What is the clinical evidence from this study?

Study design: Other. Population: Epileptical Seizures. Intervention: Gamma band (30-60 Hz) Haralick features. Primary outcome: Classification between seizures and healthy states (AUC 0.96).

What does this research mean for the field?

Using Haralick features extracted from the gamma band (30-60 Hz) of EEG data achieves a maximum AUC of 0.96 for classifying epileptic seizures, offering an effective detection method that reduces computational load compared to whole-spectrum approaches. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

February 1, 2020

ROC Analysis for detection of Epileptical Seizures using Haralick features of Gamma band

Key Result

Using Haralick features of the gamma band (30-60 Hz) from EEG data achieved a maximum area under the curve (AUC) of 0.96 for classifying between epileptic seizures and healthy states.

Structured PICO

Does using Haralick features of the Gamma band (30-60 Hz) from EEG data accurately detect epileptical seizures?

P

Population

EEG data from subjects with epileptical seizures and healthy controls

I

Intervention

Gamma band (30-60 Hz) Haralick features extracted via Short-time Fourier transform (STFT) and classified using a decision tree

O

Outcome

Detection of epileptical seizures (measured by Area Under Curve in ROC analysis)surrogate

Using Haralick features from only the high-frequency gamma band of EEG data provides high accuracy (AUC 0.96) for seizure detection while reducing computational load.

Main Result

Effect estimate: AUC 0.96

Abstract

In this study, gamma band (30-60 Hz) is used for detection of epileptical seizures using Haralick features. Most of the previous methods are based on the whole frequency spectrum for detection. This work use only high frequency electroencephalogram (EEG) subband for seizure detection using image descriptors. To convert one dimensional EEG data into image Short-time Fourier transform (STFT) has been used. Gamma band is cut from the time frequency (t-f) plane and Haralick features is used as image descriptors to fed in the decision tree classifier. The results have been evaluated using receiver operating characteristic (ROC) analysis. Maximum area under curve (AUC) of 0.96 is obtained to classify between seizures and healthy. Advantage of this work is rather using whole frequency band it utilizes only a particular band which reduces computational load. It also shows the utility of gamma band in seizure detection.