dc.contributor.author | Pattichis, Constantinos S. | en |
dc.contributor.author | Elia, Andreas G. | en |
dc.creator | Pattichis, Constantinos S. | en |
dc.creator | Elia, Andreas G. | en |
dc.date.accessioned | 2019-11-13T10:41:54Z | |
dc.date.available | 2019-11-13T10:41:54Z | |
dc.date.issued | 1999 | |
dc.identifier.issn | 1350-4533 | |
dc.identifier.uri | http://gnosis.library.ucy.ac.cy/handle/7/54799 | |
dc.description.abstract | Quantitative electromyographic signal analysis in the time domain for motor unit action potential (MUAP) classification and disease identification has been well documented over recent years. Considerable work has also been carried out in the frequency domain using classical power spectrum analysis techniques. Although MUAP autoregressive (AR) spectral analysis has been suggested as a diagnostic tool by a number of studies, it has not been thoroughly investigated yet. This work investigates the application of AR modeling and cepstral analysis for the diagnostic assessment of MUAPs recorded from normal (NOR) subjects and subjects suffering with motor neuron disease (MND) and myopathy (MYO). The following feature sets were extracted from the MUAP signal: (i) time domain measures, (ii) AR spectral measures, (iii) AR coefficients, and (iv) cepstral coefficients. Discriminative analysis of the individual features was carried out using the univariate and multiple covariance analysis methods. Both methods showed that: (i) the duration measure is the best discriminative feature among the time domain parameters, and (ii) the median frequency is the best discriminative feature among the AR spectral measures. Furthermore, the classification performance of the above four feature sets was investigated for three classes (NOR, MND and MYO) using artificial neural networks. Results showed that the highest diagnostic yield was obtained with the time domain measures followed by the cepstral coefficients, the AR spectral parameters, and the AR coefficients. In conclusion, MUAP autoregressive and cepstral analyses combined with time domain analysis provide useful information in the assessment of myopathology. Copyright (C) 1999 IPEM. | en |
dc.source | Medical Engineering and Physics | en |
dc.source.uri | https://www.scopus.com/inward/record.uri?eid=2-s2.0-0032744849&doi=10.1016%2fS1350-4533%2899%2900072-7&partnerID=40&md5=e667b303f3e21446d426a9f3f674f25e | |
dc.subject | model | en |
dc.subject | article | en |
dc.subject | Neural networks | en |
dc.subject | human | en |
dc.subject | Humans | en |
dc.subject | priority journal | en |
dc.subject | Time Factors | en |
dc.subject | data analysis | en |
dc.subject | classification | en |
dc.subject | Reference Values | en |
dc.subject | artificial neural network | en |
dc.subject | Neural Networks (Computer) | en |
dc.subject | Muscle, Skeletal | en |
dc.subject | Motor Neurons | en |
dc.subject | power spectrum | en |
dc.subject | Muscular Diseases | en |
dc.subject | Electromyography | en |
dc.subject | Action Potentials | en |
dc.subject | Models, Neurological | en |
dc.subject | action potential | en |
dc.subject | neuromuscular disease | en |
dc.subject | motor unit potential | en |
dc.subject | Autoregressive analysis | en |
dc.subject | Cepstral analysis | en |
dc.subject | Fourier Analysis | en |
dc.subject | Motor Neuron Disease | en |
dc.subject | Motor Unit Action Potential | en |
dc.title | Autoregressive and cepstral analyses of motor unit action potentials | en |
dc.type | info:eu-repo/semantics/article | |
dc.identifier.doi | 10.1016/S1350-4533(99)00072-7 | |
dc.description.volume | 21 | |
dc.description.issue | 6-7 | |
dc.description.startingpage | 405 | |
dc.description.endingpage | 419 | |
dc.author.faculty | 002 Σχολή Θετικών και Εφαρμοσμένων Επιστημών / Faculty of Pure and Applied Sciences | |
dc.author.department | Τμήμα Πληροφορικής / Department of Computer Science | |
dc.type.uhtype | Article | en |
dc.description.notes | <p>Cited By :51</p> | en |
dc.source.abbreviation | Med.Eng.Phys. | en |
dc.contributor.orcid | Pattichis, Constantinos S. [0000-0003-1271-8151] | |
dc.gnosis.orcid | 0000-0003-1271-8151 | |