scientific article; zbMATH DE number 1193394
Time-frequency representations describe the characteristics of time-varying signals by means of two-dimensional energy distributions in an excellent manner. Many examples in this book demonstrate the advantages of time-frequency transformations in different medical and biological applications. However, normally these methods do not automatically characterize signals with a set of meaningful parameters as is necessary for typical classification or discrimination tasks. For example, in our application, evoked potentials are used to discriminate children with attention deficit disorder with hyperactivity (ADDH) from a control group. To verify the clinical hypothesis of frontal brain lesions, the central nervous system is tested by auditory evoked potentials (AEPs). This activity is particularly useful because it is independent of an intact behavioral response system. However, the main problem of our study is to extract parameters from evoked potentials (EPs) which can be easily used to find group differences between normal and ADDH children. Frequency, EP peak parameters, i.e., amplitudes and latencies, are considered as relevant features. Besides this, specific parts of the signal represented by its sample points are used for the classification task. In any case additional selection procedures are required to reduce the number of features. This chapter gives a short introduction on how wavelet functions can be used for an optimal formal description of a time-varying signal. This description is achieved by only a small number of representative parameters. Particular attention is paid to the learning algorithm in order to get reliable parameter estimates. Furthermore, the derived procedure is applied to our clinical study concerning the discrimination of ADDH children by AEPs. The results are discussed in comparison with conventional heuristic strategies. For the entire collection see [Zbl 0889.00018].
- Classification of EEG signals using the wavelet transform.
- Wavelet analysis of big data contaminated by large noise in an fMRI study of neuroplasticity
- The influence of wavelets on multiscale analysis and parametrization of midlatency auditory evoked potentials
- Automated averaging of auditory evoked response waveforms using wavelet analysis
- scientific article; zbMATH DE number 5151509 (Why is no real title available?)
- Discrimination between two mental states (rest and motor image) using wavelet transform and neural networks
- scientific article; zbMATH DE number 833250 (Why is no real title available?)
- scientific article; zbMATH DE number 910693 (Why is no real title available?)
- Obtaining single stimulus evoked potentials with wavelet denoising
- Obtaining single stimulus evoked potentials with wavelet denoising
- Brain electrical activity analysis using wavelet-based informational tools
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3841870)