Fabienne Porée

ECG removal in preterm EEG combining empirical mode decomposition and adaptive filtering

In neonatal electroencephalography (EEG) heart activity is a major source of artifacts which can lead to misleading results in automated analysis if they are not properly eliminated. In this work we propose a combination of empirical mode decomposition (EMD) and adaptive filtering (AF) to cancel electrocardiogram (ECG) noise in a simplified EEG montage for preterm infants. The introduction of EMD prior to AF allows to selectively remove ECG preserving at maximum the original characteristics of EEG. Cleaned signals improved up to 17% the correlation coefficient with original datasets in comparison with signals denoised solely with AF.

Performance analysis of Hurst exponent estimators using surrogate-data and fractional lognormal noise models: Application to breathing signals from preterm infants

The use of the Hurst exponent (H) to quantify the fractal characteristics of biological signals and its potential to detect abnormalities has aroused, recently, the interest of many researchers. Numerous techniques to estimate H are described in the literature, yet the choice of the most performing one is not straightforward. In this paper, we proposed some tests using artificial signals from experimental data and stochastic models to evaluate the robustness of three estimation techniques. Different surrogate-data tests, including a novel method to parametrize the degree of correlation in experimental signals with H (Hurst-adjusted surrogates), were first carried out. Then, simulated signals with prescribed H were obtained from fractional Gaussian noise modified properly to follow the lognormal laws observed in empirical data. The tests were applied to examine detrended fluctuation analysis (DFA), discrete wavelet transform and least squares based on standard deviation (LSSD) methods in the particular case of inter-breath interval signals from preterm infants. Simulations showed that none of the estimators were robust for every breathing pattern (regular, erratic and periodic) and should not be applied blindly without performing the preliminary tests proposed here. The LSSD technique was the most precise in general, but DFA was more robust with highly spiked patterns.

Detection of Levodopa Induced Dyskinesia in Parkinson's Disease patients based on activity classification

In this paper, we present an activity classification-based algorithm for the automatic detection of Levodopa Induced Dyskinesia in Parkinsons Disease (PD) patients. Two PD patients experiencing motor fluctuations related to chronic Levodopa therapy performed a protocol of simple daily life activities on at least two different occasions. A Random Forest classifier was able to classify the performed activities by the patients with an overall accuracy of 86%. Based on the detected activity, a K Nearest Neighbor classifier detected the presence of dyskinesia with accuracy ranging from 75% to 88%.

Performance analysis of hurst's exponent estimators in higly immature breathing patterns of preterm infants

We analyzed the performance of five common estimators of the Hurst parameter (H) in simulated data, i.e. surrogate and synthetic through a fractional Gaussian noise model. Inter-breath signals of 30 preterm infants were used to generate surrogates as well as the synthetic data, which was produced according 3 typical patterns: erratic (EB), periodic (PB) and regular breathing (RB). The discrete wavelet transform was the most efficient for PB, with a concordance correlation coefficient (CCC) of 0.92. The iterative fGn-based estimator was the most efficient for EB and RB, with a CCC of 0.92 and 0.97 respectively, and performed better in the surrogates test with an estimation error of 0.008.

Multi-feature classifiers for burst detection in single EEG channels from preterm infants

OBJECTIVE The study of electroencephalographic (EEG) bursts in preterm infants provides valuable information about maturation or prognostication after perinatal asphyxia. Over the last two decades, a number of works proposed algorithms to automatically detect EEG bursts in preterm infants, but they were designed for populations under 35 weeks of post menstrual age (PMA). However, as the brain activity evolves rapidly during postnatal life, these solutions might be under-performing with increasing PMA. In this work we focused on preterm infants reaching term ages (PMA  ⩾36 weeks) using multi-feature classification on a single EEG channel. APPROACH Five EEG burst detectors relying on different machine learning approaches were compared: logistic regression (LR), linear discriminant analysis (LDA), k-nearest neighbors (kNN), support vector machines (SVM) and thresholding (Th). Classifiers were trained by visually labeled EEG recordings from 14 very preterm infants (born after 28 weeks of gestation) with 36-41 weeks PMA. MAIN RESULTS The most performing classifiers reached about 95% accuracy (kNN, SVM and LR) whereas Th obtained 84%. Compared to human-automatic agreements, LR provided the highest scores (Cohens kappa  =  0.71) using only three EEG features. Applying this classifier in an unlabeled database of 21 infants  ⩾36 weeks PMA, we found that long EEG bursts and short inter-burst periods are characteristic of infants with the highest PMA and weights. SIGNIFICANCE In view of these results, LR-based burst detection could be a suitable tool to study maturation in monitoring or portable devices using a single EEG channel.