scholarly journals Multiband entropy-based feature-extraction method for automatic identification of epileptic focus based on high-frequency components in interictal iEEG

2020 ◽  
Author(s):  
Most. Sheuli Akter ◽  
Md. Rabiul Islam ◽  
Yasushi Iimura ◽  
Hidenori Sugano ◽  
Kosuke Fukumori ◽  
...  
Keyword(s):  
Machine Learning ◽  
High Frequency ◽  
Statistical Tests ◽  
Machine Learning Techniques ◽  
Epileptic Focus ◽  
Automatic Identification ◽  
Surgical Removal ◽  
Linear Discriminant ◽  
Information Theoretic ◽  
Feature Extraction Method

Presurgical investigations for categorizing focal patterns are crucial, leading to localization and surgical removal of the epileptic focus. This paper presents a machine learning approach using information theoretic features extracted from high-frequency subbands to detect the epileptic focus from interictal intracranial electroencephalogram (iEEG). It is known that high-frequency subbands (>80 Hz) include important biomarkers such as high-frequency oscillations (HFOs) for identifying epileptic focus commonly referred to as the seizure on-set zone (SOZ). In this analysis, the multi-channel interictal iEEG signals were splitted into segments and each segment was decomposed into multiple high-frequency subbands. The different types of entropy were calculated for each of the subbands and the sparse linear discriminant analysis (sLDA) was applied to select the prominent entropy features. Due to the imbalance of SOZ and non-SOZ channels in iEEG data, the use of machine learning techniques is always tricky. To deal with the imbalanced learning problem, an adaptive synthetic oversampling approach (ADASYN) with radial basis function kernel-based SVM was used to detect the focal segments. Finally, the epileptic focus was identified based on detection of focal segments on SOZ and non-SOZ channels. Eight patients were examined to observe the efficiency of the automatic detector. The experimental results and statistical tests indicate that the proposed automatic detector can identify the epileptic focus accurately and efficiently.

ANOMALY DETECTION IN VESSEL TRACKING – A BAYESIAN NETWORKS (BNs) APPROACH

2021 ◽  
Vol 157 (A3) ◽  
Author(s):  
D Handayani ◽  
W Sediono ◽  
A Shah
Keyword(s):  
Machine Learning ◽  
Bayesian Networks ◽  
Expert Knowledge ◽  
Machine Learning Techniques ◽  
Automatic Identification ◽  
Identification System ◽  
International Convention ◽  
Learning Techniques ◽  
Reporting Data ◽  
Vessel Anomaly

The paper describes the supervised method approach to identifying vessel anomaly behaviour. The vessel anomaly behaviour is determined by learning from self-reporting maritime systems based on the Automatic Identification System (AIS). The AIS is a real world vessel reporting data system, which has been recently made compulsory by the International Convention for the Safety of Life and Sea (SOLAS) for vessels over 300 gross tons and most commercial vessels such as cargo ships, passenger vessels, tankers, etc. In this paper, we describe the use of Bayesian networks (BNs) approach to identify the behaviour of the vessel of interest. The BNs is a machine learning technique based on probabilistic theory that represents a set of random variables and their conditional independencies via directed acyclic graph (DAG). Previous studies showed that the BNs have important advantages compared to other machine learning techniques. Among them are that expert knowledge can be included in the BNs model, and that humans can understand and interpret the BNs model more readily. This work proves that the BNs technique is applicable to the identification of vessel anomaly behaviour.

Cognitive refractory state caused by spontaneous epileptic high-frequency oscillations in the human brain

2019 ◽  
Vol 11 (514) ◽  
pp. eaax7830 ◽  
Author(s):  
Su Liu ◽  
Josef Parvizi
Keyword(s):  
Cognitive Deficits ◽  
High Frequency ◽  
Medial Temporal Lobe ◽  
Memory Performance ◽  
Physiological Responses ◽  
Epileptic Activity ◽  
Epileptic Focus ◽  
Surgical Removal ◽  
High Frequency Oscillations ◽  
Refractory State

Epileptic brain tissue is often considered physiologically dysfunctional, and the optimal treatment of many patients with uncontrollable seizures involves surgical removal of the epileptic tissue. However, it is unclear to what extent the epileptic tissue is capable of generating physiological responses to cognitive stimuli and how cognitive deficits ensuing surgical resections can be determined using state-of-the-art computational methods. To address these unknowns, we recruited six patients with nonlesional epilepsies and identified the epileptic focus in each patient with intracranial electrophysiological monitoring. We measured spontaneous epileptic activity in the form of high-frequency oscillations (HFOs), recorded stimulus-locked physiological responses in the form of physiological high-frequency broadband activity, and explored the interaction of the two as well as their behavioral correlates. Across all patients, we found abundant normal physiological responses to relevant cognitive stimuli in the epileptic sites. However, these physiological responses were more likely to be “seized” (delayed or missed) when spontaneous HFOs occurred about 850 to 1050 ms before, until about 150 to 250 ms after, the onset of relevant cognitive stimuli. Furthermore, spontaneous HFOs in medial temporal lobe affected the subjects’ memory performance. Our findings suggest that nonlesional epileptic sites are capable of generating normal physiological responses and highlight a compelling mechanism for cognitive deficits in these patients. The results also offer clinicians a quantitative tool to differentiate pathological and physiological high-frequency activities in epileptic sites and to indirectly assess their possible cognitive reserve function and approximate the risk of resective surgery.

scholarly journals Study of Machine Learning Techniques for EEG Eye State Detection

Proceedings ◽  
2020 ◽  
Vol 54 (1) ◽  
pp. 53
Author(s):  
Francisco Laport ◽  
Paula M. Castro ◽  
Adriana Dapena ◽  
Francisco J. Vazquez-Araujo ◽  
Daniel Iglesia
Keyword(s):  
Machine Learning ◽  
Mental States ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
Discrete Wavelet ◽  
Linear Discriminant ◽  
State Identification ◽  
Brain Signals ◽  
State Classification ◽  
Learning Techniques

A comparison of different machine learning techniques for eye state identification through Electroencephalography (EEG) signals is presented in this paper. (1) Background: We extend our previous work by studying several techniques for the extraction of the features corresponding to the mental states of open and closed eyes and their subsequent classification; (2) Methods: A prototype developed by the authors is used to capture the brain signals. We consider the Discrete Fourier Transform (DFT) and the Discrete Wavelet Transform (DWT) for feature extraction; Linear Discriminant Analysis (LDA) and Support Vector Machine (SVM) for state classification; and Independent Component Analysis (ICA) for preprocessing the data; (3) Results: The results obtained from some subjects show the good performance of the proposed methods; and (4) Conclusion: The combination of several techniques allows us to obtain a high accuracy of eye identification.

scholarly journals A Classification Approach for Predicting COVID-19 Patient Survival Outcome with Machine Learning Techniques

2020 ◽  
Author(s):  
Abdulhameed Ado Osi ◽  
Hussaini Garba Dikko ◽  
Mannir Abdu ◽  
Auwalu Ibrahim ◽  
Lawan Adamu Isma'il ◽  
...  
Keyword(s):  
Machine Learning ◽  
Survival Outcome ◽  
Supervised Machine Learning ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
P Value ◽  
Kappa Index ◽  
Quality Health Care ◽  
Linear Discriminant ◽  
Learning Techniques

COVID-19 is an infectious disease discovered after the outbreak began in Wuhan, China, in December 2019. COVID-19 is still becoming an increasing global threat to public health. The virus has been escalated to many countries across the globe. This paper analyzed and compared the performance of three different supervised machine learning techniques; Linear Discriminant Analysis (LDA), Random Forest (RF), and Support Vector Machine (SVM) on COVID-19 dataset. The best level of accuracy between these three algorithms was determined by comparison of some metrics for assessing predictive performance such as accuracy, sensitivity, specificity, F-score, Kappa index, and ROC. From the analysis results, RF was found to be the best algorithm with 100% prediction accuracy in comparison with LDA and SVM with 95.2% and 90.9% respectively. Our analysis shows that out of these three classification models RF predicts COVID-19 patient's survival outcome with the highest accuracy. Chi-square test reveals that all the seven features except sex were significantly correlated with the COVID-19 patient's outcome (P-value < 0.005). Therefore, RF was recommended for COVID-19 patient outcome prediction that will help in early identification of possible sensitive cases for quick provision of quality health care, support and supervision.

Automatic Cognitive Load Classification Using High-Frequency Interaction Events

2013 ◽  
Vol 9 (3) ◽  
pp. 73-88 ◽  
Author(s):  
Tao Lin ◽  
Xiao Li ◽  
Zhiming Wu ◽  
Ningjiu Tang
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Cognitive Load ◽  
High Frequency ◽  
Evaluation Method ◽  
Back Propagation ◽  
Machine Learning Techniques ◽  
Selection Methods ◽  
Learning Techniques ◽  
Frequency Interaction

There is still a challenge of creating an evaluation method which can not only unobtrusively collect data without supplement equipment but also objectively, quantitatively and in real-time evaluate cognitive load of user based the data. The study explores the possibility of using the features extracted from high-frequency interaction events to evaluate cognitive load to respond to the challenge. Specifically, back-propagation neural networks, along with two feature selection methods (nBset and SFS), were used as the classifier and it was able to use a set of features to differentiate three cognitive load levels with an accuracy of 74.27%. The main contributions of the research are: (1) demonstrating the use of combining machine learning techniques and the HFI features in automatically evaluating cognitive load; (2) showing the potential of using the HFI features in discriminating different cognitive load when suitable classifier and features are adopted.

scholarly journals A Feature Extraction Method for P2P Botnet Detection Using Graphic Symmetry Concept

Symmetry ◽  
2019 ◽  
Vol 11 (3) ◽  
pp. 326 ◽  
Author(s):  
Zhixian Yang ◽  
Buhong Wang
Keyword(s):  
Machine Learning ◽  
Information Entropy ◽  
Feature Detection ◽  
Denial Of Service ◽  
Attack Detection ◽  
Machine Learning Techniques ◽  
Detection Accuracy ◽  
Feature Extraction Method ◽  
Denial Of Service Attack ◽  
Ddos Attack

A DDoS (Distributed Denial of Service) attack makes use of a botnet to launch attacks and cause node congestion of wireless sensor networks, which is a common and serious threat. Due to the various kinds of features required in a Peer-to-Peer (P2P) botnet for DDoS attack detection via current machine learning methods and the failure to effectively detect encrypted botnets, this paper extracts the data packet size and the symmetric intervals in flow according to the concept of graphic symmetry. Combined with flow information entropy and session features, the frequency domain features can be sorted so as to obtain features with better correlations, which solves the problem of multiple types of features required for detection. Information entropy corresponding to the flow size can distinguish an encrypted botnet. This method is implemented through machine learning techniques. Experimental results show that the proposed method can detect the P2P botnet for DDoS attack and the detection accuracy is higher than that of traditional feature detection.

Sign in / Sign up

Export Citation Format

Share Document