Navigation

Christine Martindale, M. Sc.

Research Focus

Objective health data about subjects outside of the laboratory is important in order to analyse symptoms that cannot be reproduced in the laboratory environment. This health data needs to also be clinically applicable and accurate in order to use it for ‘in the wild’ athlete analysis and medical analysis. In order to achieve this large and diverse data sets of human motion recorded using unobtrusive, wearable sensors need to be collected and annotated, as well as generic or easily adaptable algorithms for processing the data need to be developed. Therefore my research focuses on collecting and analyzing diverse and realistic data sets of human motion where ground truth data on a stride level is available, developing smart annotation methods in order to reduce the labeling cost of large data sets, developing generic recognition algorithms which are capable of clinically relevant annotation and testing the applicability of common machine learning algorithms on heterogeneous gait data, specifically gait data from spastic patients.

 

  • Machine learning in mobile gait analysis, especially of patients with spastic gait
  • Semi-supervised methods for annotation of daily activity data for accurate gait analysis, or repetition analysis

  • Since 02/2015
    Ph.D. student in the Digital Sports Group, Pattern Recognition Lab, Friedrich-Alexander-Universität Erlangen-Nürnberg
  • 09/2013 – 09/2014
    M. Sc. Biomedical Engineering at Imperial College LondonThesis in the Non-Invasive Surgery and Biopsy Laboratory.
  • 09/2013 – 01/2013
    M. Sc. Electrical Engineering at the University of Cape Town.Research within the Power Systems Research Group and the Robotics Research Group.
  • 01/2009 – 12/2012
    B. Sc. (Honours) Mechatronics Engineering  at the University of Cape Town
  • 01/2003 – 12/2008
    Secondary school at Dominican Convent High School Harare, Zimbabwe

2019

2018

2017

2016

2014

2013

BA 2019 Brosig, Johanna Semi-supervised techniques for combined activity recognition and gait analysis
MLTS project 2019 Folle, Lukas; Böhner, Steven; Zillig, Tobias Deep learning for HAR
MP 2019 Kratzer, Thilo Auto sk-learn for HAR
MA 2018 Aholt, Katharina A mobile solution for home-based assessment of rhythmic auditory stimulation of gait
BA 2018 Felsheim, Rebecca Power-aware on-line gait-analysis on shoe-
mounted inertial sensor
FP 2018 Junk, Florian Gait analysis in HSP patients
FP 2017 Gottschalk, Tristan Activity recognition
MP 2017 Koch, Tobias LSTMs for activity recognition
MP 2017 Skrecki, Marcel LSTMs for activity recognition
MA 2016 Bloch, Beatrice Classification and automated extraction of clinically relevant gait parameters from spastic gait
BA 2016 Strauß, Martin Comparison of Segmentation and Classification algorithms for gait sequence analysis in Hereditary Spastic Paraplegia Patients
BA 2016 Enzenhöfer, Verena Enhancement of unsupervised learning for the segmentation of cyclic data using Hierarchical Hidden Markov Models
FP 2016 Löber, Patrick Android apps for IMUs in Kivy
FP 2016 Oberländer, Jana Activity recognition
BA 2015 Dück, Kerstin Development of an Automated Training Plan Management Application for Smartphones
BA 2015 Kübler, Jonathan Time Synchronization of Wearable Sensors