Skip navigation
Please use this identifier to cite or link to this item: https://repository.esi-sba.dz/jspui/handle/123456789/378
Full metadata record
DC FieldValueLanguage
dc.contributor.authorAISSIOUENE, MAhrez-
dc.contributor.authorBOUKABRINE, FAycal Amine-
dc.date.accessioned2022-11-14T07:21:45Z-
dc.date.available2022-11-14T07:21:45Z-
dc.date.issued2022-
dc.identifier.urihttps://repository.esi-sba.dz/jspui/handle/123456789/378-
dc.descriptionEncadreur : Mme Naoum Hanaeen_US
dc.description.abstractThe advancing field of medicine and information technology has been able to provide revolutionary techniques in the localization and treatment of serious pathologies however, cancer remains till this day a difficult diagnosis despite the evolving of medical imaging machines like radigraphique imaging. Our solution is to implement mamographie alternative which is artificial intelligence in the service of breast cancer detection. Our job is to study in depth the most effective methods of artifical neural networks in order to apply them in such a delicate field in medicine. We have costructed a connected convolutional neural networks , to make the supervised classification This model take as input a vector of labeled data of suspected breast cancers and provides a prediction relating to the 2 diagnostic categories namely:malignant and benign. Our most important results relate to the precision obtained for the classification of breast cancers by applying the targeted techniques. The scope and validity of our work lies not only in the automatic calculation of its success rate based on the accuracy obtained but also to the possible application of our model on the field by medical experts. *** Le domaine ´evolutif de la m´edecine et des technologies de l’information a ´et´e en mesure de fournir des techniques r´evolutionnaires dans la localisation et le traitement des pathologies graves cependant, le cancer reste `a ce jour un diagnostic difficile malgr´e l’´evolution des machines d’imagerie m´edicale comme l’imagerie radigraphique. Notre solution consiste `a mettre en oeuvre l’alternative mammographique qui est l’intelligence artificielle au service de la d´etection du cancer du sein. Notre m´etier est d’´etudier en profondeur les m´ethodes les plus efficaces de r´eseaux de neurones artificiels afin de les appliquer dans un domaine aussi d´elicat de la m´edecine. Nous avons construit un r´eseau de neurones convolutifs connect´es , pour faire la classification supervis ´ee Ce mod`ele prend en entr´ee un vecteur de donn´ees ´etiquet´ees des cancers du sein suspect´es et fournit une pr´ediction relative aux 2 cat´egories de diagnostic `a savoir : malignes et b´enignes. Nos r´esultats les plus importants concernent la pr´ecision obtenue pour la classification des cancers du sein en appliquant les techniques cibl´ees. La port´ee et la validit´e de notre travail ne r´esident pas seulement dans le calcul automatique de sa taux de r´eussite en fonction de la pr´ecision obtenue mais aussi de l’´eventuelle application de notre mod`ele sur domaine par des experts m´edicaux.en_US
dc.language.isoenen_US
dc.subjectArtificial Intelligenceen_US
dc.subjectArtifical Neural Networksen_US
dc.subjectConnected Convolutional Neural Networksen_US
dc.subjectSupervised Classificationen_US
dc.subjectAccuracyen_US
dc.titleDétection du cancer du sein en utilisant les réseaux de neurones convolutifen_US
dc.typeThesisen_US
Appears in Collections:Ingénieur

Files in This Item:
File Description SizeFormat 
Engennering_thesis-1-1.pdf108,27 kBAdobe PDFView/Open
Show simple item record


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.