Uniform TitleA study on adaptive stimulation of the basal ganglia as a treatment for Parkinsonism
NameLeondopulos, Stathis S. (author), Micheli-Tzanakou, Evangelia (chair), Bushnell, Michael (internal member), Gajic, Zoran (internal member), Caggiano, Michael (internal member), Orfanidis, Sophocles (internal member), Nowakowski, Richard (outside member), Rutgers University, Graduate School - New Brunswick,
Degree Date2007
Date Created2007
SubjectElectrical and Computer Engineering,
Brain stimulation
DescriptionThe purpose of this dissertation is to design an automated system for the modification of Deep Brain Stimulation (DBS) parameters based on specific identifiers in the neuronal response of Parkinsonian patients undergoing DBS treatment. The neural response patterns are obtained from an artificial neural network consisting of dynamic neuron and synapse components and programmed to exhibit a response to pulse stimuli that resembles the activity in the subthalamic nucleus of Parkinsonian patients undergoing DBS treatment. Moreover, using pulse stimuli of varying specification, a band-pass filtered response of the network is subjected to a set of signal processing techniques including Linear Predictive Coding (LPC), Autoregressive Moving Average (ARMA) modeling, Discrete Fourier Transform (DFT), moments and higher order statistics, producing a set of results or features. Then, each feature is evaluated to determine the effectiveness, in terms of error probability, of discerning between different neuronal responses to pulse stimuli. Furthermore, a digital circuit is designed at the transistor level for computing the 1st LPC coefficient of recorded neural data and also autonomously regulating the specifications of the stimulus waveform based on the value of the computed coefficient. Also, the circuit design is optimized using a pipeline to reduce dynamic power dissipation. Moreover, it is suggested that a similar design may be useful in automating the administration of DBS as a treatment for Parkinsonism with only a minimal additional power demand.
NotePh.D.
NoteIncludes bibliographical references (p. 181-192).
Genretheses
Persistent URLhttp://hdl.rutgers.edu/1782.2/rucore10001600001.ETD.16722
LanguageEnglish
CollectionGraduate School - New Brunswick Electronic Theses and Dissertations
Organization NameRutgers, The State University of New Jersey
RightsThe author owns the copyright to this work.