Soutenance de thèse Panagiotis Kostopoulos
M. Panagiotis Kostopoulos soutiendra, en vue de l'obtention du grade de docteur en systèmes d'information de la Faculté d'économie et de management (GSEM), sa thèse intitulée:
From fall detection to stress pattern using smart devices
- Prof. Gilles Falquet (Président du jury)
- Dr. Michel Deriaz (Co-directeur de thèse, University of Geneva)
- Prof. Dimitri Konstantas (Co-directeur de thèse, University of Geneva)
- Prof. Jean-Henry Morin, University of Geneva
- Dr. Eric Gerstel, Clinic La Colline
Smart mobile services and applications are ubiquitous in our lives. The act of taking preventative or necessary medical procedures to improve a person's wellbeing is called healthcare. The use of smart devices is getting more attention in healthcare day by day. Medical applications make smartphones useful tools in the practice of evidence-based medicine at the point of care, in addition to their use in mobile clinical communication.
As people get older, they tend to become more and more vulnerable to physical disabilities and mental illnesses. In order to prevent the deterioration of their quality of life we have invented applications that help elderly to sustain their activities of daily living (ADL). More specifically, we have made research in two important domains of e-health which are the fall detection and the stress detection. The falls and the stress are two of the main health problems that elderly people are facing nowadays. These two serious health problems can cause a wide spectrum of other health related consequences that deteriorate the quality of life of elderly people and make them vulnerable to various health related and so problems.
The purpose of this thesis is the description of the contribution of a fall detection system and a stress detection system in the daily life of elderly people. Firstly, we present a practical real time fall detection system running on a smartwatch called F2D. Falls among older people remain a very important public healthcare issue. In the majority of fall events external support is imperative in order to avoid major consequences. Therefore, the ability to automatically detect these fall events could help reduce the response time and significantly improve the prognosis of fall victims. In F2D data from the accelerometer is collected, passing through an adaptive threshold-based algorithm which detects patterns corresponding to a fall. A decision module takes into account the residual movement of the user, matching a detected fall pattern to an actual fall. Unlike traditional systems which require a base station and an alarm central, F2D works completely independently. To the best of our knowledge, this is the first fall detection system which works on a smartwatch, being less stigmatizing for the end user. The fall detection algorithm has been tested by Fondation Suisse pour les Téléthèses (FST), the project partner who is responsible for the commercialization of our system. Moreover by testing with real data we have a fall detection system ready to be deployed on the market. Finally, the last module of F2D is the location module which makes our system very useful for nursing homes that host elderly people.
Thanks to the knowledge that we acquired by extracting useful information from the sensors of smart devices and more specifically by detecting falls from a smartwatch, we enhanced our know-how analyzing and extracting patterns from raw sensor data. The next implementation of our expertise and second main element of this thesis is the detection of stress patterns by analyzing smartphone data.
Therefore, secondly we present a novel stress detection system which aims to detect stress and burn-out risks by analyzing the behaviors of the users via their smartphone. The main purpose of our stress detection system is the use of the mobile sensor technology for detecting stress. In particular, we collect data from people’s daily phone usage, gathering information about the sleeping pattern, the social interaction and the physical activity of the user. We combine the information gathered from these main dimensions of wellbeing and we provide a relaxation score to the end-user, making him aware about his stress level. To the best of our knowledge, this is the first system that computes a stress score based on different dimensions of human wellbeing. The main innovation of this work is addressed in the fact that the way the stress level is computed is as less invasive as possible. Our solution relies only on the daily phone usage of people. Also we acquire the ground truth for the importance of each dimension of wellbeing for each individual by asking the users. This leads us to a personalized model which focuses on the personality of each individual user. Our stress detection algorithm was the key element of an Active and Assisted Living (AAL) project called StayActive as well and it has been evaluated in a real world environment with people working in the public transportation company of Geneva (Transports Publics Genevois).
Both of the systems that are presented in this thesis have been used in applications that will be available on the market, transferring directly the scientific research into a commercial product. Also both of the systems have been tested with real end-users and therefore the research has gone one step further, behind the lab trials.
Finally, people coming from the research community and the industrial world have shown great interest in our research results. Therefore, our research results led to two new Commission for Technology and Innovation (CTI) projects. We collaborate with one of the biggest clinic groups in Switzerland, Hirslanden, working on a project called Recover@home. The main idea of this project is to build a solution to monitor a patient while at home. Moreover we collaborate with Hirslanden for extending our stress detection system in a project called Stress and Burnout (SaB). The main innovation of SaB will be an algorithm computing a stress level by combining biosignals from a wearable device, behavioral information from a smartphone, as well as subjective answers to standard medical questionnaires.
To recapitulate, in this thesis we present two e-health applications. We begin with a fall detection system and we continue with a stress detection system. Last but not least we present the new research directions and projects that have been created based on our expertise of detecting patterns from raw sensor data, collected from smart devices.
Date: Jeudi 19 octobre 2017 à 10h00
Lieu: Battelle bâtiment A - Salle de réunion 432-433 (3ème étage)