Browsing by Author "Cherifi, Dalila (supervisor)"

Now showing 1 - 9 of 9

3D statistical shape modeling
(2016) Omari, Sabrina; Soual, Imene; Cherifi, Dalila (supervisor)
Statistical shape models (SSMs) have been firmly established as a robust tool for segmentation of images. Widespread utilization of three-dimensional models appeared only in recent years, primarily made possible by breakthroughs in automatic detection of shape correspondences; while 2D models have been in use since the early 1990s. The objective of this project is to build a 3D statistical shape modeling for a given data; the implemented process goes through those basic steps, first collect the given data then apply the alignment algorithm based on the ICP (iterative closest point) method which in turn relies on Procrustes analysis result as a starting point, next we apply fitting algorithm which is also based on ICP. Finally we obtain the model using PCA (principle component analysis). To achieve this work, we have implemented the above process on two different shape models, one tested with the Basel Face Model (BSF) and the other is the femur model data samples from the SICAS (Swiss Institute for Computer Assissted Surgery) Medical Image Repository which is used by the Basel University (Switzerland) for both samples, where these models allow the generation and the exploration of the possible shape variation.
A Comparative analysis of flight control for an autonomous aerial delivery vehicle
(Université M'hamed Bougara Boumerdès: Institue de génie electronic et electric, 2024) Riabi, Hanane; Cherifi, Dalila (supervisor)
This work explores a comparative analysis of fligh tcontro lfo ra nautonomou saeria ldelivry vehicle, specificall yapplie dt omedica lassistanc ei nemergenc yan dnatura ldisaste rscenarios. The study begins with an overview of drone technology, including classification san dapplications. It then delves into mathematical modeling, focusing on quadcopter dynamics and control systems. Two distinct control methodologies, linear PID control and 5th generation of Sliding mode control(Continuous Nonsingular Terminal Sliding Mode contoller), are synthesized and simulated. The comparative analysis evaluated the performance of the linear PID controller and the CNSMC approach for the QBall 2 quadrotor system. We assessed their behavior under normal conditions, and the CNSMC controller demonstrated superior resilience to perturbations compared to the PID controller. This makes CNSMC the preferred choice for real-world applications where stability and adaptability are crucial.
Design and implementation of a quadcopter with image reconstruction and object detection
(2016) Beladjeri, Mohammed Ilyes; Khadraoui, Souleymen; Cherifi, Dalila (supervisor)
Aerial robotics is seen to be one of the most dominant aspects in several areas in the last few years, and one special type of the flying robots are quadcopters, which are of an important use nowadays to ease the way of our life. This project is about building a quadcopter, and implementing a control algorithm to stabilize the system, the second aspect is to develop a ground control station equipped with two image processing applications using Emgu.CV. The GCS communicates with the quadcopter in order to take aerial pictures and perform image reconstruction from the captured ones, the second application is to detect a specific object in the reconstructed image or from any other captured scene.
EEG signal feature extraction and classification for epilepsy detection
(2020) Falkoun, Nousaaiba; Ouakouak, Ferial; Cherifi, Dalila (supervisor)
Epilepsy is a neurological disorder of the central nervous system, characterized by sudden seizures caused by abnormal electrical discharges in the brain. Electroencephalogram (EEG) is the most common technique used for Epilepsy diagnosis. Generally, it is done by the manual inspection of the EEG recordings of active seizure periods (ictal). Several techniques have been proposed throughout the years to automate this process. In this study, we have used three different approaches to extract features from the filtered EEG signals. The first approach was to extract eight statistical features directly from the time-domain signal. In the second approach, we have used only the frequency domain information by applying the Discrete Cosine Transform (DCT) to the EEG signals. In the last approach, we have used a tool that combines both time and frequency domain information, which is the Discrete Wavelet Transform (DWT). Six different wavelet families have been tested with their different orders resulting in 37 wavelets. The extracted features are then fed to three different classifiers k-Nearest Neighbor (k-NN), Support Vector Machine (SVM), and Artificial Neural Network (ANN) to perform two binary classification scenarios: healthy versus epileptic (mainly from interictal activity), and seizure-free versus ictal. We have used a benchmark database, the Bonn database, which consists of five different sets. In each scenario, we have taken different combinations of the available data. For Epilepsy detection (healthy vs epileptic), the first approach performed badly. Using the DCT improved the results, but the best accuracies were obtained with the DWT-based approach. For seizure detection, the three methods had a good performance. However, the third method had the best performance and was better than many state-of-the-art methods in terms of accuracy. After carrying out the experiments on the whole EEG signal, we separated the five rhythms and applied the DWT on them with the Daubechies7 (db7) wavelet for feature extraction. We have observed that close accuracies to those recorded before can be achieved with only the Delta rhythm in the first scenario (Epilepsy detection) and the Beta rhythm in the second scenario (seizure detection).
Implementation of determinisic and probabilisic fiber tracking algorihms for abnormal brain tissues analysis using dMRI
(Université M’Hamed BOUGARA de Boumerdes : Institut de génie electrique et electronique (IGEE), 2019) Moussaoui, Imane; Cherifi, Dalila (supervisor)
Diﬀusion magnetic resonance imaging (DMRI) is a technique that allows to probe the microstructure of materials. In our case we use it for the White Matter (WM) while tractography is a computational reconstruction method based on diﬀusion-weighted mag- netic resonance imaging (DWI)that attempts to reveal the trajectories of white matter pathways in vivo and to infer the underlying structural connectome of the human brain. The aim of our study is to reach the best reconstruction of the WM in the presence of abnormal tissues such as Astrocytoma type II and III, Glioblastoma Multiform, Menin- gioma and Oligodendrocytoma type II. For that purpose, nine data about the mentioned diseases aquired from the the UK data archive are utilised, the procedure is to apply both deterministic and probabilistic methods with two stopping criteria for each to the dataset. The analysis of the four outputs is conducted for each patient to assess the results in the region of interest (ROI). Besides the comparison between the tracts generated with the probabilistic and the deter- ministic algorithms, another comparison is performed for FA=0.2 and FA=0.4 as stopping criteria and their eﬀect on the generated ﬁbers. The main contribution of this work is the implementation of the probabilistic tracking algorithm. While searching for information concerning tractography .It is found that de- terministic tractography is widely used because of its ease and simplicity. In this repport advantages of using the probabilistic method for better results demonstrated therefore both methods were applied on the same dataset in addition to analising the eﬀect of stopping criterion on the results in the ROI and the whole brain.
Implementation of determinisic and probabilisic fiber tracking algorihms for abnormal brain tissues analysis using dMRI
(2019) Moussaoui, Imane; Cherifi, Dalila (supervisor)
Diffusion magnetic resonance imaging (DMRI) is a technique that allows to probe the microstructure of materials. In our case we use it for the White Matter (WM) while tractography is a computational reconstruction method based on diffusion-weighted mag-netic resonance imaging (DWI)that attempts to reveal the trajectories of white matter pathways in vivo and to infer the underlying structural connectome of the human brain. The aim of our study is to reach the best reconstruction of the WM in the presence of abnormal tissues such as Astrocytoma type II and III, Glioblastoma Multiform, Menin- gioma and Oligodendrocytoma type II. For that purpose, nine data about the mentioned diseases aquired from the the UK data archive are utilised, the procedure is to apply both deterministic and probabilistic methods with two stopping criteria for each to the dataset. The analysis of the four outputs is conducted for each patient to assess the results in the region of interest (ROI). Besides the comparison between the tracts generated with the probabilistic and the deter- ministic algorithms, another comparison is performed for FA=0.2 and FA=0.4 as stopping criteria and their effect on the generated fibers. The main contribution of this work is the implementation of the probabilistic tracking algorithm. While searching for information concerning tractography .It is found that de-terministic tractography is widely used because of its ease and simplicity. In this repport advantages of using the probabilistic method for better results demonstrated therefore both methods were applied on the same dataset in addition to analising the effect of stopping criterion on the results in the ROI and the whole brain.
Lung Diseases Segmentation Using Deep Learning Algorithms
(Université M’Hamed bougara : Institute de Ginie électric et électronic, 2023) Tahar, Abdelmadjid; Cherifi, Dalila (supervisor)
Lung diseases pose signifi cant challenges in diagnosing and treating patients, necessitating ac-curate and effi cient analysis of lung images to enable early detection and effective care. This research aims to address the labor-intensive and time-consuming process of detecting lung ab-normalities by employing deep learning-based segmentation algorithms. The objective is to evaluate and compare the performance of advanced models, including U-Net, U-Net++, U-Net 3+, ResU-Net, and Attention U-Net, in automating the identifi cation and delineation of lung abnormalities. The models were trained from scratch, and data preprocessing techniques such as contrast limited adaptive histogram equalization and data augmentation were implementedto enhance the results. The fi ndings consistently demonstrate that the U-Net 3+ architecturesurpasses the other models, exhibiting superior accuracy in segmentation. It achieved state-of-the-art performance with a dice score of 87.61% in COVID-19 segmentation and 89.45in lung tumor segmentation. These results showcase the potential of U-Net 3+ as a promisingsolution for automating the detection of lung abnormalities.Keywords: Image segmentation, Deep Learning, COVID-19, Lung Tumor, CT scans.
Tissue characterization of brain using MRI
(2016) Bentaleb, Djouhra Lyza; Larabi, Farida; Cherifi, Dalila (supervisor)
In the evaluation of healthcare services, there is an increasing need for effective use of imaging data in medical diagnosis, the analysis and study of the brain is of great interest due to its potential for studying early growth tumor which is an abnormal mass of tissue found in the brain. The purpose of the project is to characterize an abnormal tissue of brain MRI Imaging, especially the brain tumor. For that, we have firstly used different methods of segmentation which are thresholding, K-means and fuzzy K-means in order to extract the tumor, secondly 3D reconstruction of the tumor has been involved by applying the pre-processing and post-processing steps to get the better image’s results which are inputted as a 2D parallel slices in the marching cube algorithm to form a 3D tumor’s volume. After that the texture analysis is presented as a useful computational method for discriminating between pathologically different regions on medical images because it has been proven to perform better than human eyesight at discriminating certain classes of texture. Finally, an experimental part has been done on where the results of the extraction’s segmentation and texture analysis are discussed, in addition 3D reconstruction is implemented.
Wireless speech recognition with WIFI environment monitor
(2019) Belacel, Khalil; Cherifi, Dalila (supervisor)
This report discloses an attempt of creating a complete technical system that based mainly on speech recognition technology and environment monitor which can be used by people with physical disability, using technology that we have today is powerful tool which will allow us to accomplish and construct this system. The aim of the system is giving a disabled person the ability of controlling some surrounding devices such as lightning, air conditioner and other domestic devices which require some basic body movements which cannot be performed by a person who has special type of injury like quadriplegia or paraplegia, and that ability is given by using speech recognition system, and also an environment monitor which can be used to monitor the environment parameters in the room where the patient resides. Additionally, to all that is to encourage them morally which can be done by giving them the feeling that they are somewhat self-dependent to perform something without external intervention. This work can contribute in finding better and efficient solutions to be used by this category of people and also to eliminate the lack of the awareness about the special needs and care that they deserve.