Automatic High Frequency Oscillations Detection Research Articles

Interactive interface for spatio-temporal mapping of epileptic human brain using characteristics of high frequency oscillations (HFOs)

Spontaneous High Frequency Oscillations (HFOs) have been considered as emerging specific biomarkers of the epileptogenic region. As a first issue, a significant difference in the implementation of automatic HFOs detection methods can sometimes occur between researchers. In addition, clinicians are not even particularly familiar with the concept of signal and image processing, and programming skills. To overcome these limitations, we propose a plug-and-play interactive Graphical User Interface (GUI) that incorporates an amalgamation of six validated methods used for detecting and quantifying of HFOs events. As a second issue, the most automated HFOs detection methods to date have a high false detection rate and low specificity, ranging, in some cases up to 80% and below 37% respectively. Therefore, the eventual utilization of HFOs detection algorithms in clinical settings requires a checking step to save clinically relevant HFOs and remove spurious oscillations from the detection results. As a last issue addressed in the present study, the major previous HFOs studies have been limited only to the detection and classification of HFOs, but only a few studies have been conducted to efficiently follow the neural dynamics of epileptic focus by studying HFOs characteristics through different brain regions and clinical stages. Therefore, in our software, the brain mapping of HFOs characteristics is done based on the duration, the inter-duration, the average frequency, and the power of HFOs. The present developed software may be considered helpful for understanding the functional significance of HFOs and also to reduce the interaction gap between fundamental research and applied clinical practice related to HFOs.

Retracted: Automatic Detection of High-Frequency Oscillations Based on an End-to-End Bi-Branch Neural Network and Clinical Cross-Validation.

[This retracts the article DOI: 10.1155/2021/7532241.].

Automatic Detection of High-Frequency Oscillations With Neuromorphic Spiking Neural Networks.

Interictal high-frequency oscillations (HFO) detected in electroencephalography recordings have been proposed as biomarkers of epileptogenesis, seizure propensity, disease severity, and treatment response. Automatic HFO detectors typically analyze the data offline using complex time-consuming algorithms, which limits their clinical application. Neuromorphic circuits offer the possibility of building compact and low-power processing systems that can analyze data on-line and in real time. In this review, we describe a fully automated detection pipeline for HFO that uses, for the first time, spiking neural networks and neuromorphic technology. We demonstrated that our HFO detection pipeline can be applied to recordings from different modalities (intracranial electroencephalography, electrocorticography, and scalp electroencephalography) and validated its operation in a custom-designed neuromorphic processor. Our HFO detection approach resulted in high accuracy and specificity in the prediction of seizure outcome in patients implanted with intracranial electroencephalography and electrocorticography, and in the prediction of epilepsy severity in patients recorded with scalp electroencephalography. Our research provides a further step toward the real-time detection of HFO using compact and low-power neuromorphic devices. The real-time detection of HFO in the operation room may improve the seizure outcome of epilepsy surgery, while the use of our neuromorphic processor for non-invasive therapy monitoring might allow for more effective medication strategies to achieve seizure control. Therefore, this work has the potential to improve the quality of life in patients with epilepsy by improving epilepsy diagnostics and treatment.

An Automatic HFO Detection Method Combining Visual Inspection Features with Multi-Domain Features.

As an important promising biomarker, high frequency oscillations (HFOs) can be used to track epileptic activity and localize epileptogenic zones. However, visual marking of HFOs from a large amount of intracranial electroencephalogram (iEEG) data requires a great deal of time and effort from researchers, and is also very dependent on visual features and easily influenced by subjective factors. Therefore, we proposed an automatic epileptic HFO detection method based on visual features and non-intuitive multi-domain features. To eliminate the interference of continuous oscillatory activity in detected sporadic short HFO events, the iEEG signals adjacent to the detected events were set as the neighboring environmental range while the number of oscillations and the peak-valley differences were calculated as the environmental reference features. The proposed method was developed as a MatLab-based HFO detector to automatically detect HFOs in multi-channel, long-distance iEEG signals. The performance of our detector was evaluated on iEEG recordings from epileptic mice and patients with intractable epilepsy. More than 90% of the HFO events detected by this method were confirmed by experts, while the average missed-detection rate was < 10%. Compared with recent related research, the proposed method achieved a synchronous improvement of sensitivity and specificity, and a balance between low false-alarm rate and high detection rate. Detection results demonstrated that the proposed method performs well in sensitivity, specificity, and precision. As an auxiliary tool, our detector can greatly improve the efficiency of clinical experts in inspecting HFO events during the diagnosis and treatment of epilepsy.

A spiking neural network (SNN) for detecting high frequency oscillations (HFOs) in the intraoperative ECoG

To achieve seizure freedom, epilepsy surgery requires the complete resection of the epileptogenic brain tissue. In intraoperative electrocorticography (ECoG) recordings, high frequency oscillations (HFOs) generated by epileptogenic tissue can be used to tailor the resection margin. However, automatic detection of HFOs in real-time remains an open challenge. Here we present a spiking neural network (SNN) for automatic HFO detection that is optimally suited for neuromorphic hardware implementation. We trained the SNN to detect HFO signals measured from intraoperative ECoG on-line, using an independently labeled dataset (58 min, 16 recordings). We targeted the detection of HFOs in the fast ripple frequency range (250-500 Hz) and compared the network results with the labeled HFO data. We endowed the SNN with a novel artifact rejection mechanism to suppress sharp transients and demonstrate its effectiveness on the ECoG dataset. The HFO rates (median 6.6 HFO/min in pre-resection recordings) detected by this SNN are comparable to those published in the dataset (Spearman’s rho = 0.81). The postsurgical seizure outcome was “predicted” with 100% (CI [63 100%]) accuracy for all 8 patients. These results provide a further step towards the construction of a real-time portable battery-operated HFO detection system that can be used during epilepsy surgery to guide the resection of the epileptogenic zone.

Automatic Detection of High-Frequency Oscillations Based on an End-to-End Bi-Branch Neural Network and Clinical Cross-Validation.

Accurate identification of high-frequency oscillation (HFO) is an important prerequisite for precise localization of epileptic foci and good prognosis of drug-refractory epilepsy. Exploring a high-performance automatic detection method for HFOs can effectively help clinicians reduce the error rate and reduce manpower. Due to the limited analysis perspective and simple model design, it is difficult to meet the requirements of clinical application by the existing methods. Therefore, an end-to-end bi-branch fusion model is proposed to automatically detect HFOs. With the filtered band-pass signal (signal branch) and time-frequency image (TFpic branch) as the input of the model, two backbone networks for deep feature extraction are established, respectively. Specifically, a hybrid model based on ResNet1d and long short-term memory (LSTM) is designed for signal branch, which can focus on both the features in time and space dimension, while a ResNet2d with a Convolutional Block Attention Module (CBAM) is constructed for TFpic branch, by which more attention is paid to useful information of TF images. Then the outputs of two branches are fused to realize end-to-end automatic identification of HFOs. Our method is verified on 5 patients with intractable epilepsy. In intravalidation, the proposed method obtained high sensitivity of 94.62%, specificity of 92.7%, and F1-score of 93.33%, and in cross-validation, our method achieved high sensitivity of 92.00%, specificity of 88.26%, and F1-score of 89.11% on average. The results show that the proposed method outperforms the existing detection paradigms of either single signal or single time-frequency diagram strategy. In addition, the average kappa coefficient of visual analysis and automatic detection results is 0.795. The method shows strong generalization ability and high degree of consistency with the gold standard meanwhile. Therefore, it has great potential to be a clinical assistant tool.

Automatic detection of HFOs based on singular value decomposition and improved fuzzy c-means clustering for localization of seizure onset zones

This paper devises a new detector based on singular value decomposition (SVD) and improved fuzzy c-means (FCM) clustering for automatically detecting high-frequency oscillations (HFOs) that are used for localizing seizure onset zones (SOZs) in epilepsy. First, HFO candidates (HFOCs) are obtained by the root mean square method. Next, a time-frequency analysis method is applied to eliminate spikes from HFOCs, which consists of the Stockwell transform, SVD combined with the k-medoids clustering algorithm, Stockwell inverse transform, and threshold method. Then, four kinds of distinctive features, i.e. mean singular values, line lengths, power ratios and spectral centroid of the rest of HFOCs, are extracted and augmented as feature vectors. These vectors are used as the input of the improved FCM clustering algorithm optimized by the simulated annealing algorithm combined with the genetic algorithm. Finally, the localization of SOZs is accomplished based on the concentrations of the detected HFOs. The superiority of the devised detector over other five existing ones is demonstrated by comparing their localization performance.

Localization of the Epileptogenic Zone Using High Frequency Oscillations.

For patients with drug-resistant focal epilepsy, surgery is the therapy of choice in order to achieve seizure freedom. Epilepsy surgery foremost requires the identification of the epileptogenic zone (EZ), defined as the brain area indispensable for seizure generation. The current gold standard for identification of the EZ is the seizure-onset zone (SOZ). The fact, however that surgical outcomes are unfavorable in 40–50% of well-selected patients, suggests that the SOZ is a suboptimal biomarker of the EZ, and that new biomarkers resulting in better postsurgical outcomes are needed. Research of recent years suggested that high-frequency oscillations (HFOs) are a promising biomarker of the EZ, with a potential to improve surgical success in patients with drug-resistant epilepsy without the need to record seizures. Nonetheless, in order to establish HFOs as a clinical biomarker, the following issues need to be addressed. First, evidence on HFOs as a clinically relevant biomarker stems predominantly from retrospective assessments with visual marking, leading to problems of reproducibility and reliability. Prospective assessments of the use of HFOs for surgery planning using automatic detection of HFOs are needed in order to determine their clinical value. Second, disentangling physiologic from pathologic HFOs is still an unsolved issue. Considering the appearance and the topographic location of presumed physiologic HFOs could be immanent for the interpretation of HFO findings in a clinical context. Third, recording HFOs non-invasively via scalp electroencephalography (EEG) and magnetoencephalography (MEG) is highly desirable, as it would provide us with the possibility to translate the use of HFOs to the scalp in a large number of patients. This article reviews the literature regarding these three issues. The first part of the article focuses on the clinical value of invasively recorded HFOs in localizing the EZ, the detection of HFOs, as well as their separation from physiologic HFOs. The second part of the article focuses on the current state of the literature regarding non-invasively recorded HFOs with emphasis on findings and technical considerations regarding their localization.

A Simple Statistical Method for the Automatic Detection of Ripples in Human Intracranial EEG.

High frequency oscillations (HFOs) are a promising biomarker of epileptic tissue, but detection of these electrographic events remains a challenge. Automatic detectors show encouraging results, but they typically require optimization of multiple parameters, which is a barrier to good performance and broad applicability. We therefore propose a new automatic HFO detection algorithm, focusing on simplicity and ease of implementation. It requires tuning of only an amplitude threshold, which can be determined by an iterative process or directly calculated from statistics of the rectified filtered data (i.e. mean plus standard deviation). The iterative approach uses an estimate of the amplitude probability distribution of the background activity to calculate the optimum threshold for identification of transient high amplitude events. We tested both the iterative and non-iterative approaches using a dataset of visually marked HFOs, and we compared the performance to a commonly used detector based on the root-mean-square. When the threshold was optimized for individual channels via ROC curve, all three methods were comparable. The iterative detector achieved a sensitivity of 99.6%, false positive rate (FPR) of 1.1%, and false detection rate (FDR) of 37.3%. However, in an eight-fold cross-validation test, the iterative method had better sensitivity than the other two methods (80.0% compared to 64.4 and 65.8%), with FPR and FDR of 1.3, and 49.4%, respectively. The simplicity of this algorithm, with only a single parameter, will enable consistent application of automatic detection across research centers and recording modalities, and it may therefore be a powerful tool for the assessment and localization of epileptic activity.

Exploring the time–frequency content of high frequency oscillations for automated identification of seizure onset zone in epilepsy

Objective. High frequency oscillations (HFOs) in intracranial electroencephalography (iEEG) recordings are considered as promising clinical biomarkers of epileptogenic regions in the brain. The aim of this study is to improve and automatize the detection of HFOs by exploring the time–frequency content of iEEG and to investigate the seizure onset zone (SOZ) detection accuracy during the sleep, awake and pre-ictal states in patients with epilepsy, for the purpose of assisting the localization of SOZ in clinical practice. Approach. Ten-minute iEEG segments were defined during different states in eight patients with refractory epilepsy. A three-stage algorithm was implemented to detect HFOs in these segments. First, an amplitude based initial detection threshold was used to generate a large pool of HFO candidates. Then distinguishing features were extracted from the time and time–frequency domain of the raw iEEG and used with a Gaussian mixture model clustering to isolate HFO events from other activities. The spatial distribution of HFO clusters was correlated with the seizure onset channels identified by neurologists in seven patient with good surgical outcome. Main results. The overlapping rates of localized channels and seizure onset locations were high in all states. The best result was obtained using the iEEG data during sleep, achieving a sensitivity of 81%, and a specificity of 96%. The channels with maximum number of HFOs identified epileptogenic areas where the seizures occurred more frequently. Significance. The current study was conducted using iEEG data collected in realistic clinical conditions without channel pre-exclusion. HFOs were investigated with novel features extracted from the entire frequency band, and were correlated with SOZ in different states. The results indicate that automatic HFO detection with unsupervised clustering methods exploring the time–frequency content of raw iEEG can be efficiently used to identify the epileptogenic zone with an accurate and efficient manner.

53. Methods of high frequency oscillations detection: Advantages and disadvantages

Background High frequency oscillations (HFOs) represent new electrographic marker of epileptogenic tissue and they are considered as a surrogate marker of seizure onset and epileptogenic zones. HFOs are recorded mainly in intracranial recordings. Visual analysis of HFOs in long-term recordings is extremely difficult due to the low signal-to-noise ratio of HFOs. Successful integration of HFOs into presurgical evaluation requires development of reliable methods of automatic HFO detection and quantification. We aimed to examine performance of three new HFO detecting algorithms and compared their performance with published detectors. Methods We implemented three published detectors which utilize RMS, line length or Hilbert transform approach to detect HFOs. We have developed additional three types of detectors which utilize short time energy estimation, Hilbert envelope and Bayesian evidence. All HFO detecting algorithms were applied to gold standard datasets and their performance quantified. Results Line length and Hilbert detectors detected the highest number of HFOs. The lowest number of the detections was achieved by RMS and energy estimating detectors. According to the results, the detectors can be divided into two groups. One group is characterized by high sensitivity. These algorithms detect nearly all the labeled HFOs events, but suffer from the high false positive detection rate. Second group of detectors have high positive prediction value but lower sensitivity. Our Hilbert envelope detector demonstrated the best performance of all evaluated detectors. Conclusions To improve the performance of detectors with high sensitivity will require to develop additional post-processing steps to remove the majority of false detections. Meanwhile detectors with low sensitivity will detect only high-amplitude HFOs. Future selection of the most appropriate algorithm for HFO detection in intracranial recordings will require detail understanding of the clinical significance of low-amplitude HFOs and major sources of false positive detections. Supported by Grants from IGA NT11460, NT13357, NT14489, GACR 14-02634S and Neuron Fund (NFKJ 001/2012).

Automatic 80–250 Hz “ripple” high frequency oscillation detection in invasive subdural grid and strip recordings in epilepsy by a radial basis function neural network

ObjectiveRecent studies give evidence that high frequency oscillations (HFOs) in the range between 80Hz and 500Hz in invasive recordings of epilepsy patients have the potential to serve as reliable markers of epileptogenicity. This study presents an algorithm for automatic HFO detection. MethodsThe presented HFO detector uses a radial basis function neural network. Input features of the detector were energy, line length and instantaneous frequency. Visual marked “ripple” HFOs (80–250Hz) of 3 patients were used to train the neural network, and a further 8 patients served for the detector evaluation. ResultsDetector sensitivity and specificity were 49.1% and 36.3%. The linear and rank correlation between visual and automatic marked “ripple” HFO counts over the channels were significant for all recordings. A reference detector based on the line length achieved a sensitivity of 35.4% and a specificity of 46.8%. ConclusionsAutomatic detections corresponded only partly to visual markings for single events but the relative distribution of brain regions displaying “ripple” HFO activity is reflected by the automated system. SignificanceThe detector allows the automatic evaluation of brain areas with high HFO frequency, which is of high relevance for the demarcation of the epileptogenic zone.

A comparison between detectors of high frequency oscillations

Objective High frequency oscillations (HFOs) are a biomarker of epileptogenicity. Visual marking of HFOs is highly time-consuming and inevitably subjective, making automatic detection necessary. We compare four existing detectors on the same dataset. Methods HFOs and baselines were identified by experienced reviewers in intracerebral EEGs from 20 patients. A new feature of our detector to deal with channels where baseline cannot be found is presented. The original and an optimal configuration are implemented. Receiver operator curves, false discovery rate, and channel ranking are used to evaluate performance. Results All detectors improve performance with the optimal configuration. Our detector had higher sensitivity, lower false positives than the others, and similar false detections. The main difference in performance was in very active channels. Conclusions Each detector was developed for different recordings and with different aims. Our detector performed better in this dataset, but was developed on data similar to the test data. Moreover, optimizing on a particular data type improves performance in any detector. Significance Automatic HFO detection is crucial to propel their clinical use as biomarkers of epileptogenic tissue. Comparing detectors on a single dataset is important to analyze their performance and to emphasize the issues involved in validation.