March 21st - Transfer learning approach for fall d


Joana Silva (Fraunhofer Portugal / FEUP)

Transfer learning approach for fall detection with the FARSEEING real-world dataset and simulated falls

Abstract:

Falls are very rare and extremely difficult to acquire in free living conditions. Due to this, most of prior work on fall detection has focused on simulated datasets acquired in scenarios that mimic the real-world context, however, the validation of systems trained with simulated falls remains unclear. This work presents a transfer learning approach for combining a dataset of simulated falls and non-falls, obtained from young volunteers, with the real-world FARSEEING dataset, in order to train a set of supervised classifiers for discriminating between falls and non-falls events. The objective is to analyze if a combination of simulated and real falls could enrich the model. This work was developed in the project Symbiotic technology for societal efficiency gains: Deus ex Machina, NORTE-01-0145-FEDER-000026.

--

Bio: MSc Joana Silva obtained the degree of Master in Bioengineering from the Faculty of Engineering of the University of Porto in 2013. During the Master, she has worked in projects related to physical activity monitoring and gait data analysis using wearable devices, and her master thesis focused in the area of physical activity monitoring using smartphones. Since 2013 she is a researcher at AICOS research center from the Association Fraunhofer Portugal Research in the area of Falls and Activity Monitoring. The work developed under the master thesis was reward with the Young Scientist Best Paper Award in the p-Health Conference 2014 in Vienna. She is now a PhD candidate in the doctoral program in Biomedical Engineering, at FEUP. Her main research areas are related to Biomedical Instrumentation, Signal Processing and Machine Learning.