Wnt-3a and Wnt-3 differently stimulate proliferation and neurogenesis of spinal neural precursors and promote neurite outgrowth by canonical signaling

Wnt factors regulate neural stem cell development and neuronal connectivity. Here we investigated whether Wnt-3a and Wnt-3, expressed in the developing spinal cord, regulate proliferation and the neuronal differentiation of spinal cord neural precursors (SCNP). Wnt-3a promoted a sustained increase of SCNP proliferation, whereas Wnt-3 enhanced SCNP proliferation transiently and increased neurogenesis through β-catenin signaling. Consistent with this, Wnt-3a and Wnt-3 differently regulate the expression of Cyclin-dependent kinase inhibitors. Furthermore, Wnt-3a and Wnt-3 stimulated neurite outgrowth in SCNP-derived neurons through ß-catenin and TCF4-dependent transcription. GSK-3ß inhibitors mimicked Wnt signaling and promoted neurite outgrowth in established cultures. We conclude that Wnt-3a and Wnt-3 signal through the canonical Wnt/β-catenin pathway to regulate different aspects of SCNP development. These findings may be of therapeutic interest for the treatment of neurodegenerative diseases and nerve injury.

Disruption of embryonic blood-CSF barrier in chick embryos reveals the actual importance of this barrier to control E-CSF composition and homeostasis in early brain development

In vertebrates, early brain development takes place at the expanded anterior end of the neural tube. After closure of the anterior neuropore, the brain wall forms a physiologically sealed cavity that encloses embryonic cerebrospinal fluid (E-CSF), a complex and protein-rich fluid that is initially composed of trapped amniotic fluid. E-CSF has several crucial roles in brain anlagen development. Recently, we reported the presence of transient blood-CSF barrier located in the brain stem lateral to the ventral midline, at the mesencephalon and prosencephalon level, in chick and rat embryos by transporting proteins, water, ions and glucose in a selective manner via transcellular routes. To test the actual relevance of the control of E-CSF composition and homeostasis on early brain development by this embryonic blood-CSF barrier, we block the activity of this barrier by treating the embryos with 6-aminonicotinamide gliotoxin (6-AN). We demonstrate that 6-AN treatment in chick embryos blocks protein transport across the embryonic blood-CSF barrier, and that the disruption of the barrier properties is due to the cease transcellular caveolae transport, as detected by CAV-1 expression cease. We also show that the lack of protein transport across the embryonic blood-CSF barrier influences neuroepithelial cell survival, proliferation and neurogenesis, as monitored by neurepithelial progenitor cells survival, proliferation and neurogenesis. The blockage of embryonic blood-CSF transport also disrupts water influx to the E-CSF, as revealed by an abnormal increase in brain anlagen volume. These experiments contribute to delineate the actual extent of this blood-CSF embryonic barrier controlling E-CSF composition and homeostasis and the actual important of this control for early brain development, as well as to elucidate the mechanism by which proteins and water are transported thought transcellular routes across the neuroectoderm, reinforcing the crucial role of E-CSF for brain development.

An Advocacy for the Use of 3D Stem Cell Culture Systems for the Development of Regenerative Medicine: An Emphasis on Photoreceptor Generation

The availability of stem cells is of great promise to study early developmental stages and to generate adequate cells for cell transfer therapies. Although many researchers using stem cells were successful in dissecting intrinsic and extrinsic mechanisms and in generating specific cell phenotypes, few of the stem cells or the differentiated cells show the capacity to repair a tissue. Advances in cell and stem cell cultivation during the last years made tremendous progress in the generation of bona fide differentiated cells able to integrate into a tissue after transplantation, opening new perspectives for developmental biology studies and for regenerative medicine. In this review, we focus on the main works attempting to create in vitro conditions mimicking the natural environment of CNS structures such as the neural tube and its development in different brain region areas including the optic cup. The use of protocols growing cells in 3D organoids is a key strategy to produce cells resembling endogenous ones. An emphasis on the generation of retina tissue and photoreceptor cells is provided to highlight the promising developments in this field. Other examples are presented and discussed, such as the formation of cortical tissue, the epithelial gut or the kidney organoids. The generation of differentiated tissues and well-defined cell phenotypes from embryonic stem (ES) cells or induced pluripotent cells (iPSCs) opens several new strategies in the field of biology and regenerative medicine. A 3D organ/tissue development in vitro derived from human cells brings a unique tool to study human cell biology and pathophysiology of an organ or a specific cell population. The perspective of tissue repair is discussed as well as the necessity of cell banking to accelerate the progress of this promising field.

GLUT2-mediated glucose uptake and availability are required for embryonic brain development in zebrafish

Glucose transporter 2 (GLUT2; gene name SLC2A2) has a key role in the regulation of glucose dynamics in organs central to metabolism. Although GLUT2 has been studied in the context of its participation in peripheral and central glucose sensing, its role in the brain is not well understood. To decipher the role of GLUT2 in brain development, we knocked down slc2a2 (glut2), the functional ortholog of human GLUT2, in zebrafish. Abrogation of glut2 led to defective brain organogenesis, reduced glucose uptake and increased programmed cell death in the brain. Coinciding with the observed localization of glut2 expression in the zebrafish hindbrain, glut2 deficiency affected the development of neural progenitor cells expressing the proneural genes atoh1b and ptf1a but not those expressing neurod. Specificity of the morphant phenotype was demonstrated by the restoration of brain organogenesis, whole-embryo glucose uptake, brain apoptosis, and expression of proneural markers in rescue experiments. These results indicate that glut2 has an essential role during brain development by facilitating the uptake and availability of glucose and support the involvement of glut2 in brain glucose sensing.

Influence of the neural tube/notochord complex on MyoD expression and cellular proliferation in chicken embryos

Important advances have been made in understanding the genetic processes that control skeletal muscle formation. Studies conducted on quails detected a delay in the myogenic program of animals selected for high growth rates. These studies have led to the hypothesis that a delay in myogenesis would allow somitic cells to proliferate longer and consequently increase the number of embryonic myoblasts. To test this hypothesis, recently segmented somites and part of the unsegmented paraxial mesoderm were separated from the neural tube/notochord complex in HH12 chicken embryos. In situ hybridization and competitive RT-PCR revealed that MyoD transcripts, which are responsible for myoblast determination, were absent in somites separated from neural tube/notochord (1.06 and 0.06 10-3 attomol MyoD/1 attomol ß-actin for control and separated somites, respectively; P<0.01). However, reapproximation of these structures allowed MyoD to be expressed in somites. Cellular proliferation was analyzed by immunohistochemical detection of incorporated BrdU, a thymidine analogue. A smaller but not significant (P = 0.27) number of proliferating cells was observed in somites that had been separated from neural tube/notochord (27 and 18 for control and separated somites, respectively). These results confirm the influence of the axial structures on MyoD activation but do not support the hypothesis that in the absence of MyoD transcripts the cellular proliferation would be maintained for a longer period of time.

Pattern of Wnt ligand expression during chick eye development

The dorsoventral axis of the eye is determined prior to optic cup invagination. A variety of signaling pathways have been implicated in the maintenance of the optic dorsoventral axis, including, but not limited to, bone morphogenetic protein 4, Sonic Hedgehog and retinoic acid. Here, we investigated the possible contribution of Wnt ligands to the establishment or maintenance of the optic axis by analyzing their expression pattern during early chick optic development. We performed in situ hybridization of Wnt-1, Wnt-3a, Wnt-4, and Wnt-5a during the optic vesicle, early optic cup and established optic cup stages and focused our analysis on the optic region. Our data showed that Wnt-5a, but none of the others, is expressed in the dorsal region of the eye starting from the Hamburger and Hamilton stage 14 (HH14). These results are supported by cryosections of the labeled optic region, which further reveal that Wnt-5a is expressed only in the dorsal retinal pigmented epithelium. Thus, we propose that Wnt-5a is a marker for dorsal retinal pigmented epithelium in chick embryos from HH14 to HH19.

Cross-Linguistic Transfer (CLT) in Bilingual Speakers : Neural Correlates of Language Learning

Le but de cette thèse est d'étudier les corrélats comportementaux et neuronaux du transfert inter-linguistique (TIL) dans l'apprentissage d’une langue seconde (L2). Compte tenu de nos connaissances sur l'influence de la distance linguistique sur le TIL (Paradis, 1987, 2004; Odlin, 1989, 2004, 2005; Gollan, 2005; Ringbom, 2007), nous avons examiné l'effet de facilitation de la similarité phonologique à l’aide de la résonance magnétique fonctionnelle entre des langues linguistiquement proches (espagnol-français) et des langues linguistiquement éloignées (persan-français). L'étude I rapporte les résultats obtenus pour des langues linguistiquement proches (espagnol-français), alors que l'étude II porte sur des langues linguistiquement éloignées (persan-français). Puis, les changements de connectivité fonctionnelle dans le réseau langagier (Price, 2010) et dans le réseau de contrôle supplémentaire impliqué dans le traitement d’une langue seconde (Abutalebi & Green, 2007) lors de l’apprentissage d’une langue linguistiquement éloignée (persan-français) sont rapportés dans l’étude III. Les résultats des analyses d’IRMF suivant le modèle linéaire général chez les bilingues de langues linguistiquement proches (français-espagnol) montrent que le traitement des mots phonologiquement similaires dans les deux langues (cognates et clangs) compte sur un réseau neuronal partagé par la langue maternelle (L1) et la L2, tandis que le traitement des mots phonologiquement éloignés (non-clang-non-cognates) active des structures impliquées dans le traitement de la mémoire de travail et d'attention. Toutefois, chez les personnes bilingues de L1-L2 linguistiquement éloignées (français-persan), même les mots phonologiquement similaires à travers les langues (cognates et clangs) activent des régions connues pour être impliquées dans l'attention et le contrôle cognitif. Par ailleurs, les mots phonologiquement éloignés (non-clang-non-cognates) activent des régions usuellement associées à la mémoire de travail et aux fonctions exécutives. Ainsi, le facteur de distance inter-linguistique entre L1 et L2 module la charge cognitive sur la base du degré de similarité phonologiques entres les items en L1 et L2. Des structures soutenant les processus impliqués dans le traitement exécutif sont recrutées afin de compenser pour des demandes cognitives. Lorsque la compétence linguistique en L2 augmente et que les tâches linguistiques exigent ainsi moins d’effort, la demande pour les ressources cognitives diminue. Tel que déjà rapporté (Majerus, et al, 2008; Prat, et al, 2007; Veroude, et al, 2010; Dodel, et al, 2005; Coynel, et al ., 2009), les résultats des analyses de connectivité fonctionnelle montrent qu’après l’entraînement la valeur d'intégration (connectivité fonctionnelle) diminue puisqu’il y a moins de circulation du flux d'information. Les résultats de cette recherche contribuent à une meilleure compréhension des aspects neurocognitifs et de plasticité cérébrale du TIL ainsi que l'impact de la distance linguistique dans l'apprentissage des langues. Ces résultats ont des implications dans les stratégies d'apprentissage d’une L2, les méthodes d’enseignement d’une L2 ainsi que le développement d'approches thérapeutiques chez des patients bilingues qui souffrent de troubles langagiers.

Identification et caractérisation d’une souris mutante Skam26Jus comme un nouveau modèle des anomalies du tube neural

Les anomalies du tube neural (ATN) sont des malformations congénitales très fréquentes chez l’humain en touchant 1-2 nouveau-nés sur 1000 naissances. Elles résultent d’une fermeture incomplète du tube neural lors de l’embryogenèse. L’étiologie des ATN est complexe impliquant des facteurs environnementaux et des facteurs génétiques. La souris représente un outil puissant afin de mieux comprendre la génétique des ATN. Particulièrement, la souris modèle a impliqué fortement la voie de la polarité cellulaire planaire (PCP) dans ces malformations. Dans cette étude, nous avons identifié et caractérisé une nouvelle souris mutante, Skam26Jus dans le but d’identifier un nouveau gène causant les ATN. Skam26Jus a été générée par l’agent mutagène N-Ethyl-N-Nitrosuera. Cette souris est caractérisée par une queue en forme de boucle ou de crochet, soit un phénotype associé aux ATN. La complémentation génétique de la souris Skam26Jus avec une souris mutante d’un gène de la voie PCP Vangl2 (Looptail) a montré une interaction génétique entre le gène muté chez Skam26Jus et Vangl2, suggérant que ces deux gènes fonctionnent dans des voies de signalisation semblables ou parallèles. Un total de 50% des embryons doubles hétérozygotes avec un phénotype de la queue présentent un spina bifida. La cartographie par homozygotie du génome entier suivie par un clonage positionnel a permis d’identifier Lrp6 comme le gène muté chez Skam26Jus. Une mutation homozygote, p.Ile681Arg, a été identifiée dans Lrp6 chez les souris ayant une queue en boucle/crochet. Cette mutation était absente dans 30 souches génétiques pures indiquant que cette mutation est spécifique au phénotype observé. Une étude de phénotype-génotype évalue la pénétrance à 53 % de la mutation Ile681Arg. Lrp6 est connu pour activer la voie canonique Wnt/β-caténine et inhiber la voie non canonique Wnt/PCP. Le séquençage de la région codante et de la jonction exon-intron de LRP6 chez 268 patients a mené à l’identification de quatre nouvelles rares mutations faux sens absentes chez 272 contrôles et de toutes les bases de données publiques. Ces mutations sont p.Tyr306His ; p.Tyr373Cys ; p.Val1386Ile; p.Tyr1541Cys et leur pathogénicité prédite in silico indiquent que p.Val1386Ile est bénigne, et que p.Tyr306Hiset p.Tyr373Cys et p.Tyr1541Cys sont i possiblement dommageables. Les mutations p.Tyr306His, p.Tyr373Cys et p.Tyr1541Cys ont affecté l’habilité de LRP6 d’activer la voie Wnt/β-caténine en utilisant le système rapporteur luciférase de pTOPflash. Nos résultats suggèrent que LRP6 joue un rôle dans le développement des ATN chez une petite fraction de patients ayant une ATN. Cette étude présente aussi Skam26Jus comme un nouveau modèle pour étudier les ATN chez l’humain et fournit un outil important pour comprendre les mécanismes moléculaires à l’origine des A TN.

Modeling High-Dimensional Audio Sequences with Recurrent Neural Networks

Cette thèse étudie des modèles de séquences de haute dimension basés sur des réseaux de neurones récurrents (RNN) et leur application à la musique et à la parole. Bien qu'en principe les RNN puissent représenter les dépendances à long terme et la dynamique temporelle complexe propres aux séquences d'intérêt comme la vidéo, l'audio et la langue naturelle, ceux-ci n'ont pas été utilisés à leur plein potentiel depuis leur introduction par Rumelhart et al. (1986a) en raison de la difficulté de les entraîner efficacement par descente de gradient. Récemment, l'application fructueuse de l'optimisation Hessian-free et d'autres techniques d'entraînement avancées ont entraîné la recrudescence de leur utilisation dans plusieurs systèmes de l'état de l'art. Le travail de cette thèse prend part à ce développement. L'idée centrale consiste à exploiter la flexibilité des RNN pour apprendre une description probabiliste de séquences de symboles, c'est-à-dire une information de haut niveau associée aux signaux observés, qui en retour pourra servir d'à priori pour améliorer la précision de la recherche d'information. Par exemple, en modélisant l'évolution de groupes de notes dans la musique polyphonique, d'accords dans une progression harmonique, de phonèmes dans un énoncé oral ou encore de sources individuelles dans un mélange audio, nous pouvons améliorer significativement les méthodes de transcription polyphonique, de reconnaissance d'accords, de reconnaissance de la parole et de séparation de sources audio respectivement. L'application pratique de nos modèles à ces tâches est détaillée dans les quatre derniers articles présentés dans cette thèse. Dans le premier article, nous remplaçons la couche de sortie d'un RNN par des machines de Boltzmann restreintes conditionnelles pour décrire des distributions de sortie multimodales beaucoup plus riches. Dans le deuxième article, nous évaluons et proposons des méthodes avancées pour entraîner les RNN. Dans les quatre derniers articles, nous examinons différentes façons de combiner nos modèles symboliques à des réseaux profonds et à la factorisation matricielle non-négative, notamment par des produits d'experts, des architectures entrée/sortie et des cadres génératifs généralisant les modèles de Markov cachés. Nous proposons et analysons également des méthodes d'inférence efficaces pour ces modèles, telles la recherche vorace chronologique, la recherche en faisceau à haute dimension, la recherche en faisceau élagué et la descente de gradient. Finalement, nous abordons les questions de l'étiquette biaisée, du maître imposant, du lissage temporel, de la régularisation et du pré-entraînement.

Development and plasticity of locomotor circuits in the zebrafish spinal cord

A fundamental goal in neurobiology is to understand the development and organization of neural circuits that drive behavior. In the embryonic spinal cord, the first motor activity is a slow coiling of the trunk that is sensory-independent and therefore appears to be centrally driven. Embryos later become responsive to sensory stimuli and eventually locomote, behaviors that are shaped by the integration of central patterns and sensory feedback. In this thesis I used a simple vertebrate model, the zebrafish, to investigate in three manners how developing spinal networks control these earliest locomotor behaviors. For the first part of this thesis, I characterized the rapid transition of the spinal cord from a purely electrical circuit to a hybrid network that relies on both chemical and electrical synapses. Using genetics, lesions and pharmacology we identified a transient embryonic behavior preceding swimming, termed double coiling. I used electrophysiology to reveal that spinal motoneurons had glutamate-dependent activity patterns that correlated with double coiling as did a population of descending ipsilateral glutamatergic interneurons that also innervated motoneurons at this time. This work (Knogler et al., Journal of Neuroscience, 2014) suggests that double coiling is a discrete step in the transition of the motor network from an electrically coupled circuit that can only produce simple coils to a spinal network driven by descending chemical neurotransmission that can generate more complex behaviors. In the second part of my thesis, I studied how spinal networks filter sensory information during self-generated movement. In the zebrafish embryo, mechanosensitive sensory neurons fire in response to light touch and excite downstream commissural glutamatergic interneurons to produce a flexion response, but spontaneous coiling does not trigger this reflex. I performed electrophysiological recordings to show that these interneurons received glycinergic inputs during spontaneous fictive coiling that prevented them from firing action potentials. Glycinergic inhibition specifically of these interneurons and not other spinal neurons was due to the expression of a unique glycine receptor subtype that enhanced the inhibitory current. This work (Knogler & Drapeau, Frontiers in Neural Circuits, 2014) suggests that glycinergic signaling onto sensory interneurons acts as a corollary discharge signal for reflex inhibition during movement. v In the final part of my thesis I describe work begun during my masters and completed during my doctoral degree studying how homeostatic plasticity is expressed in vivo at central synapses following chronic changes in network activity. I performed whole-cell recordings from spinal motoneurons to show that excitatory synaptic strength scaled up in response to decreased network activity, in accordance with previous in vitro studies. At the network level, I showed that homeostatic plasticity mechanisms were not necessary to maintain the timing of spinal circuits driving behavior, which appeared to be hardwired in the developing zebrafish. This study (Knogler et al., Journal of Neuroscience, 2010) provided for the first time important in vivo results showing that synaptic patterning is less plastic than synaptic strength during development in the intact animal. In conclusion, the findings presented in this thesis contribute widely to our understanding of the neural circuits underlying simple motor behaviors in the vertebrate spinal cord.

Role of the Protein Tyrosine Kinase 7 gene in human neural tube defects

Les anomalies du tube neural (ATN) sont des anomalies développementales où le tube neural reste ouvert (1-2/1000 naissances). Afin de prévenir cette maladie, une connaissance accrue des processus moléculaires est nécessaire. L’étiologie des ATN est complexe et implique des facteurs génétiques et environnementaux. La supplémentation en acide folique est reconnue pour diminuer les risques de développer une ATN de 50-70% et cette diminution varie en fonction du début de la supplémentation et de l’origine démographique. Les gènes impliqués dans les ATN sont largement inconnus. Les études génétiques sur les ATN chez l’humain se sont concentrées sur les gènes de la voie métabolique des folates du à leur rôle protecteur dans les ATN et les gènes candidats inférés des souris modèles. Ces derniers ont montré une forte association entre la voie non-canonique Wnt/polarité cellulaire planaire (PCP) et les ATN. Le gène Protein Tyrosine Kinase 7 est un membre de cette voie qui cause l’ATN sévère de la craniorachischisis chez les souris mutantes. Ptk7 interagit génétiquement avec Vangl2 (un autre gène de la voie PCP), où les doubles hétérozygotes montrent une spina bifida. Ces données font de PTK7 comme un excellent candidat pour les ATN chez l’humain. Nous avons re-séquencé la région codante et les jonctions intron-exon de ce gène dans une cohorte de 473 patients atteints de plusieurs types d’ATN. Nous avons identifié 6 mutations rares (fréquence allélique <1%) faux-sens présentes chez 1.1% de notre cohorte, dont 3 sont absentes dans les bases de données publiques. Une variante, p.Gly348Ser, a agi comme un allèle hypermorphique lorsqu'elle est surexprimée dans le modèle de poisson zèbre. Nos résultats impliquent la mutation de PTK7 comme un facteur de risque pour les ATN et supporte l'idée d'un rôle pathogène de la signalisation PCP dans ces malformations.

Development and Evaluation of Blind Identification Techniques for Nonlinear Systems

Identification and Control of Non‐linear dynamical systems are challenging problems to the control engineers.The topic is equally relevant in communication,weather prediction ,bio medical systems and even in social systems,where nonlinearity is an integral part of the system behavior.Most of the real world systems are nonlinear in nature and wide applications are there for nonlinear system identification/modeling.The basic approach in analyzing the nonlinear systems is to build a model from known behavior manifest in the form of system output.The problem of modeling boils down to computing a suitably parameterized model,representing the process.The parameters of the model are adjusted to optimize a performanace function,based on error between the given process output and identified process/model output.While the linear system identification is well established with many classical approaches,most of those methods cannot be directly applied for nonlinear system identification.The problem becomes more complex if the system is completely unknown but only the output time series is available.Blind recognition problem is the direct consequence of such a situation.The thesis concentrates on such problems.Capability of Artificial Neural Networks to approximate many nonlinear input-output maps makes it predominantly suitable for building a function for the identification of nonlinear systems,where only the time series is available.The literature is rich with a variety of algorithms to train the Neural Network model.A comprehensive study of the computation of the model parameters,using the different algorithms and the comparison among them to choose the best technique is still a demanding requirement from practical system designers,which is not available in a concise form in the literature.The thesis is thus an attempt to develop and evaluate some of the well known algorithms and propose some new techniques,in the context of Blind recognition of nonlinear systems.It also attempts to establish the relative merits and demerits of the different approaches.comprehensiveness is achieved in utilizing the benefits of well known evaluation techniques from statistics. The study concludes by providing the results of implementation of the currently available and modified versions and newly introduced techniques for nonlinear blind system modeling followed by a comparison of their performance.It is expected that,such comprehensive study and the comparison process can be of great relevance in many fields including chemical,electrical,biological,financial and weather data analysis.Further the results reported would be of immense help for practical system designers and analysts in selecting the most appropriate method based on the goodness of the model for the particular context.

Design and Development of data mining models for the predictions of manpower placement in the technical Domain

Data mining is one of the hottest research areas nowadays as it has got wide variety of applications in common man’s life to make the world a better place to live. It is all about finding interesting hidden patterns in a huge history data base. As an example, from a sales data base, one can find an interesting pattern like “people who buy magazines tend to buy news papers also” using data mining. Now in the sales point of view the advantage is that one can place these things together in the shop to increase sales. In this research work, data mining is effectively applied to a domain called placement chance prediction, since taking wise career decision is so crucial for anybody for sure. In India technical manpower analysis is carried out by an organization named National Technical Manpower Information System (NTMIS), established in 1983-84 by India's Ministry of Education & Culture. The NTMIS comprises of a lead centre in the IAMR, New Delhi, and 21 nodal centres located at different parts of the country. The Kerala State Nodal Centre is located at Cochin University of Science and Technology. In Nodal Centre, they collect placement information by sending postal questionnaire to passed out students on a regular basis. From this raw data available in the nodal centre, a history data base was prepared. Each record in this data base includes entrance rank ranges, reservation, Sector, Sex, and a particular engineering. From each such combination of attributes from the history data base of student records, corresponding placement chances is computed and stored in the history data base. From this data, various popular data mining models are built and tested. These models can be used to predict the most suitable branch for a particular new student with one of the above combination of criteria. Also a detailed performance comparison of the various data mining models is done.This research work proposes to use a combination of data mining models namely a hybrid stacking ensemble for better predictions. A strategy to predict the overall absorption rate for various branches as well as the time it takes for all the students of a particular branch to get placed etc are also proposed. Finally, this research work puts forward a new data mining algorithm namely C 4.5 * stat for numeric data sets which has been proved to have competent accuracy over standard benchmarking data sets called UCI data sets. It also proposes an optimization strategy called parameter tuning to improve the standard C 4.5 algorithm. As a summary this research work passes through all four dimensions for a typical data mining research work, namely application to a domain, development of classifier models, optimization and ensemble methods.

Investigations On The Development Of An Ann Model & Visual Manipulation Approach For 2-D Dft Computation In Image Processing

This thesis is an outcome of the investigations carried out on the development of an Artificial Neural Network (ANN) model to implement 2-D DFT at high speed. A new definition of 2-D DFT relation is presented. This new definition enables DFT computation organized in stages involving only real addition except at the final stage of computation. The number of stages is always fixed at 4. Two different strategies are proposed. 1) A visual representation of 2-D DFT coefficients. 2) A neural network approach. The visual representation scheme can be used to compute, analyze and manipulate 2D signals such as images in the frequency domain in terms of symbols derived from 2x2 DFT. This, in turn, can be represented in terms of real data. This approach can help analyze signals in the frequency domain even without computing the DFT coefficients. A hierarchical neural network model is developed to implement 2-D DFT. Presently, this model is capable of implementing 2-D DFT for a particular order N such that ((N))4 = 2. The model can be developed into one that can implement the 2-D DFT for any order N upto a set maximum limited by the hardware constraints. The reported method shows a potential in implementing the 2-D DF T in hardware as a VLSI / ASIC

Development of a Biometric Personal Authentication System Based on Fingerprint and Speech

Biometrics deals with the physiological and behavioral characteristics of an individual to establish identity. Fingerprint based authentication is the most advanced biometric authentication technology. The minutiae based fingerprint identification method offer reasonable identification rate. The feature minutiae map consists of about 70-100 minutia points and matching accuracy is dropping down while the size of database is growing up. Hence it is inevitable to make the size of the fingerprint feature code to be as smaller as possible so that identification may be much easier. In this research, a novel global singularity based fingerprint representation is proposed. Fingerprint baseline, which is the line between distal and intermediate phalangeal joint line in the fingerprint, is taken as the reference line. A polygon is formed with the singularities and the fingerprint baseline. The feature vectors are the polygonal angle, sides, area, type and the ridge counts in between the singularities. 100% recognition rate is achieved in this method. The method is compared with the conventional minutiae based recognition method in terms of computation time, receiver operator characteristics (ROC) and the feature vector length. Speech is a behavioural biometric modality and can be used for identification of a speaker. In this work, MFCC of text dependant speeches are computed and clustered using k-means algorithm. A backpropagation based Artificial Neural Network is trained to identify the clustered speech code. The performance of the neural network classifier is compared with the VQ based Euclidean minimum classifier. Biometric systems that use a single modality are usually affected by problems like noisy sensor data, non-universality and/or lack of distinctiveness of the biometric trait, unacceptable error rates, and spoof attacks. Multifinger feature level fusion based fingerprint recognition is developed and the performances are measured in terms of the ROC curve. Score level fusion of fingerprint and speech based recognition system is done and 100% accuracy is achieved for a considerable range of matching threshold