OPTIMIZED EARLY SEIZURE DETECTION FOR A CLOSED LOOP INTERVENTION DEVICE IN EPILEPSY
Relevant for Research Area
Prof. Dr. Andreas Schulze-Bonhage (Contact PI)
Current treatment concepts for epilepsy are based on a continuous drug delivery or electrical stimulation to prevent the occurrence of seizures, exposing the brain and body to mostly unneeded risk of adverse effects. To address the infrequent occurrence and short duration of epileptic seizures, intelligent implantable closed loop devices are needed which are based on a refined analysis of ongoing brain activity and highly specific algorithms. Current devices for the treatment of epilepsy via an implanted device either use an open loop approach or a closed loop approach taking a high number of false positive detections and as consequence unnecessary interventions into account. In this project seizure detection algorithms based on features derived from experience in signal processing in EEG signals combined with state of the art machine learning methods are developed. They combine low false detection rates with mean detection delays in the range of 4 seconds, which may be in many seizures sufficient to perform an intervention before a spread of the EEG activity to adjacent brain areas.
Responsive neurostimulation needs continuous recording and analysis of brain signals for the detection of time windows for effective intervention. Feature extraction and machine learning approaches were developed for the detection of specific patterns in intracranial long-term recordings of patients with epilepsy. The new approaches for data analysis have increased the specificity of seizure detection compared to conventional threshold-based classifiers considerable while reducing the latency between the onset of epileptic discharges to their detection, allowing for a closed-loop intervention in the range of a few seconds. These improvements both reduce the risk of unwanted effects of stimulation due to false detections, and improves chances for an effective focal intervention during ongoing seizures. In addition, slower alterations in brain dynamics were addressed to track modulations of excitability. This opens up new options for the design of devices for individualized interventions.
Schulze-Bonhage A, Somerlik K, Dümpelmann M. Closed-Loop Stimulation zur Epilepsietherapie. Z Epileptologie 2014; 27: 55-59.
Dümpelmann M, Jacobs J, Schulze-Bonhage A. Temporal and spatial characteristics of high frequency oscillations as a new biomarker in epilepsy. Epilepsia 2015 56(2): 197–206.
Donos C, Dümpelmann M, Schulze-Bonhage A. Early Seizure Detection Algorithm Based on Intracranial EEG and Random Forest Classification. International Journal of Neural Systems 2015 Vol. 25, No. 5 1550023
Meisel C, Schulze-Bonhage A, Freestone D, Cook MJ, Achermann P, Plenz D. Intrinsic excitability measures track antiepileptic drug action and uncover increasing/decreasing excitability over the wake/sleep cycle. Proc Natl Acad Sci 2015 112(47):14694–9
Meisel C, Plenz D, Schulze-Bonhage A, Reichmann H. Quantifying antiepileptic drug effects using intrinsic excitability measures. Epilepsia 2016 57(11):e210–e215
Bruder JC, Dümpelmann M, Lachner D, Mader M, Schulze-Bonhage A, Jacobs-Le Van J. Physiological Ripples Associated with Sleep Spindles Differ in Waveform Morphology from Epileptic Ripples. International Journal of Neural Systems 2017 27(7): 1750011.
Zijlmans M, Worrell GA, Dümpelmann M, Stieglitz T, Barborica A, Heers M, Ikeda A, Usui N, Le Van Quyen M. How to record high frequency oscillations in epilepsy: a practical guideline. 2017 Epilepsia 58(8):1305-1315.
Cosandier-Rimélé D, Ramantani G, Zentner J, Schulze-Bonhage A, Dümpelmann M. A realistic multimodal modeling approach for the evaluation of distributed source analysis: application to sLORETA. Journal of Neural Engineering, 2017
Schulze-Bonhage A: Brain stimulation as a neuromodulatory epilepsy therapy. Seizure-eur J Epilep, 2017; 44: 169-175
Lachner Piza D, Epitashvili N, Schulze-Bonhage A, Stieglitz T, Jacobs J, Dümpelmann M. A single channel sleep-spindle detector based on multivariate classification of EEG epochs: MUSSDET. Journal of Neuroscience Methods 2018: 297:31-43.
Donos C, Malîia M, Dümpelmann M, Schulze-Bonhage A. Seizure onset predicts its type. Epilepsia 2018 59(3):650–660.
Conference proceedings (peer reviewed)
Somerlik-Fuchs KH, Christ O, Dümpelmann M, Hofmann UG, Stieglitz T, Schulze-Bonhage A. Development of a Closed Loop Stimulation System for Epilepsy Therapy. Proceedings of 6th European Conference of the International Federation for Medical and Biological Engineering (MBEC 2014), 2014
Dümpelmann M, Cosandier-Rimélé D, Ramantani G, Schulze-Bonhage A. A Novel Approach for Multiscale Source Analysis and Modeling of Epileptic Spikes. 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. 2015
Lachner Piza D, Bruder JC, Jacobs J, Schulze-Bonhage A, Stieglitz T, Dümpelmann M. Differentiation of spindle associated hippocampal HFOs based on a correlation analysis. 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. 2016
Lachner Piza D, Schulze-Bonhage A, Stieglitz T, Jacobs J, Dümpelmann M. Depuration and augmentation of training data for supervised learning based detectors of EEG patterns. 8th International IEEE/EMBS Conference on Neural Engineering
(NER), 2017, pp 497 - 500.
Manzouri F, Schulze-Bonhage A, Dümpelmann M, Heller S, Woias P. Optimized Detector for Closed-loop Devices for Neurostimulation. IEEE International Conference on Systems, Man, and Cybernetics, Banff, Canada, 2017; pp 2158-2163.
Hügle, M, Heller S, Watter M, Blum M, Manzouri F, Dümpelmann M, Schulze-Bonhage A, Woias P, Boedecker J. Early Seizure Detection with an Energy-Efficient Convolutional Neural Network on an Implantable Microcontroller. 2018 International Joint Conference on Neural Networks, Rio de Janeiro Brazil, 2018.
Heller S, Hügle M, Nematollahi I, Manzouri F, Dümpelmann M, Schulze-Bonhage A, Bödecker J, Woias P. Early Hardware Implementation of a Performance and Energy-Optimized Convolutional Neural Network for Seizure Detection. 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Honolulu , USA, 2018