In this study, we have successfully developed a high performance transmission mode Laser scanning for SHG imaging. This setup is capable of acquiring images of size 512¡Ñ512 pixels at a rate of 5.4 seconds/frame. The of samples can thus be imaged, which reflects the samples¡¦ structure and symmetry.

Scanning probe microscopes (SPMs) are the highest resolution imaging instruments available today and are among the most important tools in nanoscience. Conventional SPMs suffer from several drawbacks owing to their large and bulky construction and to the use of piezoelectric materials. Large scanners have low resonant frequencies that limit their achievable imaging bandwidth and render them susceptible to disturbance from ambient vibrations. Array approaches have been used to alleviate the bandwidth bottleneck; however as arrays are scaled upwards, the scanning speed must decline to accommodate larger payloads. In addition, the long mechanical path from the tip to the sample contributes thermal drift. Furthermore, intrinsic properties of piezoelectric materials result in creep and hysteresis, which contribute to image distortion. The tip-sample interaction signals are often measured with optical configurations that require large free-space paths, are cumbersome to align, and add to the high cost of state-of-the-art SPM systems. These shortcomings have stifled the widespread adoption of SPMs by the nanometrology community. Tiny, inexpensive, fast, stable and independent SPMs that do not incur bandwidth penalties upon array scaling would therefore be most welcome. The present research demonstrates, for the first time, that all of the mechanical and electrical components that are required for the SPM to capture an image can be scaled and integrated onto a single CMOS chip. Principles of microsystem design are applied to produce single-chip instruments that acquire images of underlying samples on their own, without the need for off-chip scanners or sensors. Furthermore, it is shown that the instruments enjoy a multitude of performance benefits that stem from CMOS-MEMS integration and volumetric scaling of scanners by a factor of 1 million. This dissertation details the design, fabrication and imaging results of the first single-chip contact-mode AFMs, with integrated piezoresistive strain sensing cantilevers and scanning in three degrees-of-freedom (DOFs). Static AFMs and quasi-static AFMs are both reported. This work also includes the development, fabrication and imaging results of the first single-chip dynamic AFMs, with integrated flexural resonant cantilevers and 3 DOF scanning. Single-chip Amplitude Modulation AFMs (AM-AFMs) and Frequency Modulation AFMs (FM-AFMs) are both shown to be capable of imaging samples without the need for any off-chip sensors or actuators. A method to increase the quality factor (Q-factor) of flexural resonators is introduced. The method relies on an internal energy pumping mechanism that is based on the interplay between electrical, mechanical, and thermal effects. To the best of the author???s knowledge, the devices that are designed to harness these effects possess the highest electromechanical Qs reported for flexural resonators operating in air; electrically measured Q is enhanced from ~50 to ~50,000 in one exemplary device. A physical explanation for the underlying mechanism is proposed. The design, fabrication, imaging, and tip-based lithographic patterning with the first single-chip Scanning Thermal Microscopes (SThMs) are also presented. In addition to 3 DOF scanning, these devices possess integrated, thermally isolated temperature sensors to detect heat transfer in the tip-sample region. Imaging is reported with thermocouple-based devices and patterning is reported with resistive heater/sensors. An ???isothermal electrothermal scanner??? is designed and fabricated, and a method to operate it is detailed. The mechanism, based on electrothermal actuation, maintains a constant temperature in a central location while positioning a payload over a range of >35??m, thereby suppressing the deleterious thermal crosstalk effects that have thus far plagued thermally actuated devices with integrated sensors. In the thesis, models are developed to guide the design of single-chip SPMs and to provide an interpretation of experimental results. The modelling efforts include lumped element model development for each component of single-chip SPMs in the electrical, thermal and mechanical domains. In addition, noise models are developed for various components of the instruments, including temperature-based position sensors, piezoresistive cantilevers, and digitally controlled positioning devices.

Les aciers inoxydables à haute teneur en molybdène présentent une excellente résistance à la corrosion et ont été récemment utilisés dans des applications marines. L'acier inoxydable de type 254 SMO contient du molybdène (6% Mo) et des quantités plus élevées en éléments d'alliages que les aciers inoxydables plus classiques tels que 304, 316 et 316L. Alors que Anees Uddin Malik et al. montrent que l’acier 254 SMO est moins sensible à la corrosion par piqûre en eau de mer du golfe persique, Abdulsalam et al. démontrent que bien que cet acier présente une bonne résistance à la corrosion, il reste encore sensible à une corrosion caverneuse (surtout en milieu à 16% en masse d’ion chlorure ou à plus de 30°C).La première partie de ce travail traite du comportement électrochimique de l’acier inoxydable 254 SMO immergé dans l’eau de mer naturelle en laboratoire. Ainsi deux sites de la Guyane ont été choisis pour effectuer des prélèvements d’eau. Des analyses physicochimiques de ces eaux sont réalisées et montrent que l’eau du site du Port Larivot contient une quantité en ions chlorures plus importante que celle de l’eau du site du Mahury. Une colonisation par des bactéries et crustacés (balanes) a été mise en évidence, à l’aide du microscope électronique à balayage environnemental. Les mesures électrochimiques, notamment la voltammétrie cyclique a montré l’existence d’une zone de passivation et l’augmentation du courant de passivation par un facteur de 10 en présence d’un biofilm pour l’acier inoxydable immergé en eau du Mahury durant 22 jours. La deuxième partie propose une étude approfondie du comportement de l’interface complexe métal/couche passive/biofilm. Trois différents coupons d’aciers inoxydables (brut, poli et grenaillé) ont été déposés sur site (in-situ) dans un premier temps. Dans une seconde étape, nous avons réalisé des expérimentations mixtes (incubation in-situ et croissance en laboratoire). L’évolution du potentiel d’abandon de ces aciers est très proche pour l’acier brut et l’acier poli. En revanche, celui de l’acier grenaillé, de par la présence des grains d’alumine à sa surface, affiche un comportement différent. Une étude comparative des deux expérimentations montre des différences concernant les paramètres électriques de l’interface complexe métal/couche passive/biofilm. / Stainless steels with high content in molybdenum present an excellent corrosion resistance and were recently used in marine applications. The stainless steel 254 SMO contains molybdenum (6 %) and higher amounts of alloying elements than conventional stainless steels such as 304, 316 and 316L. While Anees Uddin Malik et al. show that 254 SMO steel is less likely to be susceptible to pitting corrosion in Arabian Gulf sea water, Abdulsalam et al. demonstrate that although this steel has good resistance to corrosion, it remains still susceptible to crevice corrosion (especially in environment with 16 chloride percent weight or more than 30 ° C).The first part of this work deals with the electrochemical behaviour of stainless steel 254 SMO immersed in natural seawater at laboratory. Two sites in French Guyane were chosen to take samples of seawater. Physicochemical analyses of these waters are made and show than the in Port Larivot site seawater contains a higher quantity of chloride ion than to Mahury seawater. Colonization by bacteria and crustaceans (barnacles) was highlighted, using the environmental scanning electron microscope. The electrochemical measurements, including cyclic voltammetry showed the existence of a passivation region and increasing the passivation current by a factor of 10 in the presence of biofilm on stainless steel immersed in Mahury seawater during 22 days.The second part offers a detailed study of the behaviour of the interface metal/passive film/biofilm. Three different coupons of stainless steel (crude, polished and blasted) were deposited on the site (in-situ) in the first time. In a second step, we performed experiments mixed (in-situ incubation and growth in the laboratory). The evolution of the free potential of these steels is very close for crude steel and polished steel. Conversely, the blasted steel, by the presence of alumina grains on its surface, shows a different behaviour. A comparative study of those two experiments shows differences in the electrical parameters of the interface metal/passive film/biofilm.

Bien que devenant un environnement de plus en plus riche technologiquement, la salle opératoire reste un endroit où la sécurité des patients n'est pas assurée à 100% comme le montre le nombre toujours conséquent d'erreurs chirurgicales. La nécessité de développer des systèmes intelligents au bloc opératoire apparait comme croissante. Un des éléments clés pour ce type de système est la reconnaissance du processus chirurgical, passant par une identification précise des outils chirurgicaux utilisés. L'objectif de cette thèse a donc porté sur la détection en temps-réel des outils chirurgicaux dans des vidéos 2D provenant de microscopes. Devant l'absence de jeux de données de référence, qui plus est dans un contexte neurochirurgical, la première contribution de la thèse a donc été la création d'un nouvel ensemble d'images de chirurgies du cerveau et du rachis cervical, mis à disposition en ligne. Comme seconde contribution, deux approches différentes ont été proposées permettant de détecter des outils chirurgicaux via des techniques d'analyse d'image. Tout d'abord, le SquaresChnFtrs adapté, basé sur une des méthodes de détection de piétons les plus performantes de la littérature. Notre deuxième méthode, le ShapeDetector, est une approche à deux niveaux n'utilisant aucune contrainte ou hypothèse a priori sur le nombre, la position, ou la forme des outils dans l'image. Par rapport aux travaux précédents du domaine, nous avons choisi de représenter les détections potentielles par des polygones plutôt que par des rectangles, obtenant ainsi des détections plus précises. Pour intégration dans des systèmes médicaux, une optimisation de la vitesse de calcul a été effectuée via un usage optimal du CPU, du GPU, et de méthodes ad-hoc. Pour des vidéos de résolution 612x480 pixels, notre ShapeDetector est capable d'effectuer les détections à une vitesse maximale de 8 Hz. Pour la validation de nos méthodes, nous avons pris en compte trois paramètres: la position globale, la position de l'extrémité, et l'orientation des détections. Les méthodes ont été classées et comparées avec des méthodes de référence compétitives. Pour la détection des tubes d'aspiration, nous avons obtenu un taux de manqué de 15% pour un taux de faux positifs par image de 0.1, comparé à un taux de manqué de 55% pour le SquaresChnFtrs adapté. L'orientation future du travail devra porter sur l'intégration des informations 3D, l'amélioration de la couche de labélisation sémantique, et la classification des outils chirurgicaux. Pour finir, un enrichissement du jeu de données et des annotations de plus haute précision seront nécessaires. / Despite modern-life technological advances and tremendous progress made in surgical techniques including MIS, today's OR is facing many challenges remaining yet to be addressed. The development of CAS systems integrating the SPM methodology was born as a response from the medical community, with the long-term objective to create surgical cockpit systems. Being able to identify surgical tools in use is a key component for systems relying on the SPM methodology. Towards that end, this thesis work has focused on real-time surgical tool detection from microscope 2D images. From the review of the literature, no validation data-sets have been elected as benchmarks by the community. In addition, the neurosurgical context has been addressed only once. As such, the first contribution of this thesis work consisted in the creation of a new surgical tool data-set, made freely available online. Two methods have been proposed to tackle the surgical tool detection challenge. First, the adapted SquaresChnFtrs, evolution of one of the best computer vision state-of-the-art approach for pedestrian detection. Our second contribution, the ShapeDetector, is fully data-driven and performs detection without the use of prior knowledge regarding the number, shape, and position of tools in the image. Compared to previous works, we chose to represent candidate detections with bounding polygons instead of bounding boxes, hence providing more fitting results. For integration into medical systems, we performed different code optimization through CPU and GPU use. Speed gain and accuracy loss from the use of ad-hoc optimization strategies have been thoroughly quantified to find an optimal trade-off between speed and accuracy. Our ShapeDetector is running in-between 5 and 8Hz for 612x480 pixel video sequences.We validated our approaches using a detailed methodology covering the overall tool location, tip position, and orientation. Approaches have been compared and ranked conjointly with a set of competitive baselines. For suction tube detections, we achieved a 15% miss-rate at 0.1 FPPI, compared to a 55% miss-rate for the adapted SquaresChnFtrs. Future works should be directed toward the integration of 3D feature extraction to improve detection performance but also toward the refinement of the semantic labelling step. Coupling the tool detection task to the tool classification in one single framework should be further investigated. Finally, increasing the data-set in diversity, number of tool classes, and detail of annotations is of interest.

L’étude des propriétés singulières de nanoparticules colloïdales synthétisées par voie chimique et leur intégration dans des nano-composants requiert leur assemblage dirigé sur des zones parfaitement définies et localisées de surfaces solides. L’objet de cette thèse est le développement d’une méthode d’assemblage dirigé originale: la nanoxérographie par microscope à force atomique (AFM). Cette technique consiste à injecter localement, sur des zones spécifiques, des charges électrostatiques dans un matériau électret par l’intermédiaire d’une pointe d’AFM. Ces charges servent ensuite de pièges électrostatiques sur la surface pour les nanoparticules en solution. Dans le cadre de ce travail, l’injection, la rétention de charges dans de fines couches électrets de PolyMéthylMéthAcrylate (PMMA) et la quantification des densités de charges surfaciques des motifs chargés, ont été étudiées grâce au mode électrique dérivé de l’AFM, le microscope à force Kelvin (KFM). L’étude de l’assemblage de nanoparticules de différentes natures (métal, polymère (organique ou inorganique)), de taille moyenne variable dans un large domaine (2 nm - 1µm) et de potentiel zêta contrôlé a permis d’analyser les mécanismes de dépôt et de montrer les performances de la méthode et son aspect générique. Enfin deux techniques d’injection de charges parallèles ont été mises en place afin d’offrir des perspectives industrielles: le microcontact printing électrique et la nanoimpression électrique / The study of original properties of colloidal nanoparticles and their integration into nanodevices requires their assembly onto specific areas of solid surfaces. The aim of this thesis work is to develop an innovative method for the directed assembly of colloidal nanoparticles: the nanoxerography process by atomic force microscope (AFM). This technique consists in injecting charges into electrets using an AFM tip. The injected charges are then used to electrostatically trap nanoparticles from suspensions onto the surface. In this context, the charge writing and charge decay in PolyMethyMethAcrylate (PMMA) thin films were studied and the charge density of the charged patterns were quantified using Kelvin force microscope (KFM), an electrical mode of AFM. Assemblies of nanoparticles of different nature (metallic, polymeric (organic and inorganic)), with average sizes extending over a large range (2 nm to 1 µm) and controlled zeta potential were obtained on PMMA thin films. This allowed the analysis of assembly mechanisms and demonstration of the excellent performance of the method. Finally, two techniques of parallel charge writing, viz., the electrical microcontact printing and the electrical nanoimprinting were explored with the prospect of extending the nanoxerography process to industrial scale

Nous présentons un travail expérimental contribuant à l'étude de contacts ponctuels métalliques et moléculaires à l'aide de dispositifs de jonctions brisées développés dans notre équipe. Ces techniques de jonctions brisées, utilisables dans les conditions ambiantes, sont particulièrement adaptées à deux champs disciplinaires : l'électronique moléculaire et la nano mécanique.Nous avons étudié la durée de vie de contacts métalliques d'or, qui excède rarement la dizaine de millisecondes à température ambiante. Par le biais d'une analyse statistique de mesures de conductance, nous montrons que leur durée de vie est limitée par la contrainte mécanique appliquée à la jonction. Ces résultats nous ont permis de proposer un mécanisme de rupture, et de définir des conditions optimales pour la formation des contacts à température ambiante.Nous présentons ensuite une étude préliminaire de mesure de conductance d'une molécule unique, utilisant un dispositif à jonction brisée ainsi qu'un microscope à effet tunnel.Les résultats obtenus indiquent que, si la mesure de la conductance d'une molécule unique est possible, la stabilité observée est à considérer avant d'envisager des applications. / We present here an experimental work which contributes to the study of metallic and molecular point contacts using broken junctions-based devices developed in our team. Under ambient environmental conditions, these techniques are particularly adapted to two disciplinary fields: molecular electronics and nano-mechanics.We have studied the lifetime of gold contacts, which rarely exceed ten milliseconds at room temperature.Through statistical analyses of conductance measurements, we show that this lifetime is limited by the mechanical strain applied to the junction. These results allowed us to propose a breaking mechanism, and to define optimal conditions for the formation of the contacts at room temperature. We present then a preliminary study of conductance measurements of a single molecule, using a broken junction device as well as a scanning tunneling microscope.The results indicate that, despite the conductance measure of a single molecule is possible the observed stability should be considered before envisaging applications.

Les microscopes en champ proche permettent d'obtenir la topographie d'un échantillon avec une résolution pouvant atteindre la résolution atomique. Les spectroscopies de rayons-X sont des méthodes de caractérisation qui permettent de déterminer la composition et la structure élémentaire de l'échantillon avec une précision inférieure à l'Ångström. Nous avons choisi de coupler ces deux techniques en collectant localement la luminescence visible issue de l'échantillon par la pointe-sonde d'un microscope à force de cisaillement, constituée d'une fibre optique effilée de faible ouverture. Cette technique a été utilisée pour caractériser des échantillons semiconducteurs micro- et nano-structurés afin d'en obtenir simultanément la topographie et la cartographie de luminescence locale. Afin de pouvoir étendre ce concept à d'autres types de matériaux, la faisabilité de la collecte de la fluorescence X locale a été évaluée avec la microsource. Pour cela la fluorescence X émise par un échantillon a été collectée par un capillaire cylindrique équipant un détecteur EDX. L'influence du diamètre du capillaire sur le niveau de signal a été mesurée. Une simulation numérique a été développée afin d'estimer le niveau de signal obtenu en utilisant un capillaire de 1 µm de diamètre et d'optimiser la géométrie du système. En couplant la microscopie en champ proche et l'analyse XRF, à la lumière de ces résultats, il sera possible d'atteindre 100 nm de résolution latérale en environnement synchrotron et moins de 1 µm à l'aide d'une source de laboratoire. Il serait alors possible de sélectionner un objet particulier sur une surface et d'en faire l'analyse élémentaire. / Scanning Probe Microscopes allow to obtain sample topography up to atomic resolution. X-ray spectroscopies allow elemental and structural analysis of a sample with accuracy better than 1 Å. The lateral resolution is limited by the primary beam diameter, currently a few µm². We have chosen to couple this two technics. Local sample visible luminescence is collected through a low aperture sharp optical fibre, probe of a shear force microscope. This technique was used to characterize microstructured semiconducting samples to achieve simultaneously the surface topography and luminescence mapping. The results were obtained using either synchrotron radiation or a laboratory microsource equipped with a polycapillary lens. To extend this concept to a wider variety of materials, local XRF collection by an EDX detector equipped with a cylindrical X-ray capillary was tested. A cobalt sample irradiated with the microsource was used for technique evaluation. The signal magnitude dependence with the capillary diameter was measured. Modelling and numerical calculations were developed to estimate the signal magnitude that could be detected using a 1 µm diameter capillary. The optimal system geometry was determined. Scanning Probe Microscopy combined to XRF analysis could thereby lead to simultaneous acquisition of sample topography and chemical mapping. The expected lateral resolution using synchrotron radiation is 100 nm while sub 1 µm resolution is realistic with a laboratory source. This technique would allow to point a peculiar micro- or nano-object on the surface and to perform its chemical analysis.

After nearly 100 years of particle acceleration, particle accelerator experiments continue providing results within the field of high energy physics. Particle acceleration is used worldwide in practical applications such as radiation therapy and materials science research. Unfortunately, these accelerators are large and expensive. Dielectric Laser Acceleration (DLA) is a promising technique for accelerating particles with high acceleration gradients, without requiring large-scale accelerators. DLA utilizes the electric field of a high energy laser to accelerate electrons in the proximity of a nanostructured dielectric surface.The aim of this project was limited to laser beam routing and imaging techniques for a DLA experiment. The goal was to design the laser beam pathway between the laser and the dielectric sample, and testing a proposed imaging system for aiming the laser. This goal was achieved in a test setup using a low-energy laser. In the main setup including a femtosecond laser, the result indicated lack of focus. For a full experimental setup, a correction of this focus is essential and the beam path would need to be combined with a Scanning Electron Microscope (SEM) as an electron source.

Neste trabalho foram utilizadas técnicas de modelagem molecular computacional para descrever nanossuperfícies funcionalizadas com biomoléculas do sistema imunológico correlacionando resultados experimentais obtidos com o microscópio de força atômica, simulações de dinâmica molecular e dinâmica molecular direcionada. O objetivo principal proposto é avaliar as forças intermoleculares provenientes das interações antígeno-anticorpo (funcionalizados em nanossuperfícies) para aplicação no desenvolvimento de nanoimunossensores e detecção de doenças desmielinizantes, como a Neuromielite Óptica. A Neuromielite Óptica é uma doença inflamatória autoimune na qual o próprio sistema imunológico reage contra os nervos ópticos e a medula espinhal, causando lesão desmielinizante. Estudos na literatura estabeleceramo anticorpo anti-aquaporina4 como um importante biomarcador da doença. Neste contexto, um nanoimmunosensorvem sendo desenvolvido com a técnica de Microscopia de Força Atômica, o qual visa detectar o anticorpoanti-aquaporina4 no soro de portadores da doença. Tal estudo necessitou de uma nova abordagem computacional para a descrição de estruturas tridimensionais de anticorpos. Essa nova aproximação consistiu na aplicação de técnicas de computacionais de modelagem e engenharia molecular para a geração de modelos de anticorpos com base em sucessivas substituições dos resíduos componentes do sítio de interação com o antígeno. Testes realizados envolvendo modelos de anticorpos disponíveis em bancos de dados especializadosindicaram (48 ± 18) % e (65 ± 14) % de identidade das cadeias leve e pesada, respectivamente, entre os modelos gerados computacionalmente e as estruturas 3D reais de anticorpos. Por fim, para comprovar o funcionamento dos nanoimunossensores, foi desenvolvido um modelo estatístico para tratar e interpretar os dados experimentais. Este modelo foi eficiente para distinguir os pacientes soropositivos de sujeitos soronegativos para determinados biomarcadores relacionados à Neuromielite Óptica e a Esclerose Múltipla, fornecendo assim um novo e mais preciso processopara diagnóstico de doenças desmielinizantes. / Study of the biological coverage of nanosurfaces by computational modeling: application in the development of nanoimunoresensors. In this work, computational molecular modeling techniques were applied to describe nanosurfaces functionalized with immune system biomolecules, correlating data from atomic force microscope experiments, molecular dynamics, and steered molecular dynamics simulations. The main goal of this research was to evaluate intermolecular forces involved in the antigen-antibody interaction on the nanosurfaces during the development of nanoimmunosensors for demyelinating diseases detection, especially neuromyelitisoptica. The neuromyelitisoptica is an autoimmune inflammation in which components of the immune system respond against optical nerves and spinal cord, resulting in demyelinating lesions. In the literature, studies have established anti-aquaporin 4 as an important biomarker for neuromyelitisoptica. Then, a nanoimmunosensor for anti-aquaporin 4 antibodies detection in neuromyelitisoptica patients serum via Atomic Force Microscopy is in development. This study requested a computational approach for describing the tridimensional structure of antibodies. The novel approach consisted of computer molecular modeling and engineering to perform successive substitutions in residues of the antigen interaction site. Tests carried out using antibody structures available in specialized data banks demonstrated the similarity of (48 ± 18) % and (65 ± 14) % for light and heavy chains, respectively, of the computationally generated models and experimental 3D structures of antibodies. Additionally, a statistical model was developed to prove the nanoimmunosensor sensing activity, which was useful to treat and interpret the experimental data. This statistical model was efficient to distinguish seropositive patients from seronegative subjects considering specific biomarkers related to neuromyelitisoptica and multiple sclerosis, providing a novel and more precise process for demyelinating disease diagnosis.

Le développement de méthodes de la détection en temps réel de cellules anormales (pouvant être considérées comme des cellules cancéreuses) par captures et traitements bio-images sont des axes de recherche importants dans le domaine biomédical car cela contribue à diagnostiquer un cancer. C’est dans ce contexte que se situe ces travaux de thèse. Plus précisément, les travaux présentés dans ce manuscrit, se focalise sur le développement de procédures de lecture, de détection et de classification automatiques de bio-images de cellules anormales considérées comme des cellules cancéreuses. Par conséquent, une première étape du travail à consister à déterminer une solution de détection, à partir d’images microscopiques multispectrales permettant une répétitivité d’images sur une gamme de longueurs d'ondes de certains types de bio-images anormales associées à différents stades ou évolutions de cellules cancéreuses. L’approche développée dans ces travaux repose sur l’exploitation d’une nouvelle méthode de segmentation basée sur l'intensité de la couleur et pouvant être appliquée sur des séquences d'objets dans une image en reformant de manière adaptative et itérative la localisation et la couverture de contours réels de cellules. Cette étape préalable de segmentation est primordiale et permet une classification des tissus anormaux en utilisant la méthode de réseau de neurones à convolution (CNN) appliqué sur les images microscopiques segmenté de type snake. L’approche permet d’obtenir de bas résultats comparativement à une approche basée sur d’autres méthodes de segmentation de la littérature. En effet, cette méthode de classification atteint des valeurs de performance de 100% pour la phase d’apprentissage et de 99.168 % pour les phases de test. Cette méthode est comparée à différents travaux antérieurs et basée sur différentes fonctionnalités d'extraction, et a prouvé son efficacité par rapport à ces autres méthodes. En terme de perspectives, les travaux futurs visent à valider notre approche sur des ensembles de données plus larges, et à explorer différentes architectures CNN selon différents critères d’optimisation / Development of methods for help diagnosis of the real time detection of abnormal cells (which can be considered as cancer cells) through bio-image processing and detection are most important research directions in information science and technology. Our work has been concerned by developing automatic reading procedures of the normal and abnormal bio-images tissues. Therefore, the first step of our work is to detect a certain type of abnormal bio-images associated to many types evolution of cancer within a Microscopic multispectral image, which is an image, repeated in many wavelengths. And using a new segmentation method that reforms itself in an iterative adaptive way to localize and cover the real cell contour, using some segmentation techniques. It is based on color intensity and can be applied on sequences of objects in the image. This work presents a classification of the abnormal tissues using the Convolution neural network (CNN), where it was applied on the microscopic images segmented using the snake method, which gives a high performance result with respect to the other segmentation methods. This classification method reaches high performance values, where it reaches 100% for training and 99.168% for testing. This method was compared to different papers that uses different feature extraction, and proved its high performance with respect to other methods. As a future work, we will aim to validate our approach on a larger datasets, and to explore different CNN architectures and the optimization of the hyper-parameters, in order to increase its performance, and it will be applied to relevant medical imaging tasks including computer-aided diagnosis