Andrey Temko

EEG-based neonatal seizure detection with Support Vector Machines.

Objective The study presents a multi-channel patient-independent neonatal seizure detection system based on the Support Vector Machine (SVM) classifier. Methods A machine learning algorithm (SVM) is used as a classifier to discriminate between seizure and non-seizure EEG epochs. Two post-processing steps are proposed to increase both the temporal precision and the robustness of the system. The resulting system is validated on a large clinical dataset of 267 h of EEG data from 17 full-term newborns with seizures. Results The performance of the system using event-based metrics is reported. The system showed the best up-to-date performance of a neonatal seizure detection system. The system was able to achieve an average good detection rate of ∼89% with one false seizure detection per hour, ∼96% with two false detections per hour, or ∼100% with four false detections per hour. An analysis of errors revealed sources of misclassification in terms of both missed seizures and false detections. Conclusions The results obtained with the proposed SVM-based seizure detection system allow for its practical application in neonatal intensive care units. Significance The proposed SVM-based seizure detection system can greatly assist clinical staff, in a neonatal intensive care unit, to interpret the EEG. The system allows control of the final decision by choosing different confidence levels which makes it flexible for clinical needs. The obtained results may provide a reference for future seizure detection systems.

CLEAR evaluation of acoustic event detection and classification systems

In this paper, we present the results of the Acoustic Event Detection (AED) and Classification (AEC) evaluations carried out in February 2006 by the three participant partners from the CHIL project. The primary evaluation task was AED of the testing portions of the isolated sound databases and seminar recordings produced in CHIL. Additionally, a secondary AEC evaluation task was designed using only the isolated sound databases. The set of meeting-room acoustic event classes and the metrics were agreed by the three partners and ELDA was in charge of the scoring task. In this paper, the various systems for the tasks of AED and AEC and their results are presented.

Classification of acoustic events using SVM-based clustering schemes

Acoustic events produced in controlled environments may carry information useful for perceptually aware interfaces. In this paper we focus on the problem of classifying 16 types of meeting-room acoustic events. First of all, we have defined the events and gathered a sound database. Then, several classifiers based on support vector machines (SVM) are developed using confusion matrix based clustering schemes to deal with the multi-class problem. Also, several sets of acoustic features are defined and used in the classification tests. In the experiments, the developed SVM-based classifiers are compared with an already reported binary tree scheme and with their correlative Gaussian mixture model (GMM) classifiers. The best results are obtained with a tree SVM-based classifier that may use a different feature set at each node. With it, a 31.5% relative average error reduction is obtained with respect to the best result from a conventional binary tree scheme.

Performance assessment for EEG-based neonatal seizure detectors

Objective This study discusses an appropriate framework to measure system performance for the task of neonatal seizure detection using EEG. The framework is used to present an extended overview of a multi-channel patient-independent neonatal seizure detection system based on the Support Vector Machine (SVM) classifier. Methods The appropriate framework for performance assessment of neonatal seizure detectors is discussed in terms of metrics, experimental setups, and testing protocols. The neonatal seizure detection system is evaluated in this framework. Several epoch-based and event-based metrics are calculated and curves of performance are reported. A new metric to measure the average duration of a false detection is proposed to accompany the event-based metrics. A machine learning algorithm (SVM) is used as a classifier to discriminate between seizure and non-seizure EEG epochs. Two post-processing steps proposed to increase temporal precision and robustness of the system are investigated and their influence on various metrics is shown. The resulting system is validated on a large clinical dataset of 267 h. Results In this paper, it is shown how a complete set of metrics and a specific testing protocol are necessary to extensively describe neonatal seizure detection systems, objectively assess their performance and enable comparison with existing alternatives. The developed system currently represents the best published performance to date with an ROC area of 96.3%. The sensitivity and specificity were ∼90% at the equal error rate point. The system was able to achieve an average good detection rate of ∼89% at a cost of 1 false detection per hour with an average false detection duration of 2.7 min. Conclusions It is shown that to accurately assess the performance of EEG-based neonatal seizure detectors and to facilitate comparison with existing alternatives, several metrics should be reported and a specific testing protocol should be followed. It is also shown that reporting only event-based metrics can be misleading as they do not always reflect the true performance of the system. Significance This is the first study to present a thorough method for performance assessment of EEG-based seizure detection systems. The evaluated SVM-based seizure detection system can greatly assist clinical staff, in a neonatal intensive care unit, to interpret the EEG.

Acoustic event detection in meeting-room environments

Acoustic event detection (AED) aims at determining the identity of sounds and their temporal position in the signals that are captured by one or several microphones. The AED problem has been recently proposed for meeting-room or class-room environments, where a specific set of meaningful sounds has been defined, and several evaluations have been carried out (within the international CLEAR evaluation campaigns). This paper reports some work in AED done by the authors in that framework, and particularly presents the extension to the difficult problem of detecting overlapped sounds. Actually, temporal overlaps accounted for more than 70% of errors in the real-world interactive seminar recordings used in CLEAR 2007 evaluations. An attempt to deal with that problem at the level of models using our SVM-based AED system is reported in the paper. The proposed two-step system noticeably outperforms the baseline system for both an artificially generated database and a real seminar recording database. The databases and metrics developed for the CLEAR 2007 evaluations are also described. Finally, a real-time AED system implemented in the UPCs smart-room using several microphones is reported, along with a GUI-based demo that includes also the output of an acoustic source localization system.

Fuzzy integral based information fusion for classification of highly confusable non-speech sounds

Acoustic event classification may help to describe acoustic scenes and contribute to improve the robustness of speech technologies. In this work, fusion of different information sources with the fuzzy integral (FI), and the associated fuzzy measure (FM), are applied to the problem of classifying a small set of highly confusable human non-speech sounds. As FI is a meaningful formalism for combining classifier outputs that can capture interactions among the various sources of information, it shows in our experiments a significantly better performance than that of any single classifier entering the FI fusion module. Actually, that FI decision-level fusion approach shows comparable results to the high-performing SVM feature-level fusion and thus it seems to be a good choice when feature-level fusion is not an option. We have also observed that the importance and the degree of interaction among the various feature types given by the FM can be used for feature selection, and gives a valuable insight into the problem.

EEG Signal Description with Spectral-Envelope-Based Speech Recognition Features for Detection of Neonatal Seizures

In this paper, features which are usually employed in automatic speech recognition (ASR) are used for the detection of seizures in newborn EEG. In particular, spectral envelope-based features, composed of spectral powers and their spectral derivatives are compared to the established feature set which has been previously developed for EEG analysis. The results indicate that the ASR features which model the spectral derivatives, either full-band or localized in frequency, yielded a performance improvement, in comparison to spectral-power-based features. Indeed it is shown here that they perform reasonably well in comparison with the conventional EEG feature set. The contribution of the ASR features was analyzed here using the support vector machines (SVM) recursive feature elimination technique. It is shown that the spectral derivative features consistently appear among the top-rank features. The study shows that the ASR features should be given a high priority when dealing with the description of the EEG signal.

EEG in the healthy term newborn within 12 hours of birth

OBJECTIVE To characterise and quantify the EEG during sleep in healthy newborns in the early newborn period. METHODS Continuous multi-channel video-EEG data was recorded for up to 2 hours in normal newborns within 12 hours of birth. The total amount of active (AS) and quiet sleep (QS) was calculated in the first hour of recording. The EEG signal was quantitatively analysed for symmetry and synchrony. Spectral edge frequency (SEF), spectral entropy (H) and relative delta power (delta(R)) were calculated for a ten-minute segment of AS and QS in each recording. Paired t-test and Wilcoxon rank sum test were used for data analysis. RESULTS Thirty normal newborn babies were studied, 10 within 6 hours of birth and 20 between 6 and 12 hours. All babies showed continuous symmetrical and synchronous EEG activity and well-developed sleep-wake cycling (SWC) with the median percentage of AS--48.5% and QS--36.6%. Quantitative EEG analysis of sleep epochs showed that SEF and H were significantly higher (p<0.0001) and delta(R) was significantly lower (p<0.0001) in AS than in QS. CONCLUSION The normal newborn EEG shows symmetrical and synchronous continuous activity and well-developed SWC as early as within the first 6 hours of birth. Quantitative analysis of the EEG in the early postnatal period reveals differences in SEF, H and delta(R) for AS and QS periods. SIGNIFICANCE These findings may have implications for quantitative analysis of the newborn EEG, including the EEG of babies with hypoxic ischaemic encephalopathy.

Enhanced SVM Training for Robust Speech Activity Detection

Speech activity detection (SAD) is a key objective in speech-related technologies. In this work, an enhanced version of the training stage of a SAD system based on a support vector machine (SVM) classifier is presented, and its performance is tested with the RT05 and RT06 evaluation tasks. A fast algorithm of data reduction based on proximal SVM has been developed and, furthermore, the specific characteristics of the metric used in the NIST SAD evaluation have been taken into account during training. Tested with the RT06 data, the resulting SVM SAD system has shown better scores than the best GMM-based system developed by the authors and submitted to the past RT06 evaluation.

Acoustic Event Detection: SVM-Based System and Evaluation Setup in CLEAR'07

In this paper, the Acoustic Event Detection (AED) system developed at the UPC is described, and its results in the CLEAR evaluations carried out in March 2007 are reported. The system uses a set of features composed of frequency-filtered band energies and perceptual features, and it is based on SVM classifiers and multi-microphone decision fusion. Also, the current evaluation setup and, in particular, the two new metrics used in this evaluation are presented.