Data Science Seminars

Personalised Gait Recognition for People with Neurological Conditions

Transmissão através de Videoconferência

Por Leon Ingelse (LASIGE/DI-FCUL).

There is a growing interest in monitoring gait patterns in people with neurological conditions. The democratization of wearable inertial sensors has enabled the study of gait in free living environments. One pivotal aspect to gait assessment in uncontrolled environments is to accurately recognize gait instances. Previous work has focused on wavelet transform methods or general machine learning models to detect gait; the former assumes a comparable gait pattern between people and the latter assumes training datasets that represent the diverse population. In this paper, we argue that these approaches are unsuitable for people with severe motor impairments and their distinct gait patterns, and make the case for a lightweight personalized alternative. We propose an approach that builds on top of a general model, fine-tuning it with personalized data. A comparative proof-of-concept evaluation with general machine learning (NN and CNN) approaches and personalized counterparts, showed that the latter improved overall accuracy in 3.5 % for the NN; and 5.3 % for the CNN. More importantly, participants that were ill-represented by the general model (the most extreme cases) had the recognition of gait instances improved by up to 16.9 % for NN and 20.5 % for CNN, with the personalized approaches. It is common to say that people with neurological conditions, like Parkinson's disease, present very individual motor patterns, and that in a sense they are all outliers; we expect that our results motivate researchers to explore alternative approaches that value personalization rather than harvesting datasets that are (maybe) able to represent these differences.

Bio: Leon Ingelse is writing his thesis with Alcides Fonseca on Automated Feature Engineering using Genetic Programming. He was born in Amsterdam, the Netherlands, where he also did a Bachelor in Mathematics at the University of Amsterdam. Last year he started the master of Data Science at the Faculty of Science of the University of Lisbon and collaborated with Tiago Guerreiro and Diogo Branco on the personalization of gait recognition for people with neurological conditions.


Transmissão via Zoom.

14h30
Departamento de Informática | Ciências ULisboa

Seminário do Laboratório de Instrumentação e Física Experimental de Partículas, por Jonathan Hollar (LIP).

This colloquium seeks to debate ideas from the arts, humanities and sciences on the process of empathy.

Neuroscience and Neuromodeling Seminar, por Athanasios Vourvopoulos (LaSEEB, ISR, IST-ULisboa).

Banner do evento

No dia 6 de outubro, a ULisboa dá as boas-vindas aos seus estudantes na Cidade Universitária!

Geodiversidade

Palestra no âmbito das comemorações do Dia Internacional da Geodiversidade, por Carlos Marques da Silva (DG Ciências ULisboa).

Banner do evento

Celebrando o início de um novo Ano Académico, esta sessão solene é um importante marco na vida académica da ULisboa e junta toda a sua Comunidade.

Imagem ilustrativa do curso, acompanhada de diversas informações

Um curso dirigido ao público geral e estudantes de arte ou ciência (a partir dos 16 anos), com interesse em desenho de observação, ilustração, e biologia, com ou sem experiência em desenho.

Banner do evento

Talk @LASIGE, por Bruno Loff (Faculdade de Ciências, Universidade do Porto).

Logótipo do evento

The main goal of this annual conference is to expose students, engineers, scientists and the overall academic and scientific communities to the greatest recent advances in the fields of Technology, Health and Biomedical Engineering taking place in some of the most prestigious research institutions.

Banner do evento

Participe até 12 de outubro / Submit you application until 12 October.

Banner do evento

Ciências ULisboa estará presente, de 13 a 16 de outubro, no Festival Internacional de Ciência - FIC.A

Seminário Permanente de Filosofia das Ciências, por James Ladyman (University of Bristol).

Logótipo da iniciativa, sobre imagem alusiva à cibersegurança

A ciência rege-se por diferentes escalas de tempo. Diferentes áreas do conhecimento estudam processos a diferentes escalas: dos nanossegundos aos milhões de anos. Quais são os Andamentos da Ciência?

Cartaz do evento

Um evento repleto de atividades científicas, culturais e educativas, com jogos, demonstrações, palestras, entre muitas outras atividades. 

Research at the Faculty of Sciences

Join us on 26 October 2022 for the 4th edition of Ciências Research Day.