| Autor: |
Alegre-Cortés J; Neuroprosthetics and Visual Rehabilitation Research Unit, Bioengineering Institute, Miguel Hernández University, Alicante, Spain., Soto-Sánchez C; Neuroprosthetics and Visual Rehabilitation Research Unit, Bioengineering Institute, Miguel Hernández University, Alicante, Spain.; Biomedical Research Networking Center in Bioengineering, Biomaterials and Nanomedicine, Madrid, Spain.; Biotechnology Department, University of Alicante, Alicante, Spain., Albarracín AL; Laboratorio de Medios e Interfases, Departamento de Bioingeniería, Facultad de Ciencias Exactas y Tecnología, Universidad Nacional de Tucumán, Tucumán, Argentina.; Departamento de Bioingeniería, Instituto Superior de Investigaciones Biológicas, Consejo Nacional de Investigaciones Científicas y Técnicas, Tucumán, Argentina., Farfán FD; Laboratorio de Medios e Interfases, Departamento de Bioingeniería, Facultad de Ciencias Exactas y Tecnología, Universidad Nacional de Tucumán, Tucumán, Argentina.; Departamento de Bioingeniería, Instituto Superior de Investigaciones Biológicas, Consejo Nacional de Investigaciones Científicas y Técnicas, Tucumán, Argentina., Val-Calvo M; Departamento de Electrónica, Tecnología de Computadoras, Universidad Politécnica de Cartagena, Cartagena, Spain., Ferrandez JM; Departamento de Electrónica, Tecnología de Computadoras, Universidad Politécnica de Cartagena, Cartagena, Spain., Fernandez E; Neuroprosthetics and Visual Rehabilitation Research Unit, Bioengineering Institute, Miguel Hernández University, Alicante, Spain.; Biomedical Research Networking Center in Bioengineering, Biomaterials and Nanomedicine, Madrid, Spain. |
| Abstrakt: |
Machine learning and artificial intelligence have strong roots on principles of neural computation. Some examples are the structure of the first perceptron, inspired in the retina, neuroprosthetics based on ganglion cell recordings or Hopfield networks. In addition, machine learning provides a powerful set of tools to analyze neural data, which has already proved its efficacy in so distant fields of research as speech recognition, behavioral states classification, or LFP recordings. However, despite the huge technological advances in neural data reduction of dimensionality, pattern selection, and clustering during the last years, there has not been a proportional development of the analytical tools used for Time-Frequency (T-F) analysis in neuroscience. Bearing this in mind, we introduce the convenience of using non-linear, non-stationary tools, EMD algorithms in particular, for the transformation of the oscillatory neural data (EEG, EMG, spike oscillations…) into the T-F domain prior to its analysis with machine learning tools. We support that to achieve meaningful conclusions, the transformed data we analyze has to be as faithful as possible to the original recording, so that the transformations forced into the data due to restrictions in the T-F computation are not extended to the results of the machine learning analysis. Moreover, bioinspired computation such as brain-machine interface may be enriched from a more precise definition of neuronal coding where non-linearities of the neuronal dynamics are considered. |
