Spike detection algorithm - General discussion

4. General discussion

4.2. Spike detection algorithm

The custom build algorithm using in house resources enabled us to identify, mark, quantify and subsequently report the EEG alterations in terms of spikes, spike clusters, and seizures.

With above 86 – 98 % sensitivity and up to 98% specificity rate, we were able to detect spikes in all groups of animals, using very commonly used definitions for characterizing spikes, i.e. spike height (relative to baseline) and spike width. Two other easily implementable spike identifying parameters, spike slope and Teager Kaiser Energy operator were also incorporated to improve the algorithm’s specificity. The algorithm initially detected spikes that were verified by human experts to rule out the detection of artefacts, which were

not removed by the algorithm automatically. A spike detection algorithm using similar detection parameters in rats (model: i.p. administered kainate) has been described by (White et al., 2006), but no sensitivity or specificity values for spike detections has been given. This algorithm used spike slope (12 times the upslope of EEG background) and spike width (<200ms) parameters for spike detection. Spikes detected by the algorithm were then further employed to automatically detect spike clusters and seizures in animal EEGs. Range autocorrelation method, most efficient of the 4 methods used, to automatically detect seizures, resulted in 95 % positive predictive value, and 100 % sensitivity and specificity.

A more recent publication has described an algorithm using total signal variation and an and advanced wavelet transform technique to find spikes, seizures and other abnormal EEG activity (Bergstrom et al., 2013). The algorithm was established on epileptic mice (intra-hippocampal kainic acid model) and the signal was subdivided in to normal, spikes, seizures and abnormal EEG data. The authors claimed to provide an alternative in place of long used

“Racine scale” for behavioural analysis (Bergstrom et al., 2013). The algorithm used an automatic approach to remove the artifacts by employing a 2^nd empty channel to record extra cerebral input and then subsequently remove them from mouse EEG. Furthermore, the algorithm was verified on data collected with a multi-channel EEG recording system from a model of absence seizure epilepsy in γ-butyrolacetone treated mice (Bergstrom et al., 2013).

The identification of different events was with 99% specificity and 91% sensitivity.

The major challenge in the spike or seizure detection algorithms is to prevent the detection of noise or artefact data which closely resembles interictal spikes and seizures (White et al., 2006). We were able to correctly detect and mark 76 – 84 % of spikes, while 16 – 24 % were marked as artefacts. Among marked artefacts, 76 – 83 % were removed automatically by the algorithm, while the remaining had to be excluded manually. Animal studies yield huge volumes of experimental data during pre-clinical studies. To analyze this large data volume we need to use computationally efficient algorithms. Highly sophisticated, time consuming softwares and algorithms used for human EEG analysis cannot be used on large data sets due to impracticality (White et al., 2006). The solution is developing time saving, simple to execute and computationally efficient algorithms like White et al., (2006) and ours. With our simple, though efficient algorithm, we were able to reduce the analysis time by up to 80%.

Furthermore, our algorithm can be customized using various data processing techniques according to the individual requirements, examine bulk EEGs in short time, detect events of interest using common parameters, like amplitude, duration, spike frequency and power etc.

43 5. Summary

IN-SILICO ANALYSIS, CHARACTERIZATION AND QUANTIFICATION OF EEG ALTERATIONS IN A MOUSE MODELS OF TEMPORAL LOBE EPILEPSY.

Syed Muhammad Muneeb Anjum

Epilepsy is one of the most common chronic neurologic disorders that affects approximately 1% of the general population. Brain insults such as viral encephalitis may initiate the process of epilepsy development known as epileptogenesis. The Theiler’s murine encephalomyelitis virus (TMEV) animal model of viral encephalitis-induced epilepsy, first described by Libbey et. al. (2008), is the first proposed infection driven animal model of epilepsy. This model was reproduced in our lab and we have recorded vEEG (Video-EEG) using cortical electrodes.

This vEEG recording is a primary tool used to characterize the electrophysiological parameters in animal models of epilepsy. Interictal epileptiform discharges have been associated with epilepsy, and EEG alterations such as spikes have widely been accepted diagnostically as sign of epilepsy in humans.

While characterizing the EEG of this animal model of epilepsy, we could show that animals suffer from early seizures after infection. These early seizures recorded in EEGs were either focal or generalized. We have reported that the percentage of animals suffering from seizures after acute infection is about 77%, verifying the findings of others. With visual observations this percentage values is usually lower. An average latent period of 61 days has been recorded and reported using long term continuous (24/7) vEEG recordings with cortical electrodes. Late seizure frequency during chronic phase of the disease was low in this model:

only 38% of infected animals developed epilepsy during this study. In late EEG recording of the animals, we could, just like acute phase after infection, record both focal and generalized seizures. The low frequency of late seizures as well as epileptic animals prompted us to look for alternative read outs as a biomarker for developing epilepsy. We could verify the presence of inter-ictal spikes and spike clusters in all the infected animals. Quantification of spikes in humans and experimental models of epilepsy is often done manually by visual inspection, which requires intensive training, is very laborious, subjective, and error-prone.

However, the number of spikes and spike cluster was too high for visual analysis and hence a computer assisted algorithm was needed to objectively quantify them. We have developed a novel method to reliably and objectively quantify EEG spikes and spike clusters automatically. The idea of this detection technique is to analyse bulk EEG data recorded over

an extended time period and comparing the results with the standardized visual inspection of specialists.

A multitude of constantly updated algorithms over a period of time using EEG recording software LabChart^® (AD Instruments) helped us develop a novel arithmetic & macro based signal processing method to quantify and characterize the inter-ictal spikes with high precision (90-99%). Although spikes and spike clusters have been seen in all the animals (mock & infected) during the early (0-7 days post infection) and late phase (91-97 days post infection) following infection, it is found that they are more frequent in TMEV-infected animals with seizures (early or late). The average number of spikes and spike clusters during the acute and chronic phase of epilepsy development was significantly higher in infected animals compared to controls. A comprehensive analysis of EEGs recorded in the late phase after infection showed that infected mice without early or late seizures were indifferent in spike frequency from controls. However, mice with any type of seizure, early or late, exhibited significantly increased spike and spike cluster frequencies in the late phase. On the contrary, few of the epileptic animals showed low number of spikes and spike clusters, with in the range of control animals. This suggested that spike and spike cluster frequency alone cannot predict individual animals to be epileptic or not. However, this can be used along with other previously identified surrogate markers to access the disease development such as seizure threshold, behaviour and cognitive studies.

Our results suggest that we can use increased spike and spike cluster frequencies in groups of infected animals as a new readout for disease modification or antiepileptogenesis studies to evaluate the effects of pharmacological compounds.

45 6. Zusammenfassung

IN-SILICO-ANALYSE, CHARAKTERISIERUNG UND QUANTIFIZIERUNG VON EEG-ÄNDERUNGEN IN EINER MAUSMODELLE DER TEMPORALLAPPEN-EPILEPSIE.

Syed Muhammad Muneeb Anjum

Epilepsie ist eine der häufigsten chronisch-neurologischen Störungen, die etwa 1% der Allgemeinbevölkerung betrifft. Gehirnschäden wie virale Enzephalitis können den Prozess der Epileptogenese auslösen. Das von Libbey et al. (2008) beschriebene Theilers Murines Enzephalomyelitis-Virus- (TMEV-) Modell der viralen Enzephalitis-induzierten Epilepsie ist das erste infektionsgetriebene Tiermodell für Epilepsie. Dieses Modell wurde in unserem Labor etabliert, und wir haben vEEG mittels kortikaler Elektroden aufgezeichnet. Diese vEEG-Aufzeichnung ist ein primäres Werkzeug zur Charakterisierung der elektrophysiologischen Parameter in Tiermodellen der Epilepsie. Interiktale epileptiforme Entladungen werden mit Epilepsien in Verbindung gebracht, und EEG-Veränderungen wie Spikes werden diagnostisch als Zeichen von Epilepsie beim Menschen akzeptiert.

Während wir das EEG dieses Tiermodells der Epilepsie charakterisieren, konnten wir zeigen, dass Tiere nach der Infektion an frühen Anfällen leiden. Akute Anfälle, die in EEGs aufgezeichnet wurden, konnten fokal oder generalisiert sein. Wir haben berichtet, dass der Prozentsatz der Tiere, die nach einer akuten Infektion akute Anfälle erleiden, etwa 77%

beträgt, was die Ergebnisse anderer Arbeitsgruppen bestätigt. Bei visuellen Beobachtungen ist dieser Prozentsatz normalerweise niedriger. Eine durchschnittliche Latenzzeit von 61 Tagen wurde aufgezeichnet und unter Verwendung von chronische-kontinuierlichen (24/7) vEEG-Aufzeichnungen mit kortikalen Elektroden ermittelt. Bei hippocampalen Tiefenelektroden könnte diese Latenzzeit jedoch aufgrund möglicher fokaler Anfälle viel kürzer sein. Die Häufigkeit von spontanen Anfällen während der chronischen Phase der Erkrankung ist in diesem Modell gering: nur 38% der Tiere entwickelten während dieser Studie Epilepsie. Bei der chronischen EEG-Aufzeichnung der Tiere konnten sowohl fokale als auch generalisierte Anfälle, genau wie während der akuten Phase nach der Infektion, aufgezeichnet werden. Die geringe Häufigkeit von späten Anfällen sowie epileptischen Tieren veranlasste uns, nach alternativen Auslesemöglichkeiten als Biomarker für die Entwicklung von Epilepsie in diesem Modell zu suchen. Wir konnten das Vorhandensein von inter-iktalen Spikes und Clustern in allen infizierten Tieren verifizieren. Die Quantifizierung von Spikes bei Menschen und experimentellen Epilepsiemodellen erfolgt für gewöhnlich

manuell durch visuelle Inspektion, was ein intensives Training erfordert, sehr mühsam, subjektiv und fehleranfällig ist.

Die Anzahl der Spikes und Spike-Cluster war sehr hoch, so dass ein computergestützter Algorithmus erforderlich war, um diese objektiv zu quantifizieren. Dafür haben wir eine neuartige Methode entwickelt, mit der EEG-Spikes und Spike-Cluster zuverlässig und objektiv quantifiziert werden können. Die Idee dieser Detektionstechnik besteht darin, über einen längeren Zeitraum aufgenommene Langzeit-EEG-Daten zu analysieren und die Ergebnisse mit einer standardisierten manuellen Inspektion von Spezialisten zu vergleichen.

Eine Vielzahl von ständig aktualisierten Algorithmen in der EEG-Aufzeichnungssoftware LabChart® (AD Instruments) half uns bei der Entwicklung einer neuartigen arithmetischen und VB-Makro-basierten Signalverarbeitungs-Methode zur Quantifizierung und Charakterisierung der inter-iktalen Spikes mit hoher Präzision (90-99 %). Obwohl Spikes und Spike-Cluster bei allen Tieren (uninfiziert und infiziert) während der frühen (0-7 Tage post infectionem) und späten Phase (91-97 Tage post infectionem) nach der Infektion beobachtet wurden, wurde festgestellt, dass sie bei TMEV-infizierten Tieren mit Anfällen (früh oder spät) vermehrt auftreten. Die durchschnittliche Anzahl von Spikes und Spike-Clustern während der akuten und chronischen Phase der Epilepsieentwicklung war bei infizierten Tieren signifikant höher als bei den Kontrollen. Eine umfassende Analyse von EEGs, die in der späten Phase nach der Infektion aufgezeichnet wurden, zeigte, dass sich infizierte Mäuse ohne frühe oder späte Anfälle bezüglich der Spike-Häufigkeit nicht von Kontrollen unterschieden, während Mäuse mit jeder Art von Anfall, früh oder spät, signifikant erhöhte Spike- und Spike-Cluster-Frequenzen in der späten Phase zeigten. Im Gegensatz dazu zeigten wenige der epileptischen Tiere eine geringe Anzahl von Spikes und Spike-Clustern, welche im Bereich der Kontrolltiere lag. Dies legt nahe, dass die Häufigkeit von und Spike-Clustern alleine nicht vorhersagen kann, ob ein individuelles Tier epileptisch ist oder nicht.

Dies kann jedoch zusammen mit anderen zuvor identifizierten Surrogatmarkern, wie Anfallsschwelle, Verhalten und kognitiven Studien, verwendet werden, um die Krankheitsentwicklung zu bewerten.

Unsere Ergebnisse legen nahe, dass wir erhöhte Spike- und Spike-Cluster-Häufigkeiten in einer Gruppe von infizierten Tieren als einen neue Parameter für Krankheitsmedikations- oder Antiepileptogenese-Studien verwenden können, um die Wirkungen von pharmakologischen Verbindungen zu bewerten.

47 References

Abad, V. C., & Guilleminault, C. (2003). Diagnosis and treatment of sleep disorders: a brief review for clinicians. Dialogues in Clinical Neuroscience, 5(4), 371-388.

Abend, N. S., Sanchez, S. M., Berg, R. A., Dlugos, D. J., & Topjian, A. A. (2013). Current Treatment of Electrographic Seizures and Status Epilepticus in Critically Ill Children: a Single Center Experience.

Seizure : the journal of the British Epilepsy Association, 22(6), 467-471. doi:

10.1016/j.seizure.2013.03.008

Abou-Khalil, B. W. (2012). Do We Need EEGs After Temporal Lobe Epilepsy Surgery, and How Many?

Epilepsy Currents, 12(1), 29-31. doi: 10.5698/1535-7511-12.1.29

Aldenkamp, A. P., & Arends, J. (2004). Effects of epileptiform EEG discharges on cognitive function: is the concept of “transient cognitive impairment” still valid? Epilepsy & Behavior, 5, 25-34.

American Clinical Neurophysiology Society, A. (2006). Guideline 6: A proposal for standard montages to be used in clinical EEG. Journal of clinical neurophysiology: official publication of the American Electroencephalographic Society, 23(2), 111.

Amzica, F. (2015). What does burst suppression really mean? Epilepsy & Behavior, 49, 234-237.

Arciniegas, D. B. (2011). Clinical electrophysiologic assessments and mild traumatic brain injury: state-of-the-science and implications for clinical practice. International Journal of Psychophysiology, 82(1), 41-52.

doi: 10.1016/j.ijpsycho.2011.03.004

Authier, S., Delatte, M. S., Kallman, M.-J., Stevens, J., & Markgraf, C. (2016). EEG in non-clinical drug safety assessments: Current and emerging considerations. Journal of Pharmacological and Toxicological Methods, 81, 274-285. doi: https://doi.org/10.1016/j.vascn.2016.03.002

Barker-Haliski, Wolfgang, L., Steve, W. H., & S., G. A. (2017). Neuroinflammation in epileptogenesis: Insights and translational perspectives from new models of epilepsy. Epilepsia, 58(S3), 39-47. doi:

doi:10.1111/epi.13785

Barker-Haliski, M. L., Dahle, E. J., Heck, T. D., Pruess, T. H., Vanegas, F., Wilcox, K. S., & White, H. S.

(2015). Evaluating an etiologically relevant platform for therapy development for temporal lobe epilepsy: effects of carbamazepine and valproic acid on acute seizures and chronic behavioral comorbidities in the Theiler’s murine encephalomyelitis virus mouse model. Journal of Pharmacology and Experimental Therapeutics, 353(2), 318-329.

Barker‐Haliski, M. L., Heck, T. D., Dahle, E. J., Vanegas, F., Pruess, T. H., Wilcox, K. S., & White, H. S.

(2016). Acute treatment with minocycline, but not valproic acid, improves long‐term behavioral outcomes in the Theiler's virus model of temporal lobe epilepsy. Epilepsia, 57(12), 1958-1967.

Bauer, G. (1982). EEG, drug effects and central nervous system poisonging. Electroencephalography: basic principles, clinical applications and related fields, 479-489.

Beghi, E., Carpio, A., Forsgren, L., Hesdorffer, D. C., Malmgren, K., Sander, J. W., . . . Hauser, W. A. (2010).

Recommendation for a definition of acute symptomatic seizure. Epilepsia, 51(4), 671-675. doi:

10.1111/j.1528-1167.2009.02285.x

Bekkers, J. M. (2011). Pyramidal neurons. Current Biology, 21(24), R975.

Bell, M. A., & Cuevas, K. (2012). Using EEG to Study Cognitive Development: Issues and Practices. Journal of cognition and development : official journal of the Cognitive Development Society, 13(3), 281-294.

doi: 10.1080/15248372.2012.691143

Berendt, M., Farquhar, R. G., Mandigers, P. J., Pakozdy, A., Bhatti, S. F., De Risio, L., . . . Muñana, K. (2015).

International veterinary epilepsy task force consensus report on epilepsy definition, classification and terminology in companion animals. BMC Veterinary Research, 11(1), 182.

Beres, A. M. (2017). Time is of the Essence: A Review of Electroencephalography (EEG) and Event-Related Brain Potentials (ERPs) in Language Research. Applied Psychophysiology and Biofeedback, 42(4), 247-255. doi: 10.1007/s10484-017-9371-3

Beretta, S., Carone, D., Zanchi, C., Bianchi, E., Pirovano, M., Trentini, C., . . . Group, P.-L. S. (2017). Long-term applicability of the new ILAE definition of epilepsy. Results from the PRO-LONG study.

Epilepsia, 58(9), 1518-1523. doi: 10.1111/epi.13854

Berger, H. (1929). Über das elektrenkephalogramm des menschen. Archiv für Psychiatrie und Nervenkrankheiten, 87(1), 527-570.

Bergstrom, R. A., Choi, J. H., Manduca, A., Shin, H.-S., Worrell, G. A., & Howe, C. L. (2013). Automated identification of multiple seizure-related and interictal epileptiform event types in the EEG of mice.

Scientific Reports, 3, 1483. doi: 10.1038/srep01483

Bhat, S., Acharya, U. R., Adeli, H., Bairy, G. M., & Adeli, A. (2014). Automated diagnosis of autism: in search of a mathematical marker. Reviews in the Neurosciences, 25(6), 851-861.

Binnie, C. D., & Prior, P. F. (1994). Electroencephalography. Journal of Neurology, Neurosurgery and Psychiatry, 57(11), 1308-1319.

Bonello, M., Michael, B. D., & Solomon, T. (2015). Infective Causes of Epilepsy. Seminars in Neurology, 35(3), 235-244. doi: 10.1055/s-0035-1552619

Bourien, J., Bartolomei, F., Bellanger, J. J., Gavaret, M., Chauvel, P., & Wendling, F. (2005). A method to identify reproducible subsets of co-activated structures during interictal spikes. Application to intracerebral EEG in temporal lobe epilepsy. Clinical Neurophysiology, 116(2), 443-455. doi:

10.1016/j.clinph.2004.08.010

Bragin, A., Engel, J., Wilson, C. L., Fried, I., & Mathern, G. W. (1999). Hippocampal and entorhinal cortex high‐frequency oscillations (100–500 Hz) in human epileptic brain and in kainic acid‐treated rats with chronic seizures. Epilepsia, 40(2), 127-137.

Bragin, A., Jando, G., Nadasdy, Z., Van Landeghem, M., & Buzsáki, G. (1995). Dentate EEG spikes and associated interneuronal population bursts in the hippocampal hilar region of the rat. Journal of Neurophysiology, 73(4), 1691-1705.

Bridgers, S. L. (1987). Electroencephalographic Screening of Psychiatric Inpatients. Archives of Neurology, 44, 312-316.

Brodie, M. J., Barry, S. J., Bamagous, G. A., Norrie, J. D., & Kwan, P. (2012). Patterns of treatment response in newly diagnosed epilepsy. Neurology, 78(20), 1548-1554. doi: 10.1212/WNL.0b013e3182563b19 Bröer, S., Hage, E., Käufer, C., Gerhauser, I., Anjum, M., Li, L., . . . Löscher, W. (2017). Viral mouse models of

multiple sclerosis and epilepsy: Marked differences in neuropathogenesis following infection with two naturally occurring variants of Theiler's virus BeAn strain. Neurobiology of Disease, 99, 121-132.

Bröer, S., Kaufer, C., Haist, V., Li, L., Gerhauser, I., Anjum, M., . . . Loscher, W. (2016). Brain inflammation, neurodegeneration and seizure development following picornavirus infection markedly differ among virus and mouse strains and substrains. Experimental Neurology, 279, 57-74. doi:

10.1016/j.expneurol.2016.02.011

Burkholder, D. B., Sulc, V., Hoffman, E. M., Cascino, G. D., Britton, J. W., So, E. L., . . . Worrell, G. A.

(2014). Interictal Scalp Electroencephalography and Intraoperative Electrocorticography in Magnetic Resonance Imaging–Negative Temporal Lobe Epilepsy Surgery. JAMA neurology, 71(6), 702-709. doi:

10.1001/jamaneurol.2014.585

Buzsaki, G. (2006). Rhythms of the Brain: Oxford University Press.

Buzsaki, G., Anastassiou, C. A., & Koch, C. (2012). The origin of extracellular fields and currents--EEG, ECoG, LFP and spikes. Nature Reviews: Neuroscience, 13(6), 407-420. doi: 10.1038/nrn3241

Cambiaghi, M., Magri, L., & Cursi, M. (2015). Importance of EEG in validating the chronic effects of drugs:

Suggestions from animal models of epilepsy treated with rapamycin. Seizure, 27, 30-39. doi:

https://doi.org/10.1016/j.seizure.2015.02.015

Carpentier, P. A., Getts, M. T., & Miller, S. D. (2008). Pro-inflammatory functions of astrocytes correlate with viral clearance and strain-dependent protection from TMEV-induced demyelinating disease. Virology, 375(1), 24-36. doi: https://doi.org/10.1016/j.virol.2008.01.024

Chauviere, L., Doublet, T., Ghestem, A., Siyoucef, S. S., Wendling, F., Huys, R., . . . Bernard, C. (2012).

Changes in interictal spike features precede the onset of temporal lobe epilepsy. Annals of Neurology, 71(6), 805-814. doi: 10.1002/ana.23549

Chauvière, L., Doublet, T., Ghestem, A., Siyoucef, S. S., Wendling, F., Huys, R., . . . Bernard, C. (2012).

Changes in interictal spike features precede the onset of temporal lobe epilepsy. Annals of Neurology, 71(6), 805-814.

Chemaly, N., Nehlig, A., Chiron, C., & Nabbout, R. (2018). Electrocorticographic telemetric recording in unrestrained mouse pups. Journal of Neuroscience Methods.

Chen, Z., Cao, J., Cao, Y., Zhang, Y., Gu, F., Zhu, G., . . . Cichocki, A. (2008). An empirical EEG analysis in brain death diagnosis for adults. Cognitive Neurodynamics, 2(3), 257-271. doi: 10.1007/s11571-008-9047-z

Cherian, A., & Thomas, S. V. (2009). Status epilepticus. Annals of Indian Academy of Neurology, 12(3), 140-153. doi: 10.4103/0972-2327.56312

Chez, M. G., Chang, M., Krasne, V., Coughlan, C., Kominsky, M., & Schwartz, A. (2006). Frequency of epileptiform EEG abnormalities in a sequential screening of autistic patients with no known clinical epilepsy from 1996 to 2005. Epilepsy & Behavior, 8(1), 267-271.

Christensen, J. (2012). Traumatic brain injury: risks of epilepsy and implications for medicolegal assessment.

Epilepsia, 53(s4), 43-47.

Claes, L., Del-Favero, J., Ceulemans, B., Lagae, L., Van Broeckhoven, C., & De Jonghe, P. (2001). De novo mutations in the sodium-channel gene SCN1A cause severe myoclonic epilepsy of infancy. The American Journal of Human Genetics, 68(6), 1327-1332.

Clément, H., Pascal, B., Gabriel, D., Arnaud, B., Benoît, M., & Fabrice, W. (2013). Shape features of epileptic spikes are a marker of epileptogenesis in mice. Epilepsia, 54(12), 2219-2227. doi:

doi:10.1111/epi.12406

Cossu, M., Cardinale, F., Colombo, N., Mai, R., Nobili, L., Sartori, I., & Lo Russo, G. (2005).

Stereoelectroencephalography in the presurgical evaluation of children with drug-resistant focal epilepsy. Journal of Neurosurgery: Pediatrics, 103(4), 333-343.

Crespel, A. (2000). Proposal for revised classification of epilepsies and epileptic syndromes. Clinical Neurophysiology, 111, S54-S59.

Cusick, M. F., Libbey, J. E., Patel, D. C., Doty, D. J., & Fujinami, R. S. (2013). Infiltrating macrophages are key to the development of seizures following virus infection. Journal of Virology, 87(3), 1849-1860. doi:

10.1128/JVI.02747-12

Da Silva, F. L. (2009). EEG: origin and measurement EEg-fMRI (pp. 19-38): Springer.

Daniels, J. B., Pappenheimer, A. M., & Richardson, S. (1952). OBSERVATIONS ON ENCEPHALOMYELITIS OF MICE (DA STRAIN). The Journal of Experimental Medicine, 96(6), 517-530.

DePaula-Silva, A. B., Hanak, T. J., Libbey, J. E., & Fujinami, R. S. (2017). Theiler's murine encephalomyelitis virus infection of SJL/J and C57BL/6J mice: Models for multiple sclerosis and epilepsy. Journal of Neuroimmunology, 308, 30-42. doi: 10.1016/j.jneuroim.2017.02.012

DePaula-Silva, A. B., Sonderegger, F. L., Libbey, J. E., Doty, D. J., & Fujinami, R. S. (2018). The immune response to picornavirus infection and the effect of immune manipulation on acute seizures. Journal of Neurovirology. doi: 10.1007/s13365-018-0636-2

Drury, I., & Beydoun, A. (1998). Interictal epileptiform activity in elderly patients with epilepsy. Clinical Neurophysiology, 106(4), 369-373.

Duncan, J. S. (1987). Antiepileptic drugs and the electroencephalogram. Epilepsia, 28(3), 259-266.

Ebersole, J. S., & Pedley, T. A. (2003). Drug effects and toxic encephalopathies. In A. C. Van Gott & R. P.

Brenner (Eds.), Current practice of clinical electroencephalography, 3rd edn (Vol. 10, pp. 463-482):

Wiley/Blackwell (10.1111).

El-Gohary, M., McNames, J., & Elsas, S.-M. (2008). User-Guided Interictal Spike Detection. Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference, 2008, 821-824. doi:

10.1109/IEMBS.2008.4649280

Engel, J., Pitkänen, A., Loeb, J. A., Dudek, F. E., Bertram, E. H., Cole, A. J., . . . Vezzani, A. (2013).

EPILEPSY BIOMARKERS. Epilepsia, 54(0 4), 61-69. doi: 10.1111/epi.12299

Falco-Walter, J. J., Scheffer, I. E., & Fisher, R. S. (2018). The new definition and classification of seizures and

Im Dokument In-Silico analysis, characterization and quantification of EEG alterations in a mouse model of temporal lobe epilepsy (Seite 51-69)