作者: G. Higgins , B. McGinley , S. Faul , R. P. McEvoy , M. Glavin
DOI: 10.1109/TITB.2012.2222426
关键词:
摘要: This paper examines the effects of compression on electroencephalogram (EEG) signals, in context automated detection epileptic seizures. Specifically, it use lossy EEG signals order to reduce amount data which has be transmitted or stored, while having as little impact possible information signal relevant diagnosing Two popular methods, JPEG2000 and SPIHT, were used. A range levels was selected for both algorithms compress with varying degrees loss. applied database epileptiform provided by University Freiburg, Germany. The real-time analysis event seizure system used place a trained clinician scoring reconstructed data. Results demonstrate that factor up 120:1 can achieved, minimal loss performance measured area under receiver operating characteristic curve system.