Seizure forecasting using single robust linear feature as correlation vector of seizure-like events in brain slices preparation in vitro

ABSTRACT

Objective: Epilepsy is a neurological disorder affecting 50 million individuals globally. Modern research has inspected the likelihood of forecasting epileptic seizures. Algorithmic investigations are giving promising results for seizure prediction. Though mostly seizure prediction algorithm uses pre-ictal (prodromal symptoms) events for prediction. On the contrary, prodromal symptoms may not necessarily be present in every patient or subject. This paper focuses on seizure forecasting regardless of the presence of pre-ictal (prodromal symptoms) using the single robust feature with maximum accuracy. Method: We evaluated datasets having 4-aminopydine induced seizure-like events rat’s hippocampa slices and cortical tissue from pharmacoresistant epilepsy patients. The proposed methodology applies the Discrete Wavelet Transform (DWT) at levels 1-5 utilizing 'Daubechies-4'. Linear Discriminant classifier (LDC), Quadratic Discriminant Classifier (QDC) and Support Vector Machine (SVM) were used to classify each signal using eight discriminative features. Results: Classifier performance was assessed by parameters like true detections (TD), false detection (FD), accuracy (ACC), sensitivity (SEN), specificity (SPF), and positive predicted value (PPC), negative predicted value (NPV). Highest classification feature was selected as a seizure forecasting correlation vector and decision rule was formulated for seizure forecasting. Correlation vector served as a forecaster for current EEG activity. Proposed decision rule forecasted ongoing signal activity towards possible seizure condition true or false. The suggested framework revealed forecasting of ictal events at 10 seconds before the actual seizure. Conclusion: It is worth mentioning that the proposed study utilized a single linear feature to predict seizures precisely. Moreover, utilization of single feature encouraged in subsiding system complexity, processing delays, and system latency.

KEYWORDS:

• Epilepsy
• seizure
• prediction
• forecasting
• wavelet transform
• EEG feature extraction
• LDC
• QDC
• SVM

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Muhammad Khizar Abbas

Muhammad Khizar Abbas completed his Masters in Electrical (Communication) Engineering from Pakistan Navy Engineering College, Karachi, National University of Sciences & Technology, Islamabad, Pakistan. Currently, involved in research activities at Power Research Laboratory, Pakistan Navy Engineering College - NUST. His research area includes signal processing, wireless networks, machine learning applications in biomedical sciences, and power systems.

Muhammad Liaquat Raza

Muhammad Liaquat Raza is an associate professor and currently serving as Chairman Department of Pharmacology, Faculty of Pharmacy, Hamdard University. Dr Raza has acquired B.Pharmacy, M.Phil (Pharmacology) degrees from University of Karachi and PhD in Medical Neurosciences from Charite Universitatmedizine, Berlin, Germany. He is trained pharmacologist with focus on CNS disease. He is also associate editor of BMC neuroscience & BMC Pharmacology and Toxicology. He has contributed several peer reviewed article in reputed impact factor journals.

Syed Sajjad Haider Zaidi

Syed Sajjad Haider Zaidi received his Doctorate in Electrical from Michigan State University, East Lansing, in 2010. Since 2010, he is affiliated with the National University of Sciences and Technology, Islamabad, as Assistant Professor. He is appointed in the Department of Electronics and Power Engineering, Pakistan Navy Engineering College, Karachi. He is a member of the National task force for OBE implementation. He has established Power Research Laboratory (PRL) at PNEC with prime aim to undertake projects related to Power and related systems. He is presently working on funded projects from Higher Education Commission, US Aid, National Engineering and Scientific Commission, and local industry. Among his diversified research interests are power systems, acoustic intelligence, time–frequency distribution, pattern recognition, fault prognosis.

Bilal Muhammad Khan

Bilal Muhammad Khan holds PhD and Post Doc in wireless communication networks and controls from the University of Sussex UK. He was affiliated with Sussex University UK as an Associate Lecturer and Visiting Research Fellow. Currently he is working as Assistant Professor at National University of Sciences and Technology. He is involved in various projects on design of wireless sensor networks, programmable logic controllers, Microcontrollers, Systems administration and Software training. He has published number of journal papers and written many book chapters and also serving in the editorial boards of journals. His research interests are in the area of wireless sensor networks, wireless local area networks.

Uwe Heinemann

Uwe Heinemann Late Prof. was renowned German neuroscientist, having more than 500 publications towards his credit. He served at various capacities during his career in scientific societies, grant agencies and editorial bodies. His main focus of research wwas learning and memory, pharmacoresistant epilepsy, stroke, sharp wave ripple. Before his retirement he served as Director Institute of Neurophysiology at Charite Universitatmedizine, Berlin, Germany. He has supervised several Phd students who are serving as faculty members at various universities around the world.