L’électro-encéphalogramme (EEG) est un examen qui permet l’enregistrement de l’activité cérébrale et reste incontournable pour les diagnostics et pronostics des comas, le suivi des épilepsies et de la maturation du système nerveux central. Actuellement, la pénurie nationale de médecins neurophysiologistes ne permet plus l’accès aux soins pour tous. Il existe actuellement des solutions de transmission à distance des EEG, non informatisées pour 34 % d’entre elles. Après analyse des problèmes que rencontrent ces « plateformes », nous décrivons le projet SMART-EEG qui vise à résoudre quelques-uns des freins technologiques et à proposer une plateforme intégrée qui devrait permettre d’apporter des solutions techniques et pratiques innovantes aux soucis majeurs de la réalisation et de l’interprétation des examens d’EEG par télémédecine en France.

This paper focuses on the study of fibrosis induced by an implanted medical device and explores the possibility of characterizing this process by in situ measurement of electrical impedance. The approach combines concurrent electrical and biological characterizations of fibrotic tissue, applied to electrodes implanted in an animal model. The methodology used is described. Initial results evidence the strong correlation between fibrotic tissue and electrical parameters. Identification of such electrical parameters will enable the establishment of a reliable monitoring method.

This paper presents a new embeddable method for polyp detections in Wireless Capsule Endoscopic - WCE images. this approach consists first of extracting candidate polyps within the image using geometric considerations about related shape, and second, in classifying (polyp/non-polyp) obtained candidates by a boosting-based method using texture features. The proposed approach has been designed in accordance with the hardware constraints related to FPGA implementation for integration within WCE imaging device. The classification performance of the method have been evaluated on a large dataset of 300 polyps, and 1200 non-polyps images. Experiments show interesting and promising performance: the boosting-based classification is characterized by a sensitivity of 91%, a specificity of 95% and a false detection rate of 4.8%, the detection rate of the overall processing chain being of 68%. The performance of the boosting-based classification are in accordance with the most recent reference on this particular topic using the same dataset. Building of a dedicated WCE image database should permit the improvement of the global detection rate.

Wireless capsule endoscopy (WCE) is commonly used for noninvasive gastrointestinal tract evaluation, including the identification of polyps. In this paper, a new multimodal embeddable method for polyp detection and classification in wireless capsule endoscopic images was developed and tested. The multimodal wireless capsule used both 2D and 3D data to identify possible polyps and to deliver cancerous information of the polyps based on 3D geometric features. Possible polyps within the image (2D) were extracted using simple geometric shape features and, in a second step, the candidate regions of interest (ROI) were evaluated with a boosting-based method using textural features. Once the 2D identification of polyps has been performed, the two-class (“malignant” or “begnin”) classification of the polyps is achieved using the 3D parameters computed from the preselected ROI using an active stereo vision system. At this stage, a Support Vector Machine (SVM) classifier is used to proceed to the final classification and to make possible a pre diagnosis. The new proposed multimodal approach based on 2D-3D feature extraction improves WCE capabilities to identify and classify polyps: The boosting-based polyp classification demonstrated a sensitivity of 91%, a specificity of 95% and a false detection rate of 4.8% on a database composed of 300 hundred positive examples and 1200 negative ones; Considering the 3D performance, a large scale demonstrator was evaluated and tested to perform in vitro experiments on an ad hoc polyp database. The performance of the 3D approach achieved a correct classification rate (malignant or benin) of approximately 95%.

The work presented here is related to Cyclope, an embedded active 3D vision system. The challenge is to realize an integrated sensor which achieve real-time depth vision with constraints on the size, consumption and computational ressources. Cyclope is based on active vision and is integrated in a VSiP (Visual System in Package) allowing multiple technology cohabitation, such as a CMOS imager or an AsGa VSCEL, in the same chip size package.In this paper we present an acquisition method which allows to grab both texture and IR pattern images at a 25 image/s frame rate. The method uses a pulsed IR laser pattern projector and CMOS image sensor with programmable integration time. This sensor is realized in a standard silicon technology and without additional optical filtering.The first experimental results are presented to validate our approach with the realization of a functional prototype using a SOPC board, a structured pattern projector and the designed image sensor.

Recent advances in vision system integration allows to design new integrated sensors which are able to achieve complexes tasks. The 'Cyclope' project aims to develop a novel integrated sensor for real-time 3D reconstruction. It includes an active vision sensor, a digital centralized processing architecture and a communication block. This sensor is designed to obtain a monolithic integrate vision system. This paper presents the digital processing architecture which is a massively-parallel architecture in a totally scalable IP to assure the compatibility with application constraints. This architecture has been implemented on an FPGA circuit and we demonstrate, with accurate results, the realtime operations.

Many systems and data processing for 3D vision have been designed but few have targeted real-time embedded systems. We present in this article Cyclope, an original real-time embedded multi-spectral 3D vision sensor. For this, we have developed a new digital processing architecture to reconstruct the 3D information from images. This architecture conforms to embedded system constrains as power dissipation and limited resources. It's a massively-parallel architecture in a totally scalable IP to assure the compatibility with several processing schemes. This architecture has been implemented on a FPGA target and we have obtained accurate results while ensuring realtime operations.

The design of mixed signal systems on chip is in a continuous growth these last few years. So the technical progress for systems integration allows the implementation of millions of active or passive pixel sensors (APS or PPS) on a same chip in order to define image matrix (system on chip). It has become crucial for the design of these systems to accurately predict their behavior prior to manufacture. Now the emergence of standard description languages as VHDL-AMS offers an effective and efficient way to describe these multiple domain and mixed-signal electronic systems. The aim of this work is the modeling of pixel sensors (APS and PPS), basic cells of CMOS image systems. Pixel sensors are composed of photodetectors for photogeneration of current and photocircuit for conversion and/or pre amplification of this current. PPS are usually associated to an operational amplifier (OPA). So in this paper the authors present models for each one of these parts and we instantiate them to simulate PPS and APS cells

This paper presents noise characterization of a CMOS 0.6-µm image sensors system. The integrated readout circuit adaptability allows measuring the photocurrent magnitude of each pixel in charge integration mode or in transimpedance mode. In order to find the minimum detectable signal of this system and to identify the main dominant noise sources, an accurate noise analysis is performed under dark and illumination conditions. Noise measurement and simulation with Spectre present a good agreement with the values predicted by the proposed analytical equations.

This paper describes the implementation of MPEG4 encoder on embedded ALTERA RISCS processors Nios 32. The purpose of this implementation is to validate the platform MERITE. It is an architecture of a versatile platform for wireless sensor networks. This platform acquires environmental data and detects important variations in order to trigger some alarm. When some alarm is triggered on, data of the environment are compressed and sent on a wireless network. The MPEG4 algorithm is used to compress the video data.

In this paper, we introduce the architecture of a versatile platform named MERITE for wireless sensor networks. This platform acquires environmental data and detects important variations in order to trigger some alarm. Finally, when some alarm is triggered, data of the environment are sent on a wireless network. The beacon is implemented on a platform including a microprocessor core and several dedicated IP (intellectual property).

In this paper, to be able to simulate imaging systems containing APS cells including phototransistors, an electric model based on a physical approach was elaborated and it was written in the VHDL-AMS language. It allows the simulation of the spectral response of phototransistors sensibilities and the study of their linearity according to the power of the incident light.

We present here a new paradigm based on a centralized architecture to realize electronic artificial retina. This original architecture, named connectionist retina, can execute in real time RBF and MLP neural networks applications. We demonstrate that this intelligent embedded system could be used for vision applications. We describe here the realized prototype system.