Selection of optimum frequency bands for detection of epileptiform patterns

Healthcare Technology Letters

Manjari Tripathi¹  Poodipedi Sarat Chandra¹  Piyush Swami²  Manvir Bhatia³  Tapan K. Gandhi⁴  Bijaya K. Panigrahi⁴ 

[1] All India Institute of Medical Sciences;Centre for Biomedical Engineering, Indian Institute of Technology – Delhi;Fortis Escorts Hospital;Indian Institute of Technology – Delhi;

关键词: neurophysiology;    electroencephalography;    medical signal processing;    medical signal detection;    signal classification;    medical disorders;    optimum frequency bands;    epileptiform patterns;    automated seizure detection system;    electrophysiological recordings;    signal decomposition;    frequency bins;    seizure episodes;    classification process;    seizure classification rates;    state-of-the-art classification accuracy;    classification performance;    scalp electroencephalograms;   

DOI  :  10.1049/htl.2018.5051

来源: DOAJ

【 摘 要 】

The significant research effort in the domain of epilepsy has been directed toward the development of an automated seizure detection system. In their usage of the electrophysiological recordings, most of the proposals thus far have followed the conventional practise of employing all frequency bands following signal decomposition as input features for a classifier. Although seemingly powerful, this approach may prove counterproductive since some frequency bins may not carry relevant information about seizure episodes and may, instead, add noise to the classification process thus degrading performance. A key thesis of the work described here is that the selection of frequency subsets may enhance seizure classification rates. Additionally, the authors explore whether a conservative selection of frequency bins can reduce the amount of training data needed for achieving good classification performance. They have found compelling evidence that using spectral components with <25 Hz frequency in scalp electroencephalograms can yield state-of-the-art classification accuracy while reducing training data requirements to just a tenth of those employed by current approaches.

【 授权许可】

Unknown