腦磁源定位技術(shù)研究.pdf

1、浙江大學(xué)博士學(xué)位論文腦磁源定位技術(shù)研究姓名：霍小林申請學(xué)位級別：博士專業(yè)：生物醫(yī)學(xué)工程指導(dǎo)教師：汪元美2001.3.1浙江大學(xué)博士學(xué)位論文___————●__________●●___●●_______________________________________———_—————————————————ABSTRACTThenoninvasivedetectionofthebrainfunctionispursuedallalong

2、bythehumanbeingMagnetoencephalography(MEG)isbecominganimportantmethodofthediagnosisofhumanbrainfunctionaldiseaseandthedetectionofadvancedcognitiveactionforitsexcellentSpatio—TemporalresolutionsThelocalizationofthecurrent

3、sourceproducingtheexternalmagneticfieldsisthekeyproblemofMEGManyresearchersofMEGintheworldareengagedinfindingmoreeffectivemethodsofsourcelOCalizationBasedontheresearchworkofMEGintheworld，thebasicequationsofelectromagneti

4、csinthefieldofbiomagnetismareelaboratedinthispaperThescopeofapplicationandthedeductionprocessofthemainsourcelocalizationequationsaregivenAnoverviewiSpresentedontheclassicalmethodsoftheMEGforwardandinverseproblemsThescwor

5、ksarebeneficialtesearchtheMEGsystematicallyThesourcelocalizationbelongstotheinverseproblemoftheMEGInfacttheMEGinverscproblemiSi11posedbecausedifferentneuralcurrentdistributionsmayproducethesatneexternalfieldmeasurementsA

6、methodoflocalizationofmultiplecurrentdipolesbasedontIleminimumdipolesolutionsiSstudiedandanewcostfunctionisproposedWhroughthecomputersimulation，thescopesofthemainparametersofthenewmethodaregivenComparedwiththeincremental

7、methodthenewmethodiSemcientto也elocalizationofmultiplecurrentdipolesthroughonlyonceinversionTherevisedsimulatedannealingalgorithmastheoptimizationstrategyisusedinthesimulationoftheinstantaneousmodeloftheMEGandsomegoodresu

8、ltsarereceivedTheiudgmentrulesofthetherrualequilibriumandthebackfirearepresentedTheSpatioTemporalSourceModeI(STSM)hasbecomethemaintrendinthefieldofMEGtodayAmetllodcombiningSpatioTemporalmodelwithSimulatedAnnealingtolocat

9、emultiplecurrentdipolesispresentedthroughstudyingtheSTSMofMEGThismethodcanovercometheshortcomingsofotheroptimalmethodsandavoidbeingtrappedinalocalminimumnledipoleparameterscanbeseparatedintolinearandnonlinearcomponentsTh

10、eoptimizationdimensionscanbereducedgreatlybyoptimizingthenonlinearcomponentsonlyComparedwiththeMUSIC(MUltipleSignalClassification)，thismethodcandecreasetherequirementsfortheindependenceofthedipolesourcescorrespondinglyTh

11、enoisehasagreatinfluencetotheaccuracyoftheMEGsourcelocalizationInthispaper，theproblemofthenoiseprocessingisdiscussedThefieldofspontaneousbrainactivityiSnotrelatedtothefieldofevokedbrainactivityTherefore，suchspontaneousbr