Robust Wavelet Stabilized ‘Footprints of Uncertainty’ for Fuzzy System Classifiers to Automatically Detect Sharp Waves in the EEG after Hypoxia Ischemia

• Published in: International journal of neural systems 2017-05, Vol.27 (3), p.1650051
• Main Authors: Abbasi, Hamid, Bennet, Laura, Gunn, Alistair J., Unsworth, Charles P.
• Format: Article
• Language: English
• Subjects: Animals; Biomarkers; Brain Ischemia - diagnostic imaging; Brain Ischemia - physiopathology; Classifiers; Cohort Studies; Disease Models, Animal; Electroencephalography; Electroencephalography - methods; Female; Fetal Hypoxia - diagnostic imaging; Fetal Hypoxia - physiopathology; Footprints; Fuzzy Logic; Fuzzy systems; Hypoxia; Ischemia; Machine Learning; Occlusion; Pregnancy; Prenatal Diagnosis - methods; Sensitivity and Specificity; Sheep; Sheep, Domestic; Uncertainty; Wavelet Analysis; Wavelet transforms
• ISSN: 0129-0657; 1793-6462
• DOI: 10.1142/S0129065716500519

Currently, there are no developed methods to detect sharp wave transients that exist in the latent phase after hypoxia-ischemia (HI) in the electroencephalogram (EEG) in order to determine if these micro-scale transients are potential biomarkers of HI. A major issue with sharp waves in the HI-EEG is that they possess a large variability in their sharp wave profile making it difficult to build a compact ‘footprint of uncertainty’ (FOU) required for ideal performance of a Type-2 fuzzy logic system (FLS) classifier. In this paper, we develop a novel computational EEG analysis method to robustly detect sharp waves using over 30 h of post occlusion HI-EEG from an equivalent, in utero, preterm fetal sheep model cohort. We demonstrate that initial wavelet transform (WT) of the sharp waves stabilizes the variation in their profile and thus permits a highly compact FOU to be built, hence, optimizing the performance of a Type-2 FLS. We demonstrate that this method leads to higher overall performance of 9 4 % ± 1 for the clinical 6 4 Hz sampled EEG and 9 7 % ± 1 for the high resolution 1 0 2 4 Hz sampled EEG that is improved upon over conventional standard wavelet 6 7 % ± 5 and 8 2 % ± 3 , respectively, and fuzzy approaches 8 8 % ± 2 and 9 0 % ± 3 , respectively, when performed in isolation.