Since the 1970s advances in science and technology during each succeeding decade have renewed the expectation of efficient, reliable automatic epileptiform spike detection (AESD). But even when reinforced with better, faster tools, clinically reliable unsupervised spike detection remains beyond our reach.
Expert-selected spike parameters were the first and still most widely used for AESD. Thresholds for amplitude, duration, sharpness, rise-time, fall-time, after-coming slow waves, background frequency, and more have been used. It is still unclear which of these wave parameters are essential, beyond peak-peak amplitude and duration. Wavelet parameters are very appropriate to AESD but need to be combined with other parameters to achieve desired levels of spike detection efficiency.
Artificial Neural Network (ANN) and expert-system methods may have reached peak efficiency. Support Vector Machine (SVM) technology focuses on outliers rather than centroids of spike and nonspike data clusters and should improve AESD efficiency. An exemplary spike/non-spike database is suggested as a tool for assessing parameters and methods for AESD and is available in CSV or Matlab formats from the author at brainvue@gmail.com. Exploratory Data Analysis (EDA) is presented as a graphic method for finding better spike parameters and for the step-wise evaluation of the spike detection process.
HarnerRHalbergF. Electrocorticographic differences in D8 mice at times of daily high and low susceptibility to audiogenic convulsion. The Physiologist1958; 1: 34–35.
2.
HarnerRN. Electrocorticography and frequency analysis in mice: circadian periodicity in electrocerebral activity. Electroencephalogr Clin Neurophysiol1961; 13: 752–761.
3.
CelesiaGGChenRC. Parameters of spikes in human epilepsy. Dis Nerv Syst1976; 37: 277–281.
4.
GotmanJ. Quantitative measurements of epileptic spike morphology in the human EEG. Electroencephalogr Clin Neurophysiol1980; 48: 551–557.
5.
FrostJDJr. Automatic recognition and characterization of epileptiform discharges in the human EEG. J Clin Neurophysiol1985; 2: 231–249.
6.
KtonasPYLuohWMKejariwalMLReillyELSewardMA. Computeraided quantification of EEG spike and sharp wave characteristics. Electroencephalogr Clin Neurophysiol1981; 51: 237–243.
GotmanJ. Automatic detection of seizures and spikes in the EEG In: LudersHO, (ed). Epilepsy Surgery. New York: Raven; 1991: 307–316.
10.
WilsonSBEmersonR. Spike detection: a review and comparison of algorithms. Clin Neurophysiol2002; 113: 1873–1881.
11.
WilsonSBHarnerRNDuffyFHTharpBRNuwerMRSperlingMR. Spike detection. I. correlation and reliability of human experts. Electroencephalogr Clin Neurophysiol1996; 98: 186–198.
12.
CassonAJLunaERodriguez-VillegasE. Performance metrics for the accurate characterisation of interictal spike detection algorithms. J Neurosci Methods2009; 177: 479–487.
13.
ArgoudFIDe AzevedoFMNetoJMGrilloE. SADE3: an effective system for automated detection of epileptiform events in long-term EEG based on context information. Med Biol Eng Comput2006; 44: 459–470.
14.
CarrieJR. A technique for analyzing transient EEG abnormalities. Electroencephalogr Clin Neurophysiol1972; 32: 199–201.
15.
CarrieJR. A hybrid computer technique for detecting sharp EEG transients. Electroencephalogr Clin Neurophysiol1972; 33: 336–338.
16.
CarrieJRFrostJDJrWavelength -amplitude profile analysis in clinical EEG. In: KellawayPPetersenI, (eds). Automation of Clinical Electroencephalography. New York: Raven; 1973: 65–74.
17.
GotmanJGloorP. Automatic recognition and quantification of interictal epileptic activity in the human scalp EEG. Electroencephalogr Clin Neurophysiol1976; 41: 513–529.
18.
GotmanJIvesJRGloorP. Automatic recognition of inter-ictal epileptic activity in prolonged EEG recordings. Electroencephalogr Clin Neurophysiol1979; 46: 510–520.
19.
HarnerRN. Sequential analysis and quantification of the electroencephalogram. Electroencephalogr Clin Neurophysiol1973; 34: 791.
20.
HarnerRNComputer analysis and clinical EEG interpretation: perspective and application. In: DolceGKunkelH, (eds). Computerized EEG Analysis CEAN. Stuttgart: G. Fischer Verlag; 1975: 337–343.
21.
HarnerRNOstergrenKASequential analysis and quantification of quasi-stable and paroxysmal EEG activity. In: KellawayPPetersenI, (eds). Quantitative Analytic Studies in Epilepsy. New York: Raven Press; 1976: 343–353.
22.
HarnerRNEEG analysis in the time domain. In: RemondA, (ed). EEG Informatics: A Didactic Review of Methods and Applications of EEG Data Processing. Amsterdam: Elsevier; 1977: 57–82.
23.
CarrieJRThe detection and quantification of transient and parxysmal EEG abnormalitites. In: KellawayPPetersenI, (eds). Automation of Clinical Electroencephalography. New York: Raven; 1973: 217–226.
24.
CarrieJR. A hybrid computer system for detecting and quantifying spike and wave EEG patterns. Electroencephalogr Clin Neurophysiol1972; 33: 339–341.
SmithJR. Automatic analysis and detection of EEG spikes. IEEE Trans Biomed Eng1974; 21: 1–7.
29.
GoldbergPSamson-DollfusDA time domain analysis method applied to the recognition of EEG rhythms. In: DolceGKünkelH, (eds). CEAN-Computerized EEG Analysis. Stuttgart: Fischer; 1975: 19–26.
30.
ZetterbergLHSpike detection by computer and by analog equipment. In: KellawayPZetterbergI, (eds). Automation of Clinical Electroencephalography. New York: Raven; 1973: 227–234.
31.
Lopes daSilva FHDuquesnoyAvan HultenKLommenJG. The recognition of patterns in automated EEG analysis. Rev Electroencephalogr Neurophysiol Clin.1976; 6: 186–194.
32.
FaureCDubuissonBSamson-DollfusD. Aid to the diagnosis of epilepsy: codification of EEG tracings for the characterization of detectable nonstationary phenomena. Int J Biomed Comput1980; 11: 21–32.
33.
FaureC. Attributed strings for recognition of epileptic transients in EEG. Int J Biomed Comput1985; 16: 217–229.
34.
GloverJRJrKtonasPYRaghavanNUrunuelaJMVelamuriSSReillyEL. A multichannel signal processor for the detection of epileptogenic sharp transients in the EEG. IEEE Trans Biomed Eng1986; 33: 1121–1128.
35.
GloverJRJrRaghavanNKtonasPYFrostJDJr. Context-based automated detection of epileptogenic sharp transients in the EEG: elimination of false positives. IEEE Trans Biomed Eng1989; 36: 519–527.
MaKMCelesiaGGBirkemeierWP. Non-linear boundaries for differentiation between epileptic transients and background activities in EEG. IEEE Trans Biomed Eng1977; 24: 288–290.
39.
von AlbertHH. Automatic analysis of long-term EEG in epilepsy: first experiences with a new part-automatic analytic system (author's transl). EEG EMG Z Elektroenzephalogr Elektromyogr Verwandte Geb1977; 8: 105–111.
40.
GeigerLRHarnerRN. EEG patterns at the time of focal seizure onset. Arch Neurol1978; 35: 276–286.
PfurtschellerGFischerG. A new approach to spike detection using a combination of inverse and matched filter techniques. Electroencephalogr Clin Neurophysiol1978; 44: 243–247.
43.
BurrW. Spike waves pattern recognition in EEG long term registrations using a variably programmed laboratory computer. EEG EMG Z Elektroenzephalogr Elektromyogr Verwandte Geb1980; 11: 135–141.
44.
Guedes deOliveira PHLopes daSilva FH. A topographical display of epileptiform transients based on a statistical approach. Electroencephalogr Clin Neurophysiol1980; 48: 710–714.
45.
QuyRJFitchPWillisonRG. High-speed automatic analysis of EEG spike and wave activity using an analogue detection and microcomputer plotting system. Electroencephalogr Clin Neurophysiol1980; 49: 187–189.
46.
Guedes deOliveira PQueirozCLopes daSilva F. Spike detection based on a pattern recognition approach using a microcomputer. Electroencephalogr Clin Neurophysiol1983; 56: 97–103.
HarnerRTopographic analysis of multichannel EEG data. In: Samson-DollfusD, (ed). Statistics and Topography in Quantitative EEG. Paris: Elsevier; 1988: 49–61.
51.
ArleJESimonRH. An application of fractal dimension to the detection of transients in the electroencephalogram. Electroencephalogr Clin Neurophysiol1990; 75: 296–305.
52.
WitteHEiseltMPatakovaI. Use of discrete hilbert transformation for automatic spike mapping: a methodological investigation. Med Biol Eng Comput1991; 29: 242–248.
PinonJMRubelPMaravalGMauguiereFRevolM. Discriminant analysis of graphic elements of the EEG: application to the detection of episodes of diffuse spike-waves. Int J Biomed Comput1982; 13: 237–261.
JandoGSiegelRMHorvathZBuzsakiG. Pattern recognition of the electroencephalogram by artificial neural networks. Electroencephalogr Clin Neurophysiol1993; 86: 100–109.
58.
OzdamarOKalayciT. Detection of spikes with artificial neural networks using raw EEG. Comput Biomed Res1998; 31: 122–142.
59.
ParkHSLeeYHKimNGLeeDSKimSI. Detection of epileptiform activities in the EEG using neural network and expert system. Stud Health Technol Inform1998; 52 Pt 2: 1255–1259.
60.
YuAZhangYYuK. Study on EEG signals data compression and spikes recognition with wavelet neural network. Zhongguo Yi Liao Qi Xie Za Zhi1998; 22: 249–253.
61.
JamesCJJonesRDBonesPJCarrollGJ. Detection of epileptiform discharges in the EEG by a hybrid system comprising mimetic, self-organized artificial neural network, and fuzzy logic stages. Clin Neurophysiol1999; 110: 2049–2063.
62.
ZhangYLiuAYuK. A wavelet neural network algorithm of EEG signals data compression and spikes recognition. Sheng Wu Yi Xue Gong Cheng Xue Za Zhi1999; 16: 172–176.
63.
KoCWChungHW. Automatic spike detection via an artificial neural network using raw EEG data: effects of data preparation and implications in the limitations of online recognition. Clin Neurophysiol2000; 111: 477–481.
SenhadjiLDillensegerJLWendlingFRochaCKinieA. Wavelet analysis of EEG for three-dimensional mapping of epileptic events. Ann Biomed Eng1995; 23: 543–552.
66.
GrossmanAMorletJ. Decomposition of hardy functions into square integrable wavelets of constantshape. SIAM J Math Anal1984; 15: 723–736.
67.
DaubechiesI. Orthonormal bases of compactly supported wavelets. Com Pure Appl Math1988; 41: 909–996.
68.
MallatSG. A theory of multiresolution signal decomposition: the wavelet representation. IEEE Trans PAML1989; 11: 674–693.
69.
KurthCGilliamFSteinhoffBJ. EEG spike detection with a kohonen feature map. Ann Biomed Eng2000; 28: 1362–1369.
70.
TangSZhangHZhengC. Detection of spikes in epileptic EEG based on multiresolution tsallis' entropy. Sheng Wu Yi Xue Gong Cheng Xue Za Zhi2001; 18: 378–380.
71.
SugiTNakamuraMIkedaAShibasakiH. Adaptive EEG spike detection: determination of threshold values based on conditional probability. Front Med Biol Eng2002; 11: 261–277.
72.
AcirNGuzelisC. Automatic spike detection in EEG by a two-stage procedure based on support vector machines. Comput Biol Med2004; 34: 561–575.
73.
AdjouadiMSanchezDCabrenzoM. Interictal spike detection using the walsh transform. IEEE Trans Biomed Eng.2004; 51: 868–872.
74.
HuJSiJOlsonBPHeJ. Feature detection in motor cortical spikes by principal component analysis. IEEE Trans Neural Syst Rehabil Eng2005; 13: 256–262.
75.
SubasiAErcelebiE. Classification of EEG signals using neural network and logistic regression. Comput Methods Programs Biomed2005; 78: 87–99.
76.
PhlypoRVan HesePHallezH. Extracting the spike process from the EEG by spatially constrained ICA. Conf Proc IEEE Eng Med Biol Soc2006; 1: 5286–5289.
77.
TzallasATOikonomouVPFotiadisDI. Epileptic spike detection using a kalman filter based approach. Conf Proc IEEE Eng Med Biol Soc2006; 1: 501–504.
78.
UrrestarazuEIriarteJArtiedaJAlegreMValenciaMViteriC. Independent component analysis separates spikes of different origin in the EEG. J Clin Neurophysiol2006; 23: 72–78.
79.
ValentiPCazamajouEScarpettiniMAizembergASilvaWKochenS. Automatic detection of interictal spikes using data mining models. J Neurosci Methods2006; 150: 105–110.
80.
WuXGuoYQiuTBaoH. Extraction of epileptic spike waves in EEG with B-spline wavelet. Sheng Wu Yi Xue Gong Cheng Xue Za Zhi2006; 23: 947–950.
81.
ChenZChenJQiuT. Spike extraction of epileptic waves in EEG based on EMD. Sheng Wu Yi Xue Gong Cheng Xue Za Zhi2007; 24: 973–977.
82.
InanZHKuntalpM. A study on fuzzy C-means clustering-based systems in automatic spike detection. Comput Biol Med2007; 37: 1160–1166.
83.
LatkaMWasZKozikAWestBJ. Wavelet analysis of epileptic spikes. Phys Rev E Stat Nonlin Soft Matter Phys2003; 67: 052902.
84.
ShaoCLuJZhouH. Detection of epileptic waves in EEG based on wavelet transform. Sheng Wu Yi Xue Gong Cheng Xue Za Zhi2002; 19: 259–263, 272.
85.
QuirogaRQNadasdyZBen-ShaulY. Unsupervised spike detection and sorting with wavelets and superparamagnetic clustering. Neural Comput2004; 16: 1661–1687.
86.
AcirNOzturaIKuntalpMBaklanBGuzelisC. Automatic detection of epileptiform events in EEG by a three-stage procedure based on artificial neural networks. IEEE Trans Biomed Eng2005; 52: 30–40.
87.
AdjouadiMCabrerizoMAyalaM. Detection of interictal spikes and artifactual data through orthogonal transformations. J Clin Neurophysiol2005; 22: 53–64.
88.
MengHWangZGeM. An understanding for the abnormal spikes of the EEG simulation in a 2-d neural network. Conf Proc IEEE Eng Med Biol Soc2005; 3: 3008–3011.
89.
DumpelmannMElgerCE. Automatic detection of epileptiform spikes in the electrocorticogram: a comparison of two algorithms. Seizure1998; 7: 145–152.
90.
PangCCUptonARShineGKamathMV. A comparison of algorithms for detection of spikes in the electroencephalogram. IEEE Trans Biomed Eng2003; 50: 521–526.
91.
ObeidIWolfPD. Evaluation of spike-detection algorithms for a brain-machine interface application. IEEE Trans Biomed Eng2004; 51: 905–911.
92.
BrownMWIIIPorterBEDlugosDJ. Comparison of novel computer detectors and human performance for spike detection in intracranial EEG. Clin Neurophysiol2007; 118: 1744–1752.
93.
KimKHKimSSKimSJ. Improvement of spike train decoder under spike detection and classification errors using support vector machine. Med Biol Eng Comput2006; 44: 124–130.
94.
TukeyJW. Exploratory Data Analysis. Phillipines: Addison Wesley; 1977.
95.
MartinezWLMartinezAR. Exploratory Data Analysis with Matlab. New York: CRC Press; 2005.
96.
RizkMWolfPD. Optimizing the automatic selection of spike detection thresholds using a multiple of the noise level. Med Biol Eng Comput Epub, Feb. 10, 2009.
