Classifier combination supported by the sleep-wake cycle improves EEG seizure prediction performance.

OBJECTIVE: Seizure prediction is a promising solution to improve the quality of life for drug-resistant patients, which concerns nearly 30% of patients with epilepsy. The present study aimed to ascertain the impact of incorporating sleep-wake information in seizure prediction. METHODS: We developed five patient-specific prediction approaches that use vigilance state information differently: i) using it as an input feature, ii) building a pool of two classifiers, each with different weights to sleep/wake training samples, iii) building a pool of two classifiers, each with only sleep/wake samples, iv) changing the alarm-threshold concerning each sleep/wake state, and v) adjusting the alarm-threshold after a sleep-wake transition. We compared these approaches with a control method that did not integrate sleep-wake information. Our models were tested with data (43 seizures and 482 hours) acquired during presurgical monitoring of 17 patients from the EPILEPSIAE database. As EPILEPSIAE does not contain vigilance state annotations, we developed a sleep-wake classifier using 33 patients diagnosed with nocturnal frontal lobe epilepsy from the CAP Sleep database. RESULTS: Although different patients may require different strategies, our best approach, the pool of weighted predictors, obtained 65% of patients performing above chance level with a surrogate analysis (against 41% in the control method). CONCLUSION: The inclusion of vigilance state information improves seizure prediction. Higher results and testing with longterm recordings from daily-life conditions are necessary to ensure clinical acceptance. SIGNIFICANCE: As automated sleep-wake detection is possible, it would be feasible to incorporate these algorithms into future devices for seizure prediction.

Osteoporosis screening using machine learning and electromagnetic waves.

Osteoporosis is a disease characterized by impairment of bone microarchitecture that causes high socioeconomic impacts in the world because of fractures and hospitalizations. Although dual-energy X-ray absorptiometry (DXA) is the gold standard for diagnosing the disease, access to DXA in developing countries is still limited due to its high cost, being present only in specialized hospitals. In this paper, we analyze the performance of Osseus, a low-cost portable device based on electromagnetic waves that measures the attenuation of the signal that crosses the medial phalanx of a patient's middle finger and was developed for osteoporosis screening. The analysis is carried out by predicting changes in bone mineral density using Osseus measurements and additional common risk factors used as input features to a set of supervised classification models, while the results from DXA are taken as target (real) values during the training of the machine learning algorithms. The dataset consisted of 505 patients who underwent osteoporosis screening with both devices (DXA and Osseus), of whom 21.8% were healthy and 78.2% had low bone mineral density or osteoporosis. A cross-validation with k-fold = 5 was considered in model training, while 20% of the whole dataset was used for testing. The obtained performance of the best model (Random Forest) presented a sensitivity of 0.853, a specificity of 0.879, and an F1 of 0.859. Since the Random Forest (RF) algorithm allows some interpretability of its results (through the impurity check), we were able to identify the most important variables in the classification of osteoporosis. The results showed that the most important variables were age, body mass index, and the signal attenuation provided by Osseus. The RF model, when used together with Osseus measurements, is effective in screening patients and facilitates the early diagnosis of osteoporosis. The main advantages of such early screening are the reduction of costs associated with exams, surgeries, treatments, and hospitalizations, as well as improved quality of life for patients.

The goal of explaining black boxes in EEG seizure prediction is not to explain models' decisions.

Many state-of-the-art methods for seizure prediction, using the electroencephalogram, are based on machine learning models that are black boxes, weakening the trust of clinicians in them for high-risk decisions. Seizure prediction concerns a multidimensional time-series problem that performs continuous sliding window analysis and classification. In this work, we make a critical review of which explanations increase trust in models' decisions for predicting seizures. We developed three machine learning methodologies to explore their explainability potential. These contain different levels of model transparency: a logistic regression, an ensemble of 15 support vector machines, and an ensemble of three convolutional neural networks. For each methodology, we evaluated quasi-prospectively the performance in 40 patients (testing data comprised 2055 hours and 104 seizures). We selected patients with good and poor performance to explain the models' decisions. Then, with grounded theory, we evaluated how these explanations helped specialists (data scientists and clinicians working in epilepsy) to understand the obtained model dynamics. We obtained four lessons for better communication between data scientists and clinicians. We found that the goal of explainability is not to explain the system's decisions but to improve the system itself. Model transparency is not the most significant factor in explaining a model decision for seizure prediction. Even when using intuitive and state-of-the-art features, it is hard to understand brain dynamics and their relationship with the developed models. We achieve an increase in understanding by developing, in parallel, several systems that explicitly deal with signal dynamics changes that help develop a complete problem formulation.

Unsupervised EEG preictal interval identification in patients with drug-resistant epilepsy.

Typical seizure prediction models aim at discriminating interictal brain activity from pre-seizure electrographic patterns. Given the lack of a preictal clinical definition, a fixed interval is widely used to develop these models. Recent studies reporting preictal interval selection among a range of fixed intervals show inter- and intra-patient preictal interval variability, possibly reflecting the heterogeneity of the seizure generation process. Obtaining accurate labels of the preictal interval can be used to train supervised prediction models and, hence, avoid setting a fixed preictal interval for all seizures within the same patient. Unsupervised learning methods hold great promise for exploring preictal alterations on a seizure-specific scale. Multivariate and univariate linear and nonlinear features were extracted from scalp electroencephalography (EEG) signals collected from 41 patients with drug-resistant epilepsy undergoing presurgical monitoring. Nonlinear dimensionality reduction was performed for each group of features and each of the 226 seizures. We applied different clustering methods in searching for preictal clusters located until 2 h before the seizure onset. We identified preictal patterns in 90% of patients and 51% of the visually inspected seizures. The preictal clusters manifested a seizure-specific profile with varying duration (22.9 ± 21.0 min) and starting time before seizure onset (47.6 ± 27.3 min). Searching for preictal patterns on the EEG trace using unsupervised methods showed that it is possible to identify seizure-specific preictal signatures for some patients and some seizures within the same patient.

Development of a Cyclic Voltammetry-Based Method for the Detection of Antigens and Antibodies as a Novel Strategy for Syphilis Diagnosis.

The improvement of laboratory diagnosis is a critical step for the reduction of syphilis cases around the world. In this paper, we present the development of an impedance-based method for detecting T. pallidum antigens and antibodies as an auxiliary tool for syphilis laboratory diagnosis. We evaluate the voltammetric signal obtained after incubation in carbon or gold nanoparticle-modified carbon electrodes in the presence or absence of Poly-L-Lysine. Our results indicate that the signal obtained from the electrodes was sufficient to distinguish between infected and non-infected samples immediately (T0') or 15 min (T15') after incubation, indicating its potential use as a point-of-care method as a screening strategy.

On the clinical acceptance of black-box systems for EEG seizure prediction.

Seizure prediction may be the solution for epileptic patients whose drugs and surgery do not control seizures. Despite 46 years of research, few devices/systems underwent clinical trials and/or are commercialized, where the most recent state-of-the-art approaches, as neural networks models, are not used to their full potential. The latter demonstrates the existence of social barriers to new methodologies due to data bias, patient safety, and legislation compliance. In the form of literature review, we performed a qualitative study to analyze the seizure prediction ecosystem to find these social barriers. With the Grounded Theory, we draw hypotheses from data, while with the Actor-Network Theory we considered that technology shapes social configurations and interests, being fundamental in healthcare. We obtained a social network that describes the ecosystem and propose research guidelines aiming at clinical acceptance. Our most relevant conclusion is the need for model explainability, but not necessarily intrinsically interpretable models, for the case of seizure prediction. Accordingly, we argue that it is possible to develop robust prediction models, including black-box systems to some extent, while avoiding data bias, ensuring patient safety, and still complying with legislation, if they can deliver human- comprehensible explanations. Due to skepticism and patient safety reasons, many authors advocate the use of transparent models which may limit their performance and potential. Our study highlights a possible path, by using model explainability, on how to overcome these barriers while allowing the use of more computationally robust models.

Heart rate variability analysis for the identification of the preictal interval in patients with drug-resistant epilepsy.

Electrocardiogram (ECG) recordings, lasting hours before epileptic seizures, have been studied in the search for evidence of the existence of a preictal interval that follows a normal ECG trace and precedes the seizure's clinical manifestation. The preictal interval has not yet been clinically parametrized. Furthermore, the duration of this interval varies for seizures both among patients and from the same patient. In this study, we performed a heart rate variability (HRV) analysis to investigate the discriminative power of the features of HRV in the identification of the preictal interval. HRV information extracted from the linear time and frequency domains as well as from nonlinear dynamics were analysed. We inspected data from 238 temporal lobe seizures recorded from 41 patients with drug-resistant epilepsy from the EPILEPSIAE database. Unsupervised methods were applied to the HRV feature dataset, thus leading to a new perspective in preictal interval characterization. Distinguishable preictal behaviour was exhibited by 41% of the seizures and 90% of the patients. Half of the preictal intervals were identified in the 40 min before seizure onset. The results demonstrate the potential of applying clustering methods to HRV features to deepen the current understanding of the preictal state.

A personalized and evolutionary algorithm for interpretable EEG epilepsy seizure prediction.

Seizure prediction may improve the quality of life of patients suffering from drug-resistant epilepsy, which accounts for about 30% of the total epileptic patients. The pre-ictal period determination, characterized by a transitional stage between normal brain activity and seizure, is a critical step. Past approaches failed to attain real-world applicability due to lack of generalization capacity. More recently, deep learning techniques may outperform traditional classifiers and handle time dependencies. However, despite the existing efforts for providing interpretable insights, clinicians may not be willing to make high-stake decisions based on them. Furthermore, a disadvantageous aspect of the more usual seizure prediction pipeline is its modularity and significant independence between stages. An alternative could be the construction of a search algorithm that, while considering pipeline stages' synergy, fine-tunes the selection of a reduced set of features that are widely used in the literature and computationally efficient. With extracranial recordings from 19 patients suffering from temporal-lobe seizures, we developed a patient-specific evolutionary optimization strategy, aiming to generate the optimal set of features for seizure prediction with a logistic regression classifier, which was tested prospectively in a total of 49 seizures and 710 h of continuous recording and performed above chance for 32% of patients, using a surrogate predictor. These results demonstrate the hypothesis of pre-ictal period identification without the loss of interpretability, which may help understanding brain dynamics leading to seizures and improve prediction algorithms.

Method for estimating average grey-level's measurement uncertainty from ultrasound images for non-invasive estimation of temperature in different tissue types.

The objective of this work is to assess, on metrological basis, the average grey-levels (AVGL) calculated from B-Mode images for estimating temperature variations non-invasively in different kinds of tissues. Thermal medicine includes several thermal therapies, being hyperthermia the most noted and well known. Recently, efforts have been made to understand the benefits of ultrasound hyperthermia at mild temperature levels, i.e., between 39 °C and 41 °C. Moreover, the best practices on ultrasound bio-effects research have been encouraged by recommending that temperature rise in the region of interest should be measured even when a thermal mechanism is not being tested. In this work, the average grey-levels (AVGL) calculated from B-Mode images were assessed for non-invasive temperature estimation in a porcine tissue sample containing two different tissue types, fat and muscle, with temperature varying from 35 °C to 41 °C. The sample was continuously imaged with an ultrasound scanner, and simultaneously the temperature was measured. The achieved results were assessed under the light of the measurement uncertainty in order to allow comparability among different ultrasound thermometry methods. The highest expanded uncertainty of estimating temperature variation using AVGL was determined as 0.68 °C.

A Realistic Seizure Prediction Study Based on Multiclass SVM.

A patient-specific algorithm, for epileptic seizure prediction, based on multiclass support-vector machines (SVM) and using multi-channel high-dimensional feature sets, is presented. The feature sets, combined with multiclass classification and post-processing schemes aim at the generation of alarms and reduced influence of false positives. This study considers 216 patients from the European Epilepsy Database, and includes 185 patients with scalp EEG recordings and 31 with intracranial data. The strategy was tested over a total of 16,729.80[Formula: see text]h of inter-ictal data, including 1206 seizures. We found an overall sensitivity of 38.47% and a false positive rate per hour of 0.20. The performance of the method achieved statistical significance in 24 patients (11% of the patients). Despite the encouraging results previously reported in specific datasets, the prospective demonstration on long-term EEG recording has been limited. Our study presents a prospective analysis of a large heterogeneous, multicentric dataset. The statistical framework based on conservative assumptions, reflects a realistic approach compared to constrained datasets, and/or in-sample evaluations. The improvement of these results, with the definition of an appropriate set of features able to improve the distinction between the pre-ictal and nonpre-ictal states, hence minimizing the effect of confounding variables, remains a key aspect.

On the proper selection of preictal period for seizure prediction.

Supervised machine learning-based seizure prediction methods consider preictal period as an important prerequisite parameter during training. However, the exact length of the preictal state is unclear and varies from seizure to seizure. We propose a novel statistical approach for proper selection of the preictal period, which can also be considered either as a measure of predictability of a seizure or as the prediction capability of an understudy feature. The optimal preictal periods (OPPs) obtained from the training samples can be used for building a more accurate classifier model. The proposed method uses amplitude distribution histograms of features extracted from electroencephalogram (EEG) recordings. To evaluate this method, we extract spectral power features in different frequency bands from monopolar and space-differential EEG signals of 18 patients suffering from pharmacoresistant epilepsy. Furthermore, comparisons among monopolar channels with space-differential channels, as well as intracranial EEG (iEEG) and surface EEG (sEEG) signals, indicate that while monopolar signals perform better in iEEG recordings, no significant difference is noticeable in sEEG recordings.

Early Seizure Detection Using Neuronal Potential Similarity: A Generalized Low-Complexity and Robust Measure.

A novel approach using neuronal potential similarity (NPS) of two intracranial electroencephalogram (iEEG) electrodes placed over the foci is proposed for automated early seizure detection in patients with refractory partial epilepsy. The NPS measure is obtained from the spectral analysis of space-differential iEEG signals. Ratio between the NPS values obtained from two specific frequency bands is then investigated as a robust generalized measure, and reveals invaluable information about seizure initiation trends. A threshold-based classifier is subsequently applied on the proposed measure to generate alarms. The performance of the method was evaluated using cross-validation on a large clinical dataset, involving 183 seizure onsets in 1785 h of long-term continuous iEEG recordings of 11 patients. On average, the results show a high sensitivity of 86.9% (159 out of 183), a very low false detection rate of 1.4 per day, and a mean detection latency of 13.1 s from electrographic seizure onsets, while in average preceding clinical onsets by 6.3 s. These high performance results, specifically the short detection latency, coupled with the very low computational cost of the proposed method make it adequate for using in implantable closed-loop seizure suppression systems.

Epileptic Seizure Prediction based on Ratio and Differential Linear Univariate Features.

Bivariate features, obtained from multichannel electroencephalogram recordings, quantify the relation between different brain regions. Studies based on bivariate features have shown optimistic results for tackling epileptic seizure prediction problem in patients suffering from refractory epilepsy. A new bivariate approach using univariate features is proposed here. Differences and ratios of 22 linear univariate features were calculated using pairwise combination of 6 electroencephalograms channels, to create 330 differential, and 330 relative features. The feature subsets were classified using support vector machines separately, as one of the two classes of preictal and nonpreictal. Furthermore, minimum Redundancy Maximum Relevance feature reduction method is employed to improve the predictions and reduce the number of false alarms. The studies were carried out on features obtained from 10 patients. For reduced subset of 30 features and using differential approach, the seizures were on average predicted in 60.9% of the cases (28 out of 46 in 737.9 h of test data), with a low false prediction rate of 0.11 h(-1). Results of bivariate approaches were compared with those achieved from original linear univariate features, extracted from 6 channels. The advantage of proposed bivariate features is the smaller number of false predictions in comparison to the original 22 univariate features. In addition, reduction in feature dimension could provide a less complex and the more cost-effective algorithm. Results indicate that applying machine learning methods on a multidimensional feature space resulting from relative/differential pairwise combination of 22 univariate features could predict seizure onsets with high performance.

Early detection of epilepsy seizures based on a weightless neural network.

This work introduces a new methodology for the early detection of epileptic seizure based on the WiSARD weightless neural network model and a new approach in terms of preprocessing the electroencephalogram (EEG) data. WiSARD has, among other advantages, the capacity of perform the training phase in a very fast way. This speed in training is due to the fact that WiSARD's neurons work like Random Access Memories (RAM) addressed by input patterns. Promising results were obtained in the anticipation of seizure onsets in four representative patients from the European Database on Epilepsy (EPILEPSIAE). The proposed seizure early detection WNN architecture was explored by varying the detection anticipation (δ) in the 2 to 30 seconds interval, and by adopting 2 and 3 seconds as the width of the Sliding Observation Window (SOW) input. While in the most challenging patient (A) one obtained accuracies from 99.57% (δ=2s; SOW=3s) to 72.56% (δ=30s; SOW=2s), patient D seizures could be detected in the 99.77% (δ=2s; SOW=2s) to 99.93% (δ=30s; SOW=3s) accuracy interval.

Epileptic seizure prediction using relative spectral power features.

OBJECTIVE: Prediction of epileptic seizures can improve the living conditions for refractory epilepsy patients. We aimed to improve sensitivity and specificity of prediction methods, and to reduce the number of false alarms. METHODS: Relative combinations of sub-band spectral powers of electroencephalogram (EEG) recordings across all possible channel pairs were utilized for tracking gradual changes preceding seizures. By using a specifically developed feature selection method, a set of best candidate features were fed to support vector machines in order to discriminate cerebral state as preictal or non-preictal. RESULTS: Proposed algorithm was evaluated on continuous long-term multichannel scalp and invasive recordings (183 seizures, 3565 h). The best results demonstrated a sensitivity of 75.8% (66 out of 87 seizures) and a false prediction rate of 0.1h(-1). Performance was validated statistically, and was superior to that of analytical random predictor. CONCLUSION: Applying machine learning methods on a reduced subset of proposed features could predict seizure onsets with high performance. SIGNIFICANCE: Our method was evaluated on long-term continuous recordings of overall about 5 months, contrary to majority of previous studies using short-term fragmented data. It is of very low computational cost, while providing acceptable levels of alarm sensitivity and specificity.

Robust and low complexity algorithms for seizure detection.

This paper presents two low complexity and yet robust methods for automated seizure detection using a set of 2 intracranial Electroencephalogram (iEEG) recordings. Most current seizure detection methods suffer from high number of false alarms, even when designed to be subject-specific. In this study, the ratios of power between pairs of frequency bands are used as features to detect epileptic seizures. For comparison, these features are calculated from monopolar and bipolar iEEG recordings. Optimal thresholds are individually determined and used for each feature. Alarms are generated when the measure passes the threshold. The detector was applied to long-term continuous invasive recordings from 5 patients with refractory partial epilepsy, containing 54 seizures in 780 hours. On average, the results revealed 88.9% sensitivity, a very low false detection rate of 0.041 per hour (h(-1)) and detection latency of 9.4 seconds.

Pre-ictal heart rate variability assessment of epileptic seizures by means of linear and non-linear analyses.

OBJECTIVE: The purpose of the present study was to analyze the effects of epilepsy on the autonomic control of the heart in pre-ictal phase in order to find an algorithm of early detection of seizure onset. METHODS: Overall 133 epileptic seizures were analyzed from 12 patients with epilepsy (seven males and five females; mean age 43.91 years, SD: 10.16) participated in this study. Single lead electrocardiogram recordings of epileptic patients were compiled. 240, 90-30, 30-10 and 5 minutes heart rate variability (HRV) signals of preseizure were chosen for analysis of heart rate. As HRV signals are non-stationary, a set of time and frequency domain features (Mean HR, Triangular Index, LF, HF, LF/HF) and nonlinear parameters (SD1, SD2 and SD2/SD1 indices derived from Poincare plots) extracted from HRV is analyzed. Statistical analysis was performed using paired sample t-test for comparisons of the segments and differences between pre-ictal segments were evaluated by Tukey tests. RESULTS: There was slight tachycardia in segments near the seizure (30 minutes before: 85.3517 bpm, 5 minutes before: 119.3630.82 bpm, p=0.0207) which significantly differ from baseline in segments far from seizure (240 minutes before: 66.5211.7 bpm). Also there was significant increase in LF/HF ratio (30 minutes before: 1.10.22, 5 minutes before: 2.120.5, p=0.0332) and SD2/SD1 ratio (30 minutes before: 1.20.15, 5 minutes before: 2.030.55, p=0.0431) when compared to segments far from the seizure (240 minutes before: 0.780.24 and 0.780.14) respectively. Although there was about decrease of triangular index in segments near the seizure the percentage of decrease was not comparable to segments far from the seizure. CONCLUSION: Significant changes of HRV parameters in pre-ictal (5 minutes before the seizure) are obviously higher in comparison to interictal baseline. Pre-ictal significant changes of HRV suggesting that this time can be considered as prediction time for designing an algorithm of early detection of seizure onset based on HRV.

Seizure prediction with bipolar spectral power features using Adaboost and SVM classifiers.

This paper presents the results of our study on finding a lower complexity and yet a robust seizure prediction method using intracranial electroencephalogram (iEEG) recordings. We compare two classifiers: a low-complexity Adaboost and the more complex support vector machine (SVM). Adaboost is a linear classier using decision stumps, and SVM uses a nonlinear Gaussian kernel. Bipolar and/or time-differential spectral power features of different sub-bands are extracted from the iEEG signal. Adaboost is used to simultaneously classify as well as rank the features. Eliminating the low discriminating features reduces computational complexity and power consumption. The top features selected by Adaboost were also used as a feature set for SVM classification. The outputs of classifiers are regularized by applying a moving-average window and a threshold is used to generate alarms. The proposed methods were applied on 8 invasive recordings selected from the EPILEPSIAE database, the European database of EEG seizure recordings. Doublecross validation is used by separating data sets for training and optimization from testing. The key conclusion is that Adaboost performs slightly better than SVM using a reduced feature set on average with significantly less complexity resulting in a sensitivity of 77.1% (27 of 35 seizures in 873 h recordings) and a false alarm rate of 0.18 per hour.

Preprocessing effects of 22 linear univariate features on the performance of seizure prediction methods.

Combining multiple linear univariate features in one feature space and classifying the feature space using machine learning methods could predict epileptic seizures in patients suffering from refractory epilepsy. For each patient, a set of twenty-two linear univariate features were extracted from 6 electroencephalogram (EEG) signals to make a 132 dimensional feature space. Preprocessing and normalization methods of the features, which affect the output of the seizure prediction algorithm, were studied in terms of alarm sensitivity and false prediction rate (FPR). The problem of choosing an optimal preictal time was tackled using 4 distinct values of 10, 20, 30, and 40 min. The seizure prediction problem has traditionally been considered a two-class classification problem, which is also exercised here. These studies have been conducted on the features obtained from 10 patients. For each patient, 48 different combinations of methods are compared to find the best configuration. Normalization by dividing by the maximum and smoothing are found to be the best configuration in most of the patients. The results also indicate that applying machine learning methods on a multidimensional feature space of 22 univariate features predicted seizure onsets with high performance. On average, the seizures were predicted in 73.9% of the cases (34 out of 46 in 737.9h of test data), with a FPR of 0.15 h(-1).

EPILEPSIAE - a European epilepsy database.

With a worldwide prevalence of about 1%, epilepsy is one of the most common serious brain diseases with profound physical, psychological and, social consequences. Characteristic symptoms are seizures caused by abnormally synchronized neuronal activity that can lead to temporary impairments of motor functions, perception, speech, memory or, consciousness. The possibility to predict the occurrence of epileptic seizures by monitoring the electroencephalographic activity (EEG) is considered one of the most promising options to establish new therapeutic strategies for the considerable fraction of patients with currently insufficiently controlled seizures. Here, a database is presented which is part of an EU-funded project "EPILEPSIAE" aiming at the development of seizure prediction algorithms which can monitor the EEG for seizure precursors. High-quality, long-term continuous EEG data, enriched with clinical metadata, which so far have not been available, are managed in this database as a joint effort of epilepsy centers in Portugal (Coimbra), France (Paris) and Germany (Freiburg). The architecture and the underlying schema are here reported for this database. It was designed for an efficient organization, access and search of the data of 300 epilepsy patients, including high quality long-term EEG recordings, obtained with scalp and intracranial electrodes, as well as derived features and supplementary clinical and imaging data. The organization of this European database will allow for accessibility by a wide spectrum of research groups and may serve as a model for similar databases planned for the future.