Technology, Robotics and Healthcare
Date: Thursday 1st October 2020. 16:00-17:30
Date: Thursday 1st October 2020. 14:00-15:30
16:00-16:20: Object Localization in Indoor Environments. Daniel Alshamaa and Farah Chehade
Abstract: This paper investigates object localization in indoor environments. This subject is of high interest, especially for its importance in the healthcare domain. The idea of our work is to equip the object to be localized by a WiFi tag and use the access points installed in the indoor environment to determine its position. We tackle the problem as a zoning localization where the objective is to determine the zone where the object resides at any instant. The proposed approach uses the belief functions theory to define an evidence framework for estimating the most probable object's zone. Real experiments demonstrate the effectiveness of this approach as compared to other localization methods.
16:30-16:50: A Detrended Fluctuation Analysis to differentiate between a turn and straight path in the Timed Up and Go test. Dona Bou Zeidan and Aly Chkeir.
Abstract: Frailty can attack older persons at any age and have serious consequences. Because it is a reversible process, detecting it as soon as possible helps the elderly from facing the symptoms. The Gait speed, one of the mobility and frailty indicators, is monitored in-home by a Doppler Radar. Eleven subjects took part in the experiments based on the Timed Up and Go (TUG) test to assess the gait speed. The Detrended Fluctuation Analysis (DFA) is used on the radar signal trying to detect the turn. The T-test proved that there was a significant difference between the straight paths and the turn, but changing the window size of the DFA changed the results of the t-test, this created doubt about several parameters that could affect the DFA like the amplitude of the signal and the window size.
17:00-17:20: Evidential prediction of atrial fibrillation with rejection. Mohamed Mroueh, Farah Chehade and Fahed Abdallah.
Abstract: Prediction of cardiovascular diseases is an important task that can save many lives. This paper study the prediction of the atrial fibrillation, an arrhythmia, by using the physiological signals. The main contribution of this paper is the reject option added to the work developed earlier. Thanks to it, the precision value of our proposed method increased from 57% to 75% at the price of not classifying some subjects.