This work presents a new method that combines symbol dynamics methodologies with an Ngram algorithm for the detection and prediction of epileptic seizures. The presented approach specifically applies Ngram-based pattern recognition, after data pre-processing, with similarity metrics, including the Hamming distance and Needlman-Wunsch algorithm, for identifying unique patterns within epochs of time. Pattern counts within each epoch are used as measures to determine seizure detection and prediction markers. Using 623 hours of intracranial electrocorticogram recordings from 21 patients containing a total of 87 seizures, the sensitivity and false prediction/detection rates of this method are quantified. Results are quantified using individual seizures within each case for training of thresholds and prediction time windows. The statistical significance of the predictive power is further investigated. We show that the method presented herein, has significant predictive power in up to 100% of temporal lobe cases, with sensitivities of up to 70–100% and low false predictions (dependant on training procedure). The cases of highest false predictions are found in the frontal origin with 0.31–0.61 false predictions per hour and with significance in 18 out of 21 cases. On average, a prediction sensitivity of 93.81% and false prediction rate of approximately 0.06 false predictions per hour are achieved in the best case scenario. This compares to previous work utilising the same data set that has shown sensitivities of up to 40–50% for a false prediction rate of less than 0.15/hour.
References
[1]
Mormann F, Andrzejak RG, Elger CE, Lehnertz K (2007) Seizure prediction: the long and winding road. Brain 130: 314–333.
[2]
Aschenbrenner-Scheibe R, Maiwald T, Winterhalder M, Voss HU, Trimmer J, et al. (2003) How well can epileptic seizures be predicted? An evaluation of a nonlinear method. Brain 126: 1–11.
[3]
Eftekhar A, Toumazou C, Drakakis EM (2013) Empirical mode decomposition: Real-time imple-mentation and applications. Journal of Signal Processing Systems.
[4]
Stein GA, Eder GH, Blum ED, Drachev A, Fisher SR (2000) An automated drug delivery system for focal epilepsy. Epilepsy Research 39.
[5]
Hill WM, Wong M, Amarakone A, Rothman MS (2000) Rapid cooling aborts seizure-like activity in rodent hippocampal-entorhinal slices. Epilepsia 41.
[6]
Osorio I, Frei MG, Sunderam S, Giftakis J, Bhavaraju NC, et al.. (2000) Automated seizure abate-ment in humans using electrical stimulation. Annals of neurology 57.
[7]
Badawi N, Kurinczuk JJ, Keogh JM, Alessandri LM, O’Sullivan, et al (1998) Risk factors for newborn encephalopathy: The Western Australian case-control study. British Medical Journal 317: 1549–53.
[8]
Trent neonatal survey report 2007 (2007) Technical report, Leicester: Dept of Health Sciences, University of Leicester.
[9]
Faul S, Boylan G, Connolly S, Marnane L, Lightbody G (2005) An evaluation of automated seizure detection methods. Clinical Neurophysiology 116.
[10]
Niran al Naqeeb A, Edwards D, Cowan FM, Azzopardi D (1999) Assessment of neonatal en-cephalopathy by amplitude-integrated electroencephalography. Paediatrics 103: 1263–1271.
[11]
Toet M, Hellstrom-Westas L, Groenendaal F, Eken P, de Vries LS (1999) Amplitude-integrated EEG 3 and 6 hours after birth in full term neonates with hypoxic-ischaemic encephalopathy. Arch Dis Child Fetal Neonatal Ed 81: F19–F23.
[12]
Rennie JM, Chorley G, Boylan GB, Pressler R, Nguyen Y, et al.. (2004) Non-expert use of the cerebral function monitor for neonatal seizure detection. Arch Dis Child fetal neonatal Edition.
[13]
Logesparan L, Casson AJ, Rodriguez-Villegas E (2012) Optimal features for online seizure detec-tion. Med Biol Eng Comput 50: 659–69.
[14]
Yadav R, Shah AK, Loeb JA, Swamy MDS, Agarwal R (2012) Morphology-based automatic seizure detector for intracerebral EEG recordings. IEEE Transactions on Biomedical Engineering 59: 1871–1881.
[15]
Aarabi A, Wallois F, Grebe R (2006) Automated neonatal seizure detection: A multistage classification system through feature selection based on relevance and redundancy analysis. Clinical Neurophysiology 117.
[16]
Feldwisch-Drentrup H, Schulze-Bonhage A, Trimmer J, Schelter B (2011) Statistical validation of event predictors: A comparitive study based on the field of seizure prediction. Physical Review E 83.
[17]
Feldwisch-Drentrup H, Ihle M, Le Van Quyen M, Teixeira C, Dourada A, et al. (2011) Anticipating the unobserved: Prediction of subclinical seizures. Epilepsy and Behavior 22: S119–S126.
[18]
Aarabi A, He B (2012) A rule-based seizure prediction method for focal neocortical epilepsy. Clinical Neurophisiology 123: 1111–1122.
[19]
Gadhoumi K, Lina JM, Gotman J (2013) Seizure prediction in patients with mesial temporal lobe epilepsy using EEG measures of state similarity. Clinical Neurophisiology.
[20]
Williamson JR, Bliss DW, Browne DW, Narayanan JT (2012) Seizure prediction using EEG spa-tiotemporal correlation structure. Epilepsy and Behavior 25: 230–238.
[21]
Zandi AS, Tafreshi R, Javidan M, Dumont GA (2013) Predicting epileptic seizures in scalp EEG based on a variational Bayesian gaussian mixture model of zero-crossing intervals. IEEE Transac-tions on Biomedical Engineering 60: 1401–1413.
[22]
Cook MJ, O’Brien TJ, Berkovic SF, Murphy M, Morokoff A, et al. (2013) Prediction of seizure likelihood with a long-term, implanted seizure advisory system in patients with drug-resistant epilepsy: a first-in-man study. The Lancet Neurology 12: 563–571.
[23]
Juffali W, El-Imad J, Eftekhar A, Toumazou C (2010) The WiNAM project: Neural data analysis with applications to epilepsy. IEEE BioCAS: 45–48.
[24]
Jain A, Duin R, Mao J (2000) Statistical Pattern Recognition: A Review. IEEE Transactions on Pattern Analysis and Machine Intelligence 22.
[25]
Paraskevopoulou SE, Barsakcioglu DY, Saberi MR, Eftekhar A, Constandinou TG (2013) Feature extraction using first and second derivative extrema (FSDE) for real-time and hardware efficient spike sorting. Journal of Neuroscience Methods 215: 29–37.
[26]
Banerjee S, Pederesen T (2003) The design, implementation and use of the ngram statistics package. Proceedings of Intelligent Text Processing and Computational Linguistics.
[27]
Bengio Y, Ducharme R, Vincent P, Jauvin C (2003) A Neural Probabilistic Language Model. Journal of Machine Learning Research 3.
[28]
Petrosian A (1995) Kolmogorov complexity of finite sequences and recognition of different preictal EEG patterns. Proc IEEE: 212–217.
[29]
Cohen P, Heeringa B, Adams NM (2002) An unsupervised algorithm for segmenting categorial timeseries into episodes. In: ESF Exploratory Workshop on Pattern Detection and Discovery in Data Mining. Springer-Verlag, 99–106.
[30]
Keller K, Lauffer H, Sinn M (2007) Ordinal analysis of EEG time series. Chaos and Complexity 2: 247–258.
[31]
Schindler K, Gast H, Stieglitz L, Stibal A, Hauf M, et al. (2011) Forbidden ordinal patterns of periictal intracranial EEG indicate deterministic dynamics in human epileptic seizures. Epilepsia 52: 1771–1780.
[32]
Daw CS, Finney CEA, Tracy ER (2003) A review of symbolic analysis of experimental data. Review Scientific Instruments 74.
[33]
Gupta S, Khatkhate A, Ray A, Keller E (2006) Identification of statistical patterns in complex systems via symbolic time series analysis. ISA Transactions 45: 477–490.
[34]
Gailey PC, Hively LM, Protopopescu VA (1999) Robust detection of dynamical change in scalp EEG. Proceedings Experimental Chaos.
[35]
Hively LM, Protopopescu VA, Munro NB (2005) Enhancements in epilepsy forewarning via phase-space dissimilarity. Journal of Clinical Neurophysiology 22: 402–9.
[36]
Eftaxias K, Minadakis G, Athanasopoulou L, Kalimeri M, Ptirakis SM, et al.. (2011) Are epileptic seizures quakes of the brain? An approach by means of nonextensive Tsallis statistics. Physics Geophysics 1.
[37]
Hively LM, Protopopescu VA, Gailey PC (2000) Timely detection of dynamical change in scalp EEG signals. Chaos 10: 864–875.
[38]
Takens F (1981) Detecting strange attractors in turbulence. Dynamical Systems and Turbulence, Lecture notes in mathematics.
[39]
Needleman SB, Wunsch CD (1970) A general method applicable to the search for similarity in the amino acid sequences of two proteins. J Mol Biol 48: 443–453.
[40]
Winterhalder M, Maiwald T, Voss HU, Aschenbrenner-Scheibe R, Trimmer J, et al. (2003) The seizure prediction characteristic: a general framework to assess and compare seizure prediction methods. Epilepsy Behaviour 3: 318–325.
[41]
Winterhalder M, Schelter B, Maiwald T, Brandt A, Schad A, et al.. (2006) Testing statistical significance of multivariate time series analysis techniques for epileptic seizure prediction. Chaos 16.
[42]
Schulze-Bonhage A, Feldwisch-Drentrup H, Ihle M (2011) The role of high-quality EEG databases in the improvement and assessment of seizure prediction methods. Epilepsy and Behavior 22: S88–S93.
[43]
Klatt J, Feldwisch-Drentrup H, Ihle M, Navarro V, Neufang M, et al. (2012) The epilepsiae database: An extensive electroencephalography database of epilepsy patients. Epilepsia 53: 1669–1676.
[44]
Zhou W, Liu Y, Yuan Q, Li X (2013) Epileptic seizure detection using lacunarity and Bayesian linear discriminant analysis in intracranial EEG. IEEE Transactions on Biomedical Engineering.
[45]
Maiwald T, Winterhalder M, Voss HU, Aschenbrenner-Scheibe R, Trimmer J, et al. (2004) Compar-ison of three nonlinear seizure prediction methods by means of the seizure rpediction characteristic. Physica D 194: 357–368.
[46]
Schelter B, Winterhalder M, Maiwald T, Brandt A, Schad A, et al. (2006) Do false predictions of seizures depend on the state of vigilance? A report from two seizure-prediction methods and proposed remedies. Epilepsia 47: 2058–2070.
