Features Fusion Using Belief Functions Theory for ARDS Prediction
Aline Taoum1,2,
Farah Mourad-Chehade1, and
Hassan Amoud2
1.Laboratoire de Modélisation et Sûreté des Systèmes, Institut Charles Delaunay, Université de Technologie de Troyes, Troyes, France
2.Azm Platform for Research in Biotechnology and its Applications, EDST, Lebanese University, Tripoli, Lebanon
2.Azm Platform for Research in Biotechnology and its Applications, EDST, Lebanese University, Tripoli, Lebanon
Abstract—Information fusion techniques are at high interest with the increase in dimensionality of the data being handled. They are applied in different applications, such as in the biomedical domain. This paper proposes an information fusion model that predicts the occurrence of ARDS using vital signs. This model uses features fusion based on the belief functions theory. Different linear and nonlinear parameters are first extracted from the signals, and a parameters selection procedure is proposed to select only pertinent parameters. These parameters are then used to construct mass functions in the belief functions framework. Afterwards, the prediction is performed in real-time by combining all the constructed mass functions. Results present the effectiveness of the belief theory predicting ARDS using the MIMIC II public database.
Index Terms—acute respiratory distress syndrome, belief functions theory, features fusion, linear and non-linear parameters
Cite: Aline Taoum, Farah Mourad-Chehade, and Hassan Amoud, "Features Fusion Using Belief Functions Theory for ARDS Prediction," International Journal of Signal Processing Systems, Vol. 7, No. 4, pp. 107-112, December 2019. doi: 10.18178/ijsps.7.4.107-112
Index Terms—acute respiratory distress syndrome, belief functions theory, features fusion, linear and non-linear parameters
Cite: Aline Taoum, Farah Mourad-Chehade, and Hassan Amoud, "Features Fusion Using Belief Functions Theory for ARDS Prediction," International Journal of Signal Processing Systems, Vol. 7, No. 4, pp. 107-112, December 2019. doi: 10.18178/ijsps.7.4.107-112
Previous paper:First page
Next paper:Development of Dementia Neurofeedback System Using EEG Brainwave Signals
Next paper:Development of Dementia Neurofeedback System Using EEG Brainwave Signals