Palestine (État) | EPFL Graph Search

3autres|la Palestine en tant quÉtat souverain'|la Palestine en tant que région géographique et historique|Palestine (région)|la Palestine en tant que territoires conquis par l’Égypte et la Jordanie de 1948 à 1967 puis par Israël|Territoires palestiniens occupés||Palestine La Palestine, en forme longue lÉtat de Palestine (ar), est un État souverain de jure (loi palestinienne). Située au Moyen-Orient, elle revendique son établissement sur les territoires palestiniens occupés, dont Jérusalem-Est. Anciennement intégré à l'Empire ottoman puis passée sous mandat britannique, l'émergence de la Palestine comme État indépendant est accélérée par le conflit israélo-arabe de 1948. La déclaration du de l'Organisation de libération de la Palestine en exil à Alger a été retenue par l'État de Palestine comme sa déclaration d'indépendance et est reconnue par la Ligue arabe et l'Organisation de la coopération islamique. En 2011, la Palestine adhère à l'UNESCO. Le , l'État de Palestine adhè

Interpreting deep learning models for epileptic seizure detection on EEG signals

David Atienza Alonso, Leila Cammoun, Valentin Alexandre Guy Gabeff, Tomas Teijeiro Campo, Marina Zapater Sancho

While Deep Learning (DL) is often considered the state-of-the art for Artificial Intelligence-based medical decision support, it remains sparsely implemented in clinical practice and poorly trusted by clinicians due to insufficient interpretability of neural network models. We have tackled this issue by developing interpretable DL models in the context of online detection of epileptic seizure, based on EEG signal. This has conditioned the preparation of the input signals, the network architecture, and the post-processing of the output in line with the domain knowledge. Specifically, we focused the discussion on three main aspects: 1) how to aggregate the classification results on signal segments provided by the DL model into a larger time scale, at the seizure-level; 2) what are the relevant frequency patterns learned in the first convolutional layer of different models, and their relation with the delta, theta, alpha, beta and gamma frequency bands on which the visual interpretation of EEG is based; and 3) the identification of the signal waveforms with larger contribution towards the ictal class, according to the activation differences highlighted using the DeepLIFT method. Results show that the kernel size in the first layer determines the interpretability of the extracted features and the sensitivity of the trained models, even though the final performance is very similar after post-processing. Also, we found that amplitude is the main feature leading to an ictal prediction, suggesting that a larger patient population would be required to learn more complex frequency patterns. Still, our methodology was successfully able to generalize patient inter-variability for the majority of the studied population with a classification F1-score of 0.873 and detecting 90% of the seizures.

2021

Chitosan Functionalization: Covalent and Non-Covalent Interactions and Their Characterization

Sandrine Gerber, Céline Marie Anne Journot, Laura Camille Louise Nicolle

Chitosan (CS) is a natural biopolymer that has gained great interest in many research fields due to its promising biocompatibility, biodegradability, and favorable mechanical properties. The versatility of this low-cost polymer allows for a variety of chemical modifications via covalent conjugation and non-covalent interactions, which are designed to further improve the properties of interest. This review aims at presenting the broad range of functionalization strategies reported over the last five years to reflect the state-of-the art of CS derivatization. We start by describing covalent modifications performed on the CS backbone, followed by non-covalent CS modifications involving small molecules, proteins, and metal adjuvants. An overview of CS-based systems involving both covalent and electrostatic modification patterns is then presented. Finally, a special focus will be given on the characterization techniques commonly used to qualify the composition and physical properties of CS derivatives.

2021

Timber Folded Plate Structures – Topological and Structural Considerations

Andrea Stitic, Yves Weinand

This paper analyses the potential of different possible folded form topologies for generating timber folded surface structures. The main advantage of such structures lies primarily in the realm of ecology and sustainability. By offering an integral way of construction, which fulfils both a supporting as well as a covering function, very lightweight structures are achieved. Also, greater degree of prefabrication is possible which leads to reduced overall cost. Timber folded structures consist of a large number of discrete, thin plane elements, mutually connected to form an overall folded surface. Therefore proper edgewise connection details are needed in order to ensure an efficient load bearing system. For structures made of wood products this presents a great challenge using the state-of-the art joining techniques. For this reason, the use of timber folded plates in civil engineering applications has been very limited. However, recently new technical solutions have been proposed for efficient edgewise joining of thin timber panels. In this paper focus is put on integrated mechanical attachment technique which utilises digital prefabrication to integrate connectors through panel geometry. Taking into consideration material, fabrication and connection detail constraints, various topologies are examined for the considered application. Furthermore, structural behaviour of folded systems is studied and three feasible forms are compared by means of Finite Element Analysis. Finally, observations are made on a case study of a built prototype structure and the structural potential of the proposed systems is outlined.

2015

Interpreting deep learning models for epileptic seizure detection on EEG signals

David Atienza Alonso, Leila Cammoun, Valentin Alexandre Guy Gabeff, Tomas Teijeiro Campo, Marina Zapater Sancho

2021