselines up to 9.0%, 6.5%, and 6.0% at upscaling factors as high as ×8
- …
