Browsing by Author "Bennai, Mohamed Tahar"

Now showing 1 - 9 of 9

Air quality monitoring using IoT : a survey
(IEEE, 2019) Mokrani, Hocine; Lounas, Razika; Bennai, Mohamed Tahar; Salhi, Dhai Eddine; Djerbi, Rachid
The increase in industrial activities and the rapidurbanization of human populations had a terrible effect onglobal air quality. Thousands of factories and billions of vehiclesrelease enormous amounts of pollutants into the air everyday; dangerously affecting human health. Many epidemiologicalstudies pointed out the responsibility of air pollution in manyhealth issues, the reason why monitoring air quality becamean obligation to prevent or limit these issues. Conventionalsystems based on measuring stations are expensive and offerlow data granularity. As a result, researchers are increasinglytargeting IOT-based systems. However, elaborating a new systemfor air quality monitoring requires an awareness of the stateof the art and the mastery of a certain amount of specificknowledge (pollutants, their health effects, the sensing equipment,the IOT possible configuration,...).This paper aims to answerthese necessities by reviewing the existing works on air qualitymonitoring using IOT with the focus on lasted trends andchallenges
A Collaborative System for Machine Learning-Based Final-Year Projects With Enhanced Dataset Accessibility
(IGI Global, 2024) Lounas, Razika; Djerbi, Rachid; Mokrani, Hocine; Bennai, Mohamed Tahar
This chapter explores the transformative impact of information and communication technology (ICT) on pedagogy, specifically focusing on the integration of collaboration tools in final year projects (FYPs). Final year projects (FYPs) represent the ultimate activity in the student's curriculum. They are designed to use, test, and enhance the knowledge students have gained over the years by confronting them with real-world projects. Despite existing systems for FYPs, the chapter identifies gaps, particularly in covering the entire FYP process and in addressing different collaborative aspects. With a focus on the rise of machine learning-based FYPs, this research aims to propose a comprehensive solution based on a proposed collaboration architecture in response to various needs such as communication, coordination, production, and resource sharing. The application is designed for multiple user roles, including students, advisors, and administrative staff, each allocated a personalized workspace. The novelty of the proposed system is its comprehensive coverage of all collaborative aspects mentioned throughout the FYP process, including proposal processing, project assignment, project completion, and evaluation. The research contributes to fostering innovation in machine learning projects by effectively managing and sharing datasets through collaboration tools. The results indicate good scores in improving collaborative aspects with a score of 98% for virtualization in coordination and 96% for communication. The results also showed that surveyed users are positively inclined to use the system as their final year project (FYP) management system, with an attention-to-use score of 90% of advisors and 92.8% of students.
A cooperative approach based on local detection of similarities and discontinuities for brain MR images segmentation
(2020) Bennai, Mohamed Tahar; Smaine, Mazouzi; Guessoum, Zahia; Mezghiche, Mohamed; Cormier, Stephane
This paper introduces a new cooperative multi-agent approach for segmenting brain Magnetic Resonance Images (MRIs). MRIs are manually processed by human radiology experts for the identification of many diseases and the monitoring of their evolution. However, such a task is time-consuming and depends on expert decision, which can be affected by many factors. Therefore, various types of research were and are still conducted to automate MRI processing, mainly MRI segmentation. The approach presented in this paper, without any parametrization or prior knowledge, uses a set of situated agents, locally interacting to segment images according to two main phases: the detection of discontinuities and the detection of similarities. An implementation of this approach was tested on phantom brain MR images to assess the results and prove its efficiency. Experimental results ensure a minimum of 89% Dice coefficient with increasing values of the noise and the intensity non-uniformity
Development of a self-adaptive multi-agent system for medical image processing
(Université M'hamad Bougara : Faculté des Sciences, 2022) Bennai, Mohamed Tahar; Mezghiche, Mohamed(Directeur de thèse)
L'imagerie m edicale fournit une repr esentation visuelle des structures ou des activit es du corps humain selon di erentes modalit es anatomiques et fonctionnelles. L'une des m ethodes de traitement les plus couramment utilis ees sur ce type d'images est la segmentation. La segmentation est un processus lors duquel une image est divis ee en un ensemble de r egions d'int er^et. La complexit e de l'anatomie humaine et les artefacts d'acquisition des images m edicales complexi e grandement la segmentation de ces derni eres. Ainsi, plusieurs solutions ont et e propos ees pour automatiser la segmentation des images. Cependant, la plupart des solutions existantes utilisent des connaissances a priori et/ou n ecessitent une forte interaction avec l'utilisateur pour r ealiser correctement cette t^ache. Dans cette th ese, nous proposons plusieurs approches multi-agents pour l'automatisation de la segmentation d'images m edicales. Ces approches, utilisant un algorithme de croissance de r egions modi e, sont bas ees sur des agents autonomes et interactifs coop erant au sein de l'image a n de correctement la segmenter. Dans un premier temps, une approche a base de r egion utilise un ensemble d'agents mobiles pour explorer l'image et d etecter les r egions homog enes qui la composent. Lors du processus de d etection des r egions, chaque agent une m ethode de croissance des r egions qui introduit l'emploi de la valeur du gradient lors de l' evaluation des similarit es. Cette m ethode est ex ecut ee de mani ere coop erative par plusieurs g en erations d'agents jusqu' a ce que l'ensemble de l'image soit trait ee. Cette approche fut test ee sur un ensemble d'IRM c er ebrale avec di erent niveau de d et erioration. Les r esultats montrent que l'interaction entre la population d'agents o re une e cacit e certaine pour la segmentation des tissus c er ebraux dans des IRM saines. Par la suite, une autre approche appel ee MLISS et utilisant simultan ement les propri et es de similarit e et de discontinuit e de l'image pour la d etection des r egions est pr esent ee. Contrairement a l'approche pr ec edente, MLISS utilise deux ensembles distincts d'agents. Le premier ensemble d'agents a pour but de pr eparer la d etection des noyaux de r egions, quand le second groupe d'agents utilise une m ethode de croissance de r egion pour d etecter les r egions nales. Cette nouvelle architecture permet d'am eliorer les r esultats de segmentation de r egions compactes comme ce fut le cas lors de la segmentation des zones de mati ere blanche dans des images IRM c er ebrales. Pour nir, une nouvelle approche multi-agents, inspir ee des deux pr ec edentes et baptis ee MAMES, est propos ee pour la segmentation de r egions tumorales dans des images IRM c er ebrales 3D. MAMES a h erit e de l'architecture a deux ensembles d'agents de MLISS. De ce fait, la premi ere population d'agents permet le placement des germes de r egions et la croissance de ces derni eres, tandis que la seconde population interagit et collabore pour permettre la nalisation de la segmentation en fusionnant les r egions sur-segment ees. Les exp erimentations men ees sur des IRM c er ebrales saines et pathologiques ont fourni des r esultats prometteurs, d emontrent ainsi l'e cacit e de notre m ethode, notamment pour la segmentation des tumeurs
A generic Multi-Agent framework for Medical-Image segmentation
(2017) Bennai, Mohamed Tahar; Guessoum, Zahia; Mazouzi, Smaine; Cormier, Stéphane; Mezghiche, Mohamed
Multi-agent medical image segmentation : a survey
(Elsevier, 2023) Bennai, Mohamed Tahar; Guessoum, Zahia; Mazouzi, Smaine; Cormier, Stéphane; Mezghiche, Mohamed
During the last decades, the healthcare area has increasingly relied on medical imaging for the diagnosis of a growing number of pathologies. The different types of medical images are mostly manually processed by human radiologists for diseases detection and monitoring. However, such a procedure is time-consuming and relies on expert judgment. The latter can be influenced by a variety of factors. One of the most complicated image processing tasks is image segmentation. Medical image segmentation consists of dividing the input image into a set of regions of interest, corresponding to body tissues and organs. Recently, artificial intelligence (AI) techniques brought researchers attention with their promising results for the image segmentation automation. Among AI-based techniques are those that use the Multi-Agent System (MAS) paradigm. This paper presents a comparative study of the multi-agent approaches dedicated to the segmentation of medical images, recently published in the literature
Réalisation d’un système multi-agents adaptatifs pour l’imagerie médicale
(2013) Bennai, Mohamed Tahar
A stochastic multi-agent approach for medical-image segmentation: Application to tumor segmentation in brain MR images
(ELSEVIER, 2020) Bennai, Mohamed Tahar; Guessoum, Zahia; Mazouzi, Smaine; Cormier, Stéphane; Mezghiche, Mohamed
According to functional or anatomical modalities, medical imaging provides a visual representation of complex structures or activities in the human body. One of the most common processing methods applied to those images is segmentation, in which an image is divided into a set of regions of interest. Human anatomical complexity and medical image acquisition artifacts make segmentation of medical images very complex. Thus, several solutions have been proposed to automate image segmentation. However, most existing solutions use prior knowledge and/or require strong interaction with the user. In this paper, we propose a multi-agent approach for the segmentation of 3D medical images. This approach is based on a set of autonomous, interactive agents that use a modified region growing algorithm and cooperate to segment a 3D image. The first organization of agents allows region seed placement and region growing. In a second organization, agent interaction and collaboration allow segmentation refinement by merging the over-segmented regions. Experiments are conducted on magnetic resonance images of healthy and pathological brains. The obtained results are promising and demonstrate the efficiency of our method.
Towards a generic Multi-agent approach for medical image segmentation
(Springer, 2017) Bennai, Mohamed Tahar; Guessoum, Zahia; Mazouzi, Smaine; Cormier, Stéphane; Mezghiche, Mohamed