Researchers in Telecom division propose topics for students interested by following training in CDTA. These subjects fit with the themes addressed within their research teams. Title 1: Web-CAD: Online Re-ranking Approach for Brain Tumor Diagnosis Supervier: Aouache Mustapha Team : Advanced Information Systems (SIA) Abstract : Accurate diagnosis is essential for the successful treatment of the brain tumor. As a result, in this work, we propose a Web-computer-assisted intelligent (Web-CAD) system that retrieves similar brain magnetic resonance images from a medical database to assist the radiologist in diagnose the brain tumor. Firstly, the proposed CAD system uses a two-step approach to retrieve similar MR images. The first step classifies the query image as benign or malignant using the features that discriminate the classes. The second step then retrieves the most similar images within the predicted class using the features that distinguish the subclasses. In order to provide faster image retrieval, we investigate of use of an indexing method to be investigated that can groups’ subclass features into clusters using a clustering approach and separately reduces the dimensionality of each cluster. The proposed CAD system should also make robust against misalignment that occurs during MR image acquisition. Experiments should be carried out on a database consisting of MR images of the brain tumor and compared with the state-of-the-art works. At the end, an integration of the proposed CAD tumor system with Web application should be performed during this study to produce an Online Re-ranking Approach for Brain Tumor Diagnosis. The experimental results should demonstrate that the effectiveness of the proposed system show the viability of clinical application. Title 2: Web-CAD: Retinal Vessel Segmentation for Diabetic Retinopathy Assessment Supervier: Aouache Mustapha Team : Advanced Information Systems (SIA) Abstract : The early detection and treatment of prevalent eye diseases, particularly diabetic retinopathy (DR), glaucoma, and age-related macular degeneration (AMD) have a significant impact on reducing the incidence of blindness and blindness worldwide. Thus, this work will be addressed to develop and new Web-CAD computerized image analysis platform to help automate the diagnosis via Retinal Vessel segmentation for Diabetic Retinopathy Assessment (DRA). To achieve this objectively, three main processes need to examine. First, the segmentation process needs to be developed to extract correctly the retinal vessel. Then, a content-based image retrieval (CBIR) process need be examined via retrieving related images from large database collections using their pictorial content is examined. The retina vessel segmentation results will be indexed via content feature list that becomes the index for storage, search, and retrieval of related images from a dataset upon specific visual characteristics. Next, retina with segmented vessel presented with/without disease will be presented in the retrieval module wherein images are measured and ranked based on the minimum distance. Retrieval performance also will be evaluated using different measures such as recall and precision, average retrieval rate (ARR), F1 or any other related approaches. In the end, the three modules will be integrated to produce the medical computer-assisted system for diseases’ detection and diagnosis. The proposed WebCAD-DRA system should be made robust against misalignment that occurs during image acquisition. Experiments were carried out on a database consisting of retina medical dataset of the images such DRIVE, START, MESSIDOR. The experimental results should demonstrate the effectiveness of the proposed system and show the viability of clinical application. Title 3: Development Of Web-CAD Platform For Clinical Image Analysis And Diagnostic Supervier: Aouache Mustapha Team : Advanced Information Systems (SIA) Abstract : This work focuses on the development of an intelligent Web-CAD platform for clinical image analysis and diagnostic from a medical database to assist the radiologist in the diagnosis of Breast Cancer. A feature vector will not perform well for finding similar images in the medical domain as images within the same disease class differ by severity, density and other such factors. To solve this issue, the proposed Web-CAD platform-based content-based image retrieval (CBIR) system uses a two-step approach to retrieve similar lesion images. First step, classifies query image with/without lesions using the features that discriminate the classes. The second step then retrieves the most similar images within the predicted class using the features that distinguish the subclasses. If the search process appear slower, an indexing method based principal component analysis (PCA) will be investigated which groups subclass features into clusters using K-means clustering and separately reduces the dimensionality of each cluster. The reduced feature set is then indexed using a instigate method. The proposed WebCAD-IAD system should be made robust against misalignment that occurs during image acquisition. Experiments were carried out on a database consisting of medical dataset of the images presented with /without lesions. The experimental results should demonstrate the effectiveness of the proposed system and show the viability of clinical application. Title 4: Decision support system: Implementation of a Data Lake Supervier: Semar Bitah Kahina, Asma Manal Kherchi Team : Advanced Information Systems (SIA) Abstract : Decision Support Systems have a major impact on the management of different organizations and provide significant assistance in steering their action plans by allowing decision-making analyzes. Currently, conventional DSSs face difficulties in making good decisions, this is due to the diversity of information sources, the data has become heterogeneous and highly scalable. In order to modernize these systems, we propose a Master thesis Project which will give rise to the implementation of a DSS using a DATA LAKE that represents a new emerging concept of Big Data, it manipulates a large mass of data where they are subjected to very little restriction in terms of their structures. This scalable system will be related to the field of hydraulics in Algeria with the aim of establishing a steering dashboard, offering a new solution to gain agility. Title 5:Decision support system using machine learning techniques: application to the field of hydraulics Supervier: Semar Bitah Kahina, Asma Manal Kherchi Team : Advanced Information Systems (SIA) Abstract : With the emergence of artificial intelligence in the field of data analysis and decision support, techniques such as machine learning and predictive analysis have become essential, in order to gain responsiveness and increase profits. In this context, we propose Master thesis Project which will give rise to the implementation of a Decision Support System using the techniques of Machine Learning. This system will relate to the field of hydraulics, used in order to facilitate the management of water drilling bases in the south of Algeria.