420 likes | 533 Views
Moderne nicht-invasive Methoden zur Erforschung des menschlichen Gehirns. Priv.-Doz. Dr. Carsten Wolters Dr.rer.nat. Harald Kugel Dr.med. Gabriel Möddel Priv.Doz. Dr. med. Christoph Kellinghaus. Vorlesung, 15.Oktober 2013. Outline.
E N D
Moderne nicht-invasive Methoden zur Erforschung des menschlichen Gehirns Priv.-Doz. Dr. Carsten Wolters Dr.rer.nat. Harald Kugel Dr.med. Gabriel Möddel Priv.Doz. Dr. med. Christoph Kellinghaus Vorlesung, 15.Oktober 2013
Outline • General planning for this lecture (language? date/time? required knowledge? Participants-Email-List!) • Literature for this lecture • Introduction to the lecture (Part 1)
Aktuelle Vorlesungsplanung • 15.Oktober: Vorbesprechung und Motivation (Wolters) • 22.Oktober: Einführung Magnetresonanztomographie (MRT) (Kugel) • 29.Oktober: Medizinische Grundlagen zur Elektro- (EEG) und Magnetoencephalography (MEG) (Wolters) • 5.Nov.: Mathematisch-physikalische Modellierungsgrundlagen zu EEG und MEG, Teil 1 (Wolters) • 12.Nov.: Mathematisch-physikalische Modellierungsgrundlagen zu EEG und MEG, Teil 2 (Wolters) • 19.Nov.: Grundlagen von Epilepsie und EEG (Kellinghaus) • 26.Nov.: Epileptische Anfälle und ihre Behandlung (Kellinghaus) • 3.Dez.: Registrierung von MRT: Teil 1 (Wolters) • 10.Dez3.: Registrierung von MRT: Teil 2 (Wolters) • 17.Dez.: Segmentierung von MRT (Wolters) • 7.Jan.: Mathematik des EEG/MEG Vorwärtsproblems, Teil 1 (Wolters) • 14.Jan.: Mathematik des EEG/MEG Vorwärtsproblems, Teil 2 (Wolters) • 21.Jan.: Mathematik des EEG/MEG inversen Problems, Teil 1 (Wolters) • 28.Jan.: Mathematik des EEG/MEG inversen Problems, Teil 2 (Wolters) • 4.Feb.: Epilepsiechirurgie, Teil 3 (Möddel)
Outline • Literature for this lecture • Introduction to the lecture (Part 1)
Literature for this lecture • Lecture webside: http://www.sci.utah.edu/~wolters/LiteraturZurVorlesung/
Outline • Literature for this lecture • Introduction to the lecture (Part 1)
Electro- (EEG) and Magneto-encephalography (MEG) 275 channel axial gradiometer whole-cortex MEG 128 channel EEG
Spatial and temporal resolution of brain imaging methods [Gazzaniga, Ivry & Mangun, Cognitive Neuroscience, 2nd ed., W.W.Norton & Company, 2002]
Grundlagen klinischer EEG und MEG Anwendungen => Warum also MEG? • EEG ist Standard in der klinischen Praxis • MEG ist kostenintensiv (Gerätekosten, Wartung, Heliumkühlung…) • Datenauswertung ist komplex (wie auch für EEG, fMRT, …) • In Deutschland bisher keine Vergütung durch die Krankenkassen
Grundlagen - MEG MEG registriert nicht-invasiv magnetische Felder neuronaler Aktivität Ähnlich dem EEG: Ableitung neuronaler Aktivität MEG und EEG messen Aktivität derselben Generatoren PET oder fMRT: Indirekte Erfassung neuronaler Aktivität 4D Neuroimaging, San Diego, CA, USA
Erfassung des magnetischen Flusses Magnetometer Superconducting quantum interference device (SQUID) Axiales Gradiometer Planares Gradiometer Papanicolaou (Ed.): Clinical Magnetoencephalography and Magnetic Source Imaging
[Lanfer, diploma thesis, 2007] MEG-System am IBB, Uni Münster Finite Elemente Knoten für die MEG Sensor-Beschreibung
Source analysis in presurgical epilepsy diagnosis • 0.5%-1% of world population suffers from epilepsy • 70-80% of patients successfully treated with drugs • For those who are still pharma-resistent after 2-3 drugs • Probability of success of a further different drug: 6% (Wiebe et al 2001) • Probability of success of a surgical treatment: 50% (Wiebe et al 2001) • Indispensable prerequisite for surgery: Focal epilepsy->Localization • Gold standard: Video-monitoring and visual inspection of the EEG (Wilson 1996) • MRI: Identification of an underlying lesion • PET and Neuropsychology: Localization of a functional deficit • Source analysis of • EEG seizure (ictal) activity(Plummer et al., 2008) • EEG/MEG interictal activity: “irritative zone” (Stefan et. al., 2003)
Epileptic spikes in EEG and MEG Clear spike in EEG Nearly no/no signal in MEG
Epileptic spikes in EEG and MEG • Clear spike in EEG • Nearly no/no signal in MEG • Deep source • Strongly radially oriented source
MEG registers mainly tangential source components: Sulci-walls: tangential pyramidal cells -> High amplitudes „Diagonal“ orientation-> Medium amplitude Radial sources hardly produce an MEG: Depth and crown of sulci: radial pyramidal cells -> Low contribution Sensitivity for radial and tangential sources
Epileptic spikes in EEG and MEG Clear signal in MEG, poor signal in EEG Explanation?
Sensitivity Sensitivity EEG > MEG in deep areas But: Sensitivity MEG > EEG in superficial areas Goldenholz et al., 2009
Spikes in EEG and MEG What should we use? MEG instead of EEG? Only EEG? Iwasaki et al., 2005
Combined EEG and MEG 275 channel axial gradiometer whole-cortex MEG 128 channel EEG
Source analysis of interictal spikes in presurgical epilepsy diagnosis
Averaged interictal EEG spikes [Wolters & Kellinghaus, 2006] Measure EEG and/or MEG
Results of combined EEG/MEG dipole fit [Wolters & Kellinghaus, 2006] EEG data and (transparent) cortex MEG data and (transparent) cortex Inverse method: Single current dipole
Results of combined EEG/MEG L1 norm current density reconstruction [Wolters & Kellinghaus, 2006] EEG data and (nontransparent) cortex MEG data and (nontransparent) cortex Inverse method: L1 norm current density
Source analysis of seizure (ictal) spikes in presurgical epilepsy diagnosis
[Gazzaniga, Ivry & Mangun, Cognitive Neuroscience, 2nd ed., W.W.Norton & Company, 2002] Typical EEG signals Gamma(30-70Hz): Starke Konzentr., Lernphase Beta (14-30Hz): Hellwach, gute Intelligenzleistung Alpha (8-13Hz): Entspannte Wachheit Theta (4-7Hz): Leichte Schlafphasen Delta (0.3-3.5Hz): Traumlose Tiefschlafphase
EEG Preprocessing [Rullmann, Anwander, Dannhauer, Warfield, Duffy & Wolters, NeuroImage, 44(2), 2009]
[Rullmann, Anwander, Dannhauer, Warfield, Duffy & Wolters, NeuroImage, 44(2), 2009] T1 MRI segmentation
[Rullmann, Anwander, Dannhauer, Warfield, Duffy & Wolters, NeuroImage, 44(2), 2009] FE mesh generation
Brain conductivity anisotropy modeling [Rullmann, Anwander, Dannhauer, Warfield, Duffy & Wolters, NeuroImage, 44(2), 2009] FA map after registration Original DTI data FA map on T1-MRI Effective medium approach model (DTI <-> CTI): Model DTI<->Conductivity Tensor Image (CTI)[Tuch et al., Ann. NYAS, 1999] Linear model DTI<->CTI [Tuch et al., PNAS, 2001] Validation of DTI<->CTI model in silk yarn phantom[Oh et al., ISMRM, 2006]
Presurgical EEG source analysis [Rullmann, Anwander, Dannhauer, Warfield, Duffy & Wolters, NeuroImage, 44(2), 2009] Goal function scan (Mosher, 1992; Knösche, 1997) MNLS (Hämäläinen & Ilmoniemi, 1984) sLORETA (Pascual-Marqui, 2002) Dipole fit (Scherg and von Cramon, 1985) Result: Behind the lesion in lateral premotor cortex
[Rullmann, Anwander, Dannhauer, Warfield, Duffy & Wolters, NeuroImage, 44(2), 2009] Validation: Intracranial EEG (iEEG)
[Rullmann, Anwander, Dannhauer, Warfield, Duffy & Wolters, NeuroImage, 44(2), 2009] CT and iEEG electrode positions
[Rullmann, Anwander, Dannhauer, Warfield, Duffy & Wolters, NeuroImage, 44(2), 2009] Validation result (localization) iEEG peaking electrodes sEEG Dipole fit result
[Rullmann, Anwander, Dannhauer, Warfield, Duffy & Wolters, NeuroImage, 44(2), 2009] Validation result (orientation) sEEG dipole fit result: Source orientation away from the lesion towards the epileptogenic tissue (Salayev et al., 2006; Plummer et al., 2008)