The paper “Personalised Gait Recognition for People with Neurological Conditions ”, authored by LASIGE’s members Leon Ingelse (MSc Student), Diogo Branco (Ph.D. student), and Tiago Guerreiro (integrated researcher) has been published in the Sensors Journal, a top-ranked journal (h5-index 126 and ranked #5 in the Google Scholar Engineering & Computer Science). The paper’s co-authors are Raquel Bouça-Machado, Joaquim J. Ferreira, the CNS Physiotherapy Study Group from CNS—Campus Neurológico, and Hristijan Gjoreski from the Faculty of Electrical Engineering and Information Technologies.
In the paper, the authors showed the relevance of personalising models for gait recognition of people with neurological conditions. The paper shows a comparative proof-of-concept evaluation with general machine learning (Neural Network – NN and Convolutional Neural Network – CNN) approaches and personalised counterparts showing that the latter improved the overall accuracy. More importantly, participants that were ill-represented by the general model (the most extreme cases) had the largest improvements. It is common to say that people with neurological conditions, such as Parkinson’s disease, present very individual motor patterns, and that in a sense their motor patterns are all outliers. We expect that our results will motivate researchers to explore alternative approaches that value personalisation rather than harvesting datasets that may be able to represent these differences.
The paper is available here.