abstract: Approximately 1% of the world population suffers from epilepsy. Continuous long-term electroencephalographic (EEG) monitoring is the gold-standard for recording epileptic seizures and assisting in the diagnosis and treatment of patients with epilepsy. However, this process still requires that seizures are visually detected and marked by experienced and trained electroencephalographers. The motivation for the development of an automated seizure detection algorithm in this research was to assist physicians in such a laborious, time consuming and expensive task. Seizures in the EEG vary in duration (seconds to minutes), morphology and severity (clinical to subclinical, occurrence rate) within the same patient and across patients. The task of seizure detection is also made difficult due to the presence of movement and other recording artifacts. An early approach towards the development of automated seizure detection algorithms utilizing both EEG changes and clinical manifestations resulted to a sensitivity of 70-80% and 1 false detection per hour. Approaches based on artificial neural networks have improved the detection performance at the cost of algorithm's training. Measures of nonlinear dynamics, such as Lyapunov exponents, have been applied successfully to seizure prediction. Within the framework of this MS research, a seizure detection algorithm based on measures of linear and nonlinear dynamics, i.e., the adaptive short-term maximum Lyapunov exponent (ASTLmax) and the adaptive Teager energy (ATE) was developed and tested. The algorithm was tested on long-term (0.5-11.7 days) continuous EEG recordings from five patients (3 with intracranial and 2 with scalp EEG) and a total of 56 seizures, producing a mean sensitivity of 93% and mean specificity of 0.048 false positives per hour. The developed seizure detection algorithm is data-adaptive, training-free and patient-independent. It is expected that this algorithm will assist physicians in reducing the time spent on detecting seizures, lead to faster and more accurate diagnosis, better evaluation of treatment, and possibly to better treatments if it is incorporated on-line and real-time with advanced neuromodulation therapies for epilepsy. / Dissertation/Thesis / M.S. Electrical Engineering 2012

Submitted by Fabio Sobreira Campos da Costa (fabio.sobreira@ufpe.br) on 2017-02-15T12:52:08Z No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Dissertação - Ricardo Batista.pdf: 10061311 bytes, checksum: ce7d296a73fc33cb8f4605b5e94a9cfb (MD5) / Made available in DSpace on 2017-02-15T12:52:08Z (GMT). No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Dissertação - Ricardo Batista.pdf: 10061311 bytes, checksum: ce7d296a73fc33cb8f4605b5e94a9cfb (MD5) Previous issue date: 2015-03-02 / CNPq / Centros de periodicidade e caos (CPCs) s˜ao pontos que podem aparecer quando projetamos certo expoente de Lyapunov λ em um plano de parˆametros de um sistema dinˆamico dissipativo. Espirais de solu¸c˜oes peri´odicas (λ < 0) e ca´oticas (λ > 0) circulam alternadamente um CPC, como aquele no ter¸co inferior direito na figura da folha de rosto. Nesta disserta¸c˜ao foi desenvolvido inicialmente um programa para o c´alculo num´erico do espectro de Lyapunov de um sistema dinˆamico tridimensional (3D) gen´erico. Em seguida, CPCs foram procurados e achados nas solu¸c˜oes das equa¸c˜oes de R¨ossler, que possuem trˆes parˆametros, a, b, e c. Em particular, para b = bc = 0.17872, o CPC foi encontrado no plano a×c com coordenadas a = ac = 0.17694 e c = cc = 10.5706. Fixando a = ac e tomando c como um parˆametro de controle no intervalo 3 < c < cc, uma sequˆencia de dobramentos de per´ıodo seguida por uma sequˆencia de janelas de adi¸c˜ao de per´ıodo dentro da regi˜ao ca´otica. Ajustes por fun¸c˜oes simples de mapas de retorno de m´aximos locais em uma das vari´aveis dinˆamicas do sistema de R¨ossler permitiram a elabora¸c˜ao de um mapa discreto unidimensional Mr(x) no intervalo unit´ario, o qual faz a m´ımica sin´optica da dinˆamica do fluxo. A raz˜ao de convergˆencia para a sequˆencia de adi¸c˜ao de per´ıodo foi estimada dos ciclos superest´aveis do mapa como um valor pouco acima de 1.7, em bom acordo com o que se obt´em do sistema de R¨ossler. Uma f´ormula para a medida invariante foi obtida de um ajuste para a distribui¸c˜ao das iteradas em regime erg´odico. O correspondente expoente de Lyapunov, 0.597, est´a em bom acordo com 0.588, valor obtido da m´edia discreta de ln|Mr(xi)|. / Aperiodicityhub(PH)isthecommoncenterofperiodic(λ < 0)andchaotic(λ > 0) spirals which show up when a characteristic Lyapunov exponent λ of a dissipative dynamical system is projected onto a planar subset of its parameter space. The color plate in a previous page of this document shows one such PH in the lower right third. In this work Lyapunov spectra of three-dimensional dynamical systems were numericallycalculatedwithastandardalgorithmwhichreliesonrepeatedapplication of the Gram-Schmidt orghonormalization procedure on certain vectors in the phase space. PHs were then searched and found in the R¨ossler system, which has three parameters, namely, a,b, and c. In particular, for b = bh = 0.17872, a PH was found in the ca-plane with coordinates a = ah = 0.17694 and c = ch = 10.5706. By fixing a = ah and taking c as a control parameter in the interval 3 < c < ch, a complete sequence , i.e., a period-doubling sequence followed by a sequence of period-adding windows within the chaotic region, was observed. Fits to tens of return maps for local maxima in one of the dynamical variables allowed the construction of a oneparameter one-dimensional discrete map in the unit interval that synoptically mimics the dynamics of the flow. The convergence ratio for the period-adding sequence was estimated from the superstable cycles as 1.7, in good agreement with the value obtained from the R¨ossler system. At full ergodicity, a formula for the invariant measurewasobtainedfromafittothedistributionoftheiterates. Fromthatformula, we estimated a Lyapunov exponent of 0.597, which is in reasonable agreement with 0.588, the value obtained straightforwardly from the discrete iterates of the map.

京都大学 / 0048 / 新制・課程博士 / 博士(情報学) / 甲第22585号 / 情博第722号 / 新制||情||124(附属図書館) / 京都大学大学院情報学研究科数理工学専攻 / (主査)教授 梅野 健, 教授 山下 信雄, 教授 守倉 正博 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

京都大学 / 新制・課程博士 / 博士(情報学) / 甲第24742号 / 情博第830号 / 新制||情||139(附属図書館) / 京都大学大学院情報学研究科数理工学専攻 / (主査)教授 梅野 健, 教授 太田 快人, 准教授 辻本 諭 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

No description available.

Bewegungsstabilität wird durch die Fähigkeit des neuromuskulären Systems adäquat auf Störungen der Bewegung antworten zu können erreicht. Einschränkungen der Stabilität werden z. B. mit Sturzrisiko in Verbindung gebracht, was schwere Konsequenzen für die Lebensqualität und Kosten im Gesundheitssystem hat. Nach wie vor wird debattiert, wie eine geeignete Bewertung von Stabilität vorgenommen werden kann. Diese Arbeit behandelt den maximalen Lyapunov Exponenten. Er drückt aus, wie sensitiv das System auf kleine Störungen eines Zustands reagiert. Eine Zeitreihe wird zunächst mittels zeitversetzter Kopien in einen mehrdimensionalen Raum eingebettet. In dieser rekonstruierten Dynamik berechnet man dann die Steigung der mittleren logarithmischen Divergenz initial naher Punkte. Die methodischen Konsequenzen für die Anwendung dieser Systemtheorie auf Bewegungen sind jedoch bislang unzureichend beleuchtet. Der experimentelle Teil zeigt klare Indizien, dass es bei Bewegungen weniger um die Analyse eines komplexen Systemdeterminismus geht, sondern um verschieden hohe dynamische Rauschlevel. Je höher das Rauschlevel, desto instabiler das System. Anwendung von Rauschreduktion führt zu kleineren Effektstärken. Das hat Folgen: Die Funktionswerte der Average Mutual Information, die bisher nur zur Bestimmung des Zeitversatzes genutzt wurden, können bereits Unterschiede in der Stabilität zeigen. Die Abschätzung der Dimension für die Einbettung (unabhängig vom verwendeten Algorithmus), ist stark von der Länge der Zeitreihe abhängig und wird bisher eher überschätzt. Die größten Effekte sind in Dimension drei zu beobachten und ein sehr früher Bereich zur Auswertung der Divergenzkurve ist zu empfehlen. Damit wird eine effiziente und standardisierte Analyse vorgeschlagen, die zudem besser imstande ist, Unterschiede verschiedener Bedingungen oder Gruppen aufzuzeigen. / Reductions of movement stability due to impairments of the motor system to respond adequately to perturbations are associated with e. g. the risk of fall. This has consequences for quality of life and costs in health care. However, there is still an debate on how to measure stability. This thesis examines the maximum Lyapunov exponent, which became popular in sports science the last two decades. The exponent quantifies how sensitive a system is reacting to small perturbations. A measured data series and its time delayed copies are embedded in a moredimensional space and the exponent is calculated with respect to this reconstructed dynamic as average slope of the logarithmic divergence curve of initially nearby points. Hence, it provides a measure on how fast two at times near trajectories of cyclic movements depart. The literature yet shows a lack of knowledge about the consequences of applying this system theory to sports science tasks. The experimental part shows strong evidence that, in the evaluation of movements, the exponent is less about a complex determinism than simply the level of dynamic noise present in time series. The higher the level of noise, the lower the stability of the system. Applying noise reduction therefore leads to reduced effect sizes. This has consequences: the values of average mutual information, which are until now only used for calculating the delay for the embedding, can already show differences in stability. Furthermore, it could be shown that the estimation of the embedding dimension d (independently of algorithm), is dependent on the length of the data series and values of d are currently overestimated. The greatest effect sizes were observed in dimension three and it can be recommended to use the very first beginning of the divergence curve for the linear fit. These findings pioneer a more efficient and standardized approach of stability analysis and can improve the ability of showing differences between conditions or groups.

Este trabalho trata das medidas de máxima entropia para certos difeomorfismos em nilvariedades. Considere um difeomorfismo parcialmente hiperbólico f definido em T3, dinamicamente coerente com folheação central compacta. Suponha ainda que a aplicação induzida por f no espaço das folhas centrais é um homeomorfismo de Anosov transitivo em T2. Mostramos que o conjunto das medidas ergódicas hiperbólicas de máxima entropia é enumerável. Usando o princípio de invariância, mostramos que se o primeiro retorno de f à alguma folha periódica tem número de rotação irracional, então, f tem no máximo duas medidas ergódicas de máxima entropia e ter apenas uma medida de máxima entropia equivale a ser extensão de rotação. Se a aplicação de primeiro retorno à alguma folha central periódica é Morse-Smale, então existe um su-toro periódico, ou temos uma cota superior para o número de medidas ergódicas de máxima entropia que depende do número de atratores da dinâmica nessa folha. Além disso, estudamos a topologia da bacia das medidas ergódicas de máxima entropia para uma outra classe de difeomorfismos especiais que são genéricos no espaço dos difeomorfismos absolutamente parcialmente hiperbólicos e denotada por SPH1(M). / This work is about maximal entropy measures for certain diffeomorphisms on nilmanifolds. Consider a partially hyperbolic diffeomorphism f on T3 , C2 , dinamically coherent with compact center foliation which is a circle bundle. Assume that the map induced by f on the space of center leaves is a transitive Anosov homeomorphism. We show that the set of hyperbolic ergodic maximal entropy measures of f is countable. Using the invariance principle, we show that if the first return map to some periodic leaf has irrational rotation number then f has at most two ergodic maximal entropy measures and, in this case, if f has only one maximal entropy measure then f is a rotation extension. If the first return map to some periodic leaf is Morse-Smale then either there exists some periodic su-torus or an upper bound for the number of ergodic maximal entropy measure depending on the number of the attractors of the dynamics in this leaf. Moreover, we study the topology of basin of ergodic maximal entropy measures of another set of special diffeomorphisms that are generic in the space of absolutely partially hyperbolic systems and denoted by SPH1(M).

Este trabalho trata das medidas de máxima entropia para certos difeomorfismos em nilvariedades. Considere um difeomorfismo parcialmente hiperbólico f definido em T3, dinamicamente coerente com folheação central compacta. Suponha ainda que a aplicação induzida por f no espaço das folhas centrais é um homeomorfismo de Anosov transitivo em T2. Mostramos que o conjunto das medidas ergódicas hiperbólicas de máxima entropia é enumerável. Usando o princípio de invariância, mostramos que se o primeiro retorno de f à alguma folha periódica tem número de rotação irracional, então, f tem no máximo duas medidas ergódicas de máxima entropia e ter apenas uma medida de máxima entropia equivale a ser extensão de rotação. Se a aplicação de primeiro retorno à alguma folha central periódica é Morse-Smale, então existe um su-toro periódico, ou temos uma cota superior para o número de medidas ergódicas de máxima entropia que depende do número de atratores da dinâmica nessa folha. Além disso, estudamos a topologia da bacia das medidas ergódicas de máxima entropia para uma outra classe de difeomorfismos especiais que são genéricos no espaço dos difeomorfismos absolutamente parcialmente hiperbólicos e denotada por SPH1(M). / This work is about maximal entropy measures for certain diffeomorphisms on nilmanifolds. Consider a partially hyperbolic diffeomorphism f on T3 , C2 , dinamically coherent with compact center foliation which is a circle bundle. Assume that the map induced by f on the space of center leaves is a transitive Anosov homeomorphism. We show that the set of hyperbolic ergodic maximal entropy measures of f is countable. Using the invariance principle, we show that if the first return map to some periodic leaf has irrational rotation number then f has at most two ergodic maximal entropy measures and, in this case, if f has only one maximal entropy measure then f is a rotation extension. If the first return map to some periodic leaf is Morse-Smale then either there exists some periodic su-torus or an upper bound for the number of ergodic maximal entropy measure depending on the number of the attractors of the dynamics in this leaf. Moreover, we study the topology of basin of ergodic maximal entropy measures of another set of special diffeomorphisms that are generic in the space of absolutely partially hyperbolic systems and denoted by SPH1(M).

L'objectif de cette thèse est de traiter de deux aspects différents de la théorie de l'exposant de Lyapunov de cocycles de Schrödinger définis par une dynamique ergodique. Dans la première partie, on s'intéresse aux estimées de grandes déviations de type Bourgain & Goldstein pour des cocycles quasi-périodiques, puis pour ceux définis par le doublement de l'angle. Après avoir montré que seule une estimée par dessus sur une bande complexe est nécessaire pour avoir la minoration, on redémontre cette inégalité pour une dynamique quasi-périodique en utilisant des techniques de mouvement brownien en lien avec des fonctions sous-harmoniques. Ensuite on adapte la méthode au cas du doublement de l'angle pour lequel on prouve des estimées de grandes déviations sur les branches inverses de cette dynamique. Dans la deuxième partie sont étudiés des cocycles de Schrödinger dont la dynamique est une somme de dynamiques quasi-périodique et aléatoire. On démontre que, dans le régime perturbatif, les développements asymptotiques de l'exposant de Lyapunov attaché à ces cocycles sont similaires à ceux déjà démontrés dans le cas aléatoire par Figotin & Pastur ou Sadel & Schulz-Baldes. L'analyse se fait en fonction du caractère diophantien ou résonant de l'énergie par rapport à la fréquence diophantienne de la partie quasi-périodique du potentiel. / In this thesis we are interested in the Lyapunov exponent of ergodic Schrödinger cocycles. These cocycles occur in the analysis of solutions to the Schrödinger equation where the potential is defined with ergodic dynamics. We study two distinct aspects related to the the Lyapunov exponent for different kinds of dynamics. First we focus on a large deviation theorem for quasi-periodic cocycles and then for potentials defined by the doubling map. We prove that estimates of Bourgain & Goldstein type are granted if an upper estimate involved in the theorem is true on a strip of the complex plane. Then we establish a new technique to prove this upper bound in the quasi-periodic setting, based on subharmonic arguments suggested by Avila, Jitomirskaya & Sadel. We adapt afterwards the method to the doubling map and prove a large deviation theorem for the inverse branches of this dynamics. In the second part, we establish an asymptotic development similar to the results of Figotin & Pastur and Sadel & Schulz-Baldes for the Lyapunov exponent of Schrödinger cocycles at small coupling when the potential is a mixture of quasi-periodic and random. The analysis distinguishes the cases when the energy is either diophantine or resonant with respect to the frequency of the quasi-periodic part of the potential.

The first part of this work gives an introduction into aperiodic order in general and the lines of research pursued. The second part consists of eight manuscripts.