S6: E-Health

13 Jul 2017

9:00

Small Hall

		1. An unobtrusive expert system to detect freezing of gait during daily living in people with Parkinson’s disease
		Lucia Pepa, Lucio Ciabattoni, Luca Spalazzi and Maria Gabriella Ceravolo (Universita Politecnica delle Marche, Italy)
		Parkinson’s disease is a common neurodegenerative disorder causing several motor symptoms. Gait disorders such as festination and freezing of gait are important clinical problems, since, despite their high impact on patients’ quality of life, they are poorly understood and counteracted phenomena due to their episodic nature, heterogeneous manifestation, and drug resistance. Automatic and objective monitoring of gait during daily living may help to increase clinical knowledge about these motor symptoms and to assess the effectiveness of pharmacological and rehabilitation therapies. Despite the effort of researchers in wearable systems for gait monitoring, their effective developing in clinical practice is still absent. An important obstacle to this goal is the aggregation of big data into meaningful information for clinicians. In this paper, an acceptable and usable architecture for 24h gait monitoring is resumed and a solution to obtain aggregate and clinical useful information from collected data is proposed.
		2. Effects of TFD Thresholding On EEG Signal Analysis Based On The Local Rényi Entropy
		Jonatan Lerga, Nicoletta Saulig and Rebeka Lerga (University of Rijeka, Croatia); Zeljka Milanovic (University of Split, Croatia)
		Analysis of electroencephalographic (EEG) signals using both,non-iterative and iterative modifications of the Rényi entropy, called,local or short-term Rényi entropy, often requires thresholding in the,time-frequency domain in order to reduce noise and low-energy,cross-terms prior to detecting the number of components present in,multicomponent EEG signals. However, preset time-frequency threshold,value is often chosen empirically. In this paper, a performance analysis,of the short-term Rényi entropy based method with regards to the chosen,time-frequency threshold is rendered. The method was applied to both,noise-free and noisy real-life EEG signals for left and right leg,movements verifying the method’s sensitivity to time-frequency,distribution (TFD) thresholding.
		3. Methodological approach in prediction of balance with machine learning applied on fMRI data
		Ranjith Steve Sivagnanaselvam and Wolfgang Taube (University of Fribourg, Switzerland); Dominique Genoud (University of Applied Sciences Western Switzerland, Switzerland)
		Machine learning is a useful method to assess biomedical data. However,,due to the complexity of these data, it is important to rely on a robust,methodology. Predictions with machine learning are not only an add-on,to statistical analysis but are a real process of discovering and,establishing clinical validity of the diagnostic techniques. This paper,describes a machine learning approach to predict balance changes in,seniors based on functional magnetic resonance imaging (fMRI) data.,Functional MRI results in a huge amount data containing some noise.,Thus, a noise filtering technique is proposed and subsequent analyses,are conducted with various machine learning algorithms. The best,prediction of the balance ability based on brain activation patterns was,obtained by applying a neural network approach. This resulted in an,area under the curve (AUC) of 0.91. Other learning algorithms such as,the tree ensemble are less predictive but allow tracking and better,understanding of the source of differentiation. The results are based on,a real dataset
		4. Comparison of Numerical Electric Field and SAR Results in Compound and Extracted Eye Models
		Mario Cvetković, Hrvoje Dodig and Dragan Poljak (University of Split, Croatia)
		This paper compares the numerical results for the induced electric field,and the specific absorption rate (SAR) obtained using the hybrid,FEM/BEM method in two detailed models of the human eye. The first model,features the human eye placed in the free space, while the second one,incorporates the eye model in the realistic head model obtained from the,magnetic resonance imaging (MRI) scans. Although the preliminary,analysis showed the similar distribution of the induced electric field,along the pupillary axis obtained in both models, the numerical results,for the SAR showed some discrepancy between the two models, which can be,attributed to the high values of induced field in the corneal and,scleral regions obtained in the compound eye model. The analysis could,provide some helpful insights when carrying out the dosimetric analysis,of the human eye exposed to electromagnetic radiation.
		5. Design and implementation of a children safety system based on IoT technologies
		Fabio Franchi, Fabio Graziosi, Claudia Rinaldi, Leonardo D’Errico and Francesco Tarquini (University of l’Aquila, Italy)
		In this paper a system for increasing children’s safety is proposed. The,focus is on the daily route from home to school and viceversa, assuming,the use of school buses. IoT paradigm is exploited together with,different localization techniques i.e. RFID and GPS, in order to design a,solution for parents willing to make certain of their child’s following,the main steps to school or home, i.e. taking the school bus and,entering school or leaving school and entering the school bus. The,chosen technology is also dependent on the idea of maintaining as low as,possible the overall costs. The proposed solution is discussed in terms,of technologies and architecture and the first prototype is presented.