Browsing by Author "Cherifi, Dalila ( supervisor)"

Now showing 1 - 5 of 5

EEG signal classification and forecasting for epileptic seizure prediction
(2019) Afoun, Laid; Iloul, Zakaria; Cherifi, Dalila ( supervisor)
EEG signal recordings are increasingly replacing the old methods of diagnosis in medical field of many neurological disorders. Our contribution in this project is the study and development of EEG signal classification and forecasting algorithms for epilepsy diagnosis using machine learningusing one rhythm; for classification, an optimum classifier is proposed with only one used rhythm so that both execution time and number of features are reduced; for forecasting, the value of RMSE is minimized when using LSTM where the best hyperparameters are found. Firstly, we used wavelet packet decomposition (WPD) to extract the five rhythms of brain activity from the public Epilepsy-EEG recordings in order to represent each signal with features vector; then we applied on it the well-known classification methods. Secondly, we implemented forecasting methods for predicting seizures states on the signals using statistics methods and LSTM. A statistical study is done to validate the different algorithms.
Ground roll attenuation using (f, k) and agora filters
(2018) Nekhoul, Mohamed Salah; Badouchi, Yassine; Cherifi, Dalila ( supervisor)
The seismic method is the most used method to capture and study the subsurface. The discipline of subsurface seismic imaging, or mapping the subsurface using seismic waves, takes a remote sensing approach to probe the Earth’s interior.The seismic method goes over three major phases: data acquisition, processing, and the interpretation processes. During the process of the acquisition and the transmission, the data is often corrupted by the different types of noises. Throughout our work, we focused on the study of the two dimensions’ filters (f, k) and AGORA.
Melanoma identification using convolutional neural networks
(2018) Louifi, Akram; Soulami, Ameur; Cherifi, Dalila ( supervisor)
Melanoma is an extremely dangerous type of skin cancer causing fatal incidences, it’s also an increasing form of cancer around the world. Since the odds of recovering for the early-diagnosed cases is very high, early detection of melanoma is vital. Computer assisted diagnosis have been used alongside traditional techniques so as to improve the reliability of detecting melanoma. In this project, a convolutional Neural network model designed from scratch as well as Transfer Learning using the pretrained model Inception v3 are used in order to develop a reliable tool able to detect melanoma that can used by
The Performance improvement using CNN model for melanoma classification
(2019) Boumali, khaled; Bourtache, Mohamed islam; Cherifi, Dalila ( supervisor)
Melanoma is a malignant skin cancer with an increasing incidence. To reduce mortalitiy rates due to melanoma, early detection must be taken into consideration. One of the fastest and most useful ways to achieve early detection is to go through Deep learning, and more speci?cically CNN model whose ouput classifies whether the patient is suffereing from melanoma or not. For a better detection, the accuracy of the CNN output has to be high enough so that the patient gets a true result about their state. One of the the hyper-paramters of the CNN model, which leads to more accuracte results, is to add more hidden layers to our model ,at the same time, apply the data augmentation technique for a more perfromant model, and that is what we did in the first part of our project In computer vision, transfer learning is usually expressed through the use of pre-trained models. A pre-trained model is a model that was trained on a large benchmark dataset to solve a problem similar to the one that we want to solve, we want to make a tool capable of detecting melanoma with a higher accuracy than the previous one. In this project, we also focused nusing thepretrained models (VGG, ResNet, Inception and Xception) we then compared the results of our work with the existing works.
Tractography of white matter fibers in presence of astrocytoma
(2018) Benouadah, Sara; Essaheli, Mohamed ebderaouf; Cherifi, Dalila ( supervisor)
Diffusion Magnetic Resonance Imaging (dMRI), a technique that maps the axonal microstructure of the brain, is often used nowadays to investigate white matter alterations. Clinicians have gained useful insights from these studies for surgical planning and demonstrating subtle abnormalities in a variety of diseases. Our work aims to identify the effects of astrocytoma, a type of tumor that affects the brain, on white matter (WM) tracts. For this purpose, three patients with different grades of astrocytomas, acquired by the UK Data Archive, are used. Constrained Spherical Deconvolution (CSD)-based deterministic tractography is applied on these datasets in order to assess the tumor-