Frontiers in Neurology (Oct 2023)
Ecological validity of a deep learning algorithm to detect gait events from real-life walking bouts in mobility-limiting diseases
- Robbin Romijnders,
- Robbin Romijnders,
- Francesca Salis,
- Clint Hansen,
- Arne Küderle,
- Anisoara Paraschiv-Ionescu,
- Andrea Cereatti,
- Lisa Alcock,
- Kamiar Aminian,
- Clemens Becker,
- Stefano Bertuletti,
- Tecla Bonci,
- Tecla Bonci,
- Philip Brown,
- Ellen Buckley,
- Ellen Buckley,
- Alma Cantu,
- Anne-Elie Carsin,
- Anne-Elie Carsin,
- Anne-Elie Carsin,
- Marco Caruso,
- Brian Caulfield,
- Brian Caulfield,
- Lorenzo Chiari,
- Lorenzo Chiari,
- Ilaria D'Ascanio,
- Silvia Del Din,
- Silvia Del Din,
- Björn Eskofier,
- Sara Johansson Fernstad,
- Marceli Stanislaw Fröhlich,
- Judith Garcia Aymerich,
- Judith Garcia Aymerich,
- Judith Garcia Aymerich,
- Eran Gazit,
- Jeffrey M. Hausdorff,
- Jeffrey M. Hausdorff,
- Hugo Hiden,
- Emily Hume,
- Alison Keogh,
- Alison Keogh,
- Cameron Kirk,
- Felix Kluge,
- Felix Kluge,
- Sarah Koch,
- Sarah Koch,
- Sarah Koch,
- Claudia Mazzà,
- Claudia Mazzà,
- Dimitrios Megaritis,
- Encarna Micó-Amigo,
- Arne Müller,
- Luca Palmerini,
- Luca Palmerini,
- Lynn Rochester,
- Lynn Rochester,
- Lars Schwickert,
- Kirsty Scott,
- Kirsty Scott,
- Basil Sharrack,
- David Singleton,
- David Singleton,
- Abolfazl Soltani,
- Abolfazl Soltani,
- Martin Ullrich,
- Beatrix Vereijken,
- Ioannis Vogiatzis,
- Alison Yarnall,
- Alison Yarnall,
- Gerhard Schmidt,
- Walter Maetzler
Affiliations
- Robbin Romijnders
- Digital Signal Processing and System Theory, Electrical and Information Engineering, Faculty of Engineering, Kiel University, Kiel, Germany
- Robbin Romijnders
- Arbeitsgruppe Neurogeriatrie, Department of Neurology, Universitätsklinikum Schleswig-Holstein, Kiel, Germany
- Francesca Salis
- Department of Biomedical Sciences, University of Sassari, Sassari, Italy
- Clint Hansen
- Arbeitsgruppe Neurogeriatrie, Department of Neurology, Universitätsklinikum Schleswig-Holstein, Kiel, Germany
- Arne Küderle
- Department of Artificial Intelligence in Biomedical Engineering, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany
- Anisoara Paraschiv-Ionescu
- Laboratory of Movement Analysis and Measurement, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland
- Andrea Cereatti
- Department of Electronics and Telecommunications, Polytechnic of Turin, Turin, Italy
- Lisa Alcock
- Faculty of Medical Sciences, Newcastle University, Newcastle upon Tyne, United Kingdom
- Kamiar Aminian
- Laboratory of Movement Analysis and Measurement, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland
- Clemens Becker
- Gesellschaft für Medizinische Forschung, Robert-Bosch Foundation GmbH, Stuttgart, Germany
- Stefano Bertuletti
- Department of Biomedical Sciences, University of Sassari, Sassari, Italy
- Tecla Bonci
- INSIGNEO Institute for In Silico Medicine, The University of Sheffield, Sheffield, United Kingdom
- Tecla Bonci
- 0Department of Mechanical Engineering, The University of Sheffield, Sheffield, United Kingdom
- Philip Brown
- 1Newcastle upon Tyne Hospitals NHS Foundation Trust, Newcastle upon Tyne, United Kingdom
- Ellen Buckley
- INSIGNEO Institute for In Silico Medicine, The University of Sheffield, Sheffield, United Kingdom
- Ellen Buckley
- 0Department of Mechanical Engineering, The University of Sheffield, Sheffield, United Kingdom
- Alma Cantu
- 2School of Computing, Newcastle University, Newcastle upon Tyne, United Kingdom
- Anne-Elie Carsin
- 3Barcelona Institute for Global Health (ISGlobal), Barcelona, Spain
- Anne-Elie Carsin
- 4Faculty of Health and Life Sciences, Universitat Pompeu Fabra, Barcelona, Spain
- Anne-Elie Carsin
- 5CIBER Epidemiología y Salud Pública, Madrid, Spain
- Marco Caruso
- Department of Electronics and Telecommunications, Polytechnic of Turin, Turin, Italy
- Brian Caulfield
- 6Insight Centre for Data Analytics, University College Dublin, Dublin, Ireland
- Brian Caulfield
- 7School of Public Health, Physiotherapy and Sports Science, University College Dublin, Dublin, Ireland
- Lorenzo Chiari
- 8Department of Electrical, Electronic and Information Engineering “Guglielmo Marconi”, University of Bologna, Bologna, Italy
- Lorenzo Chiari
- 9Health Sciences and Technologies—Interdepartmental Center for Industrial Research (CIRISDV), University of Bologna, Bologna, Italy
- Ilaria D'Ascanio
- 8Department of Electrical, Electronic and Information Engineering “Guglielmo Marconi”, University of Bologna, Bologna, Italy
- Silvia Del Din
- Faculty of Medical Sciences, Newcastle University, Newcastle upon Tyne, United Kingdom
- Silvia Del Din
- 0Translational and Clinical Research Institute, Newcastle University, Newcastle upon Tyne, United Kingdom
- Björn Eskofier
- Department of Artificial Intelligence in Biomedical Engineering, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany
- Sara Johansson Fernstad
- 2School of Computing, Newcastle University, Newcastle upon Tyne, United Kingdom
- Marceli Stanislaw Fröhlich
- 1Grünenthal GmbH, Aachen, Germany
- Judith Garcia Aymerich
- 3Barcelona Institute for Global Health (ISGlobal), Barcelona, Spain
- Judith Garcia Aymerich
- 4Faculty of Health and Life Sciences, Universitat Pompeu Fabra, Barcelona, Spain
- Judith Garcia Aymerich
- 5CIBER Epidemiología y Salud Pública, Madrid, Spain
- Eran Gazit
- 2Center for the Study of Movement, Cognition and Mobility, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel
- Jeffrey M. Hausdorff
- 2Center for the Study of Movement, Cognition and Mobility, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel
- Jeffrey M. Hausdorff
- 3Department of Physical Therapy, Sackler Faculty of Medicine & Sagol School of Neuroscience, Tel Aviv University, Tel Aviv, Israel
- Hugo Hiden
- 2School of Computing, Newcastle University, Newcastle upon Tyne, United Kingdom
- Emily Hume
- 4Department of Sport, Exercise and Rehabilitation, Northumbria University, Newcastle upon Tyne, United Kingdom
- Alison Keogh
- 6Insight Centre for Data Analytics, University College Dublin, Dublin, Ireland
- Alison Keogh
- 7School of Public Health, Physiotherapy and Sports Science, University College Dublin, Dublin, Ireland
- Cameron Kirk
- Faculty of Medical Sciences, Newcastle University, Newcastle upon Tyne, United Kingdom
- Felix Kluge
- Department of Artificial Intelligence in Biomedical Engineering, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany
- Felix Kluge
- 5Novartis Institute of Biomedical Research, Novartis Pharma AG, Basel, Switzerland
- Sarah Koch
- 3Barcelona Institute for Global Health (ISGlobal), Barcelona, Spain
- Sarah Koch
- 4Faculty of Health and Life Sciences, Universitat Pompeu Fabra, Barcelona, Spain
- Sarah Koch
- 5CIBER Epidemiología y Salud Pública, Madrid, Spain
- Claudia Mazzà
- INSIGNEO Institute for In Silico Medicine, The University of Sheffield, Sheffield, United Kingdom
- Claudia Mazzà
- 0Department of Mechanical Engineering, The University of Sheffield, Sheffield, United Kingdom
- Dimitrios Megaritis
- 4Department of Sport, Exercise and Rehabilitation, Northumbria University, Newcastle upon Tyne, United Kingdom
- Encarna Micó-Amigo
- Faculty of Medical Sciences, Newcastle University, Newcastle upon Tyne, United Kingdom
- Arne Müller
- 5Novartis Institute of Biomedical Research, Novartis Pharma AG, Basel, Switzerland
- Luca Palmerini
- 8Department of Electrical, Electronic and Information Engineering “Guglielmo Marconi”, University of Bologna, Bologna, Italy
- Luca Palmerini
- 9Health Sciences and Technologies—Interdepartmental Center for Industrial Research (CIRISDV), University of Bologna, Bologna, Italy
- Lynn Rochester
- Faculty of Medical Sciences, Newcastle University, Newcastle upon Tyne, United Kingdom
- Lynn Rochester
- 1Newcastle upon Tyne Hospitals NHS Foundation Trust, Newcastle upon Tyne, United Kingdom
- Lars Schwickert
- Gesellschaft für Medizinische Forschung, Robert-Bosch Foundation GmbH, Stuttgart, Germany
- Kirsty Scott
- INSIGNEO Institute for In Silico Medicine, The University of Sheffield, Sheffield, United Kingdom
- Kirsty Scott
- 0Department of Mechanical Engineering, The University of Sheffield, Sheffield, United Kingdom
- Basil Sharrack
- 6Department of Neuroscience and Sheffield NIHR Translational Neuroscience BRC, Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, United Kingdom
- David Singleton
- 6Insight Centre for Data Analytics, University College Dublin, Dublin, Ireland
- David Singleton
- 7School of Public Health, Physiotherapy and Sports Science, University College Dublin, Dublin, Ireland
- Abolfazl Soltani
- Laboratory of Movement Analysis and Measurement, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland
- Abolfazl Soltani
- 7Digital Health Department, CSEM SA, Neuchâtel, Switzerland
- Martin Ullrich
- Department of Artificial Intelligence in Biomedical Engineering, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany
- Beatrix Vereijken
- 8Department of Neuromedicine and Movement Science, Norwegian University of Science and Technology, Trondheim, Norway
- Ioannis Vogiatzis
- 4Department of Sport, Exercise and Rehabilitation, Northumbria University, Newcastle upon Tyne, United Kingdom
- Alison Yarnall
- Faculty of Medical Sciences, Newcastle University, Newcastle upon Tyne, United Kingdom
- Alison Yarnall
- 0Department of Mechanical Engineering, The University of Sheffield, Sheffield, United Kingdom
- Gerhard Schmidt
- Digital Signal Processing and System Theory, Electrical and Information Engineering, Faculty of Engineering, Kiel University, Kiel, Germany
- Walter Maetzler
- Arbeitsgruppe Neurogeriatrie, Department of Neurology, Universitätsklinikum Schleswig-Holstein, Kiel, Germany
- DOI
- https://doi.org/10.3389/fneur.2023.1247532
- Journal volume & issue
-
Vol. 14
Abstract
IntroductionThe clinical assessment of mobility, and walking specifically, is still mainly based on functional tests that lack ecological validity. Thanks to inertial measurement units (IMUs), gait analysis is shifting to unsupervised monitoring in naturalistic and unconstrained settings. However, the extraction of clinically relevant gait parameters from IMU data often depends on heuristics-based algorithms that rely on empirically determined thresholds. These were mainly validated on small cohorts in supervised settings.MethodsHere, a deep learning (DL) algorithm was developed and validated for gait event detection in a heterogeneous population of different mobility-limiting disease cohorts and a cohort of healthy adults. Participants wore pressure insoles and IMUs on both feet for 2.5 h in their habitual environment. The raw accelerometer and gyroscope data from both feet were used as input to a deep convolutional neural network, while reference timings for gait events were based on the combined IMU and pressure insoles data.Results and discussionThe results showed a high-detection performance for initial contacts (ICs) (recall: 98%, precision: 96%) and final contacts (FCs) (recall: 99%, precision: 94%) and a maximum median time error of −0.02 s for ICs and 0.03 s for FCs. Subsequently derived temporal gait parameters were in good agreement with a pressure insoles-based reference with a maximum mean difference of 0.07, −0.07, and <0.01 s for stance, swing, and stride time, respectively. Thus, the DL algorithm is considered successful in detecting gait events in ecologically valid environments across different mobility-limiting diseases.
Keywords
- deep learning (artificial intelligence)
- free-living
- gait analysis
- gait events detection
- inertial measurement unit (IMU)
- mobility