Adaptive frequency decomposition of EEG with subsequent expert system analysis

• Published in: Computers in biology and medicine Vol. 31; no. 6; pp. 407 - 427
• Main Authors: Herrmann, C.S, Arnold, T, Visbeck, A, Hundemer, H.-P, Hopf, H.C
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01-11-2001; New York, NY Elsevier Science
• Subjects: Adolescent; Adult; Aged; Aged, 80 and over; Alpha Rhythm; Biological and medical sciences; Biometry; Brain Diseases - diagnosis; Brain Diseases - physiopathology; Computerized, statistical medical data processing and models in biomedicine; Delta Rhythm; Diagnosis, Computer-Assisted; EEG; Electrodiagnosis. Electric activity recording; Electroencephalography - statistics & numerical data; Expert system; Expert Systems; Frequency domain analysis; Fuzzy Logic; General aspects. Methods; Humans; Investigative techniques, diagnostic techniques (general aspects); Medical sciences; Middle Aged; Nervous system; Neurology; Pathology; Spectral analysis; Spectrum analysis; Theta Rhythm; Time domain analysis; Fuzzy logic; EEG; Expert system; Spectral analysis; Human; Evaluation; Alpha rhythm; Time frequency domain method; Artefact; Decomposition; Electroencephalography; Algorithm; Automatic processing; Electrodiagnosis; Biomedical data processing; Sensitivity; Frequency; Signal analysis; Biomedical engineering
• ISSN: 0010-4825; 1879-0534
• DOI: 10.1016/S0010-4825(01)00017-8

We present a hybrid system for automatic analysis of clinical routine EEG, comprising a spectral analysis and an expert system. EEG raw data are transformed into the time–frequency domain by the so-called adaptive frequency decomposition. The resulting frequency components are converted into pseudo-linguistic facts via fuzzification. Finally, an expert system applies symbolic rules formulated by the neurologist to evaluate the extracted EEG features. The system detects artefacts, describes alpha rhythm by frequency, amplitude, and stability and after artefact rejection detects pathologic slow activity. All results are displayed as linguistic terms, numerical values and maps of temporal extent, giving an overview about the clinical routine EEG.