Integrated accelerometry-based Fall Detection System
Background:
Falls have an increasing social and economic impact, specially in the developed countries, as life expectancy of our elderly rises dramatically. Fall-related injuries increases morbidity, mortality and premature use of health care services reducing also the time the elderly can live autonomously at their own homes. Fall detection systems may notify emergency responders when no one apart from the injured is present. However, their real-world application is limited by a number of factors such as high false positive rates, low-compliance, poor usability and short battery lifetime. In order to improve these aspects we developed a fall detection algorithm with a low false alarm rate to be used in a miniaturized 3D accelerometer integrated in a belt buckle – the actibelt® – developed at the Sylvia Lawry Center for Multiple Sclerosis Research . The following thesis consist of implementing the fall detection algorithm in the microcontroller of the accelerometer and programming an Android-based application in order to link the actibelt® to a smart phone for fall detection management and notification tasks.
Master thesis 1: Customized Android Interface for a fall detection system based on a smart phone and a wireless acceleration sensor – the actibelt®BLU
Tasks:
Use case/Scenario description. Analyze the possible scenarios in which the fall detector can be used and the interaction with the user in order to define the requirements of the system.
Customize Android ROM (e.g., customize the status bar, restrict specific user applications)
User management (authentification, authorization, privacy management)
User-centered design and Android programming of interface adapted to disabled and elderly people. (It can include voice user interface)
Definition of the communication protocol
Fall alarm management (alarm prioritisation, alarm disabling, audible warning, alarm sending)
Emulator programming
Prerequisites: good knowledge of programming (preferably Android)
Master thesis /Bachelor thesis 2: Real-time implementation of an accelerometry-based fall detection algorithm on the MSP430 microcontroller of the actibelt®BLU
Tasks:
Rewrite/ refine the fall detection algorithm written in R-code into C code
Redesign plug-in interface for the communication protocol. Establish connection between actibelt-BLU and mobile phone (authentification, encryption)
Fall alarm management (according to the model which will be described in Master thesis 1)
Prerequisites: Good knowledge of C
References:
[1] M. Daumer, K. Thaler, E. Kruis, W. Feneberg, G. Staude, and M. Scholz. Steps towards a miniaturized, robust and autonomous measurement device for the long-term monitoring of the activity of patients - ActiBelt. Biomedizinische Technik/Biomedical Engineering. 2007; 52: 149-155.
[2] A new method to estimate the real upper limit of the false alarm rate in a 3D accelerometry-based fall detector for the elderly. C. Soaz, C. Lederer, M. Daumer. In the Proceedings of the 34th IEEE Annual International Conference of the Engineering in Medicine and Biology Society (EMBS). San Diego, California, USA. August 2012. DOI: 10.1109/EMBC.2012.6345915
Supervisors:
Cristina Soaz, M.Eng.,
Lehrstuhl für Datenverarbeitung (Prof. Diepold)
E-Mail: cristina.soaz@tum.de , Tel.: +49 (0)89 289 23608
M.Sc. Munish Jassi,
Chair of Electronic Design Automation (Prof. Schlichtmann)
E-Mail: munish.jassi@mytum.de, Tel.: +49 (89) 289 – 23651
External supervisor :
Dr. Martin Daumer
Scientific Director of Sylvia Lawry Centre – The Human Motion Institute
E-Mail: daumer@slcmsr.org,Tel.: +49 89 2060 269 50
Dr.-Ing. Dmitry Zhelondz
Universität der Bundeswehr München
Technology provider:
Trium Analysis Online GmbH
Advisor:
Prof. Dr.-Ing. Klaus Diepold
Lehrstuhl für Datenverarbeitung