Brain-machine interface based on transfer-learning for detecting the appearance of obstacles during exoskeleton-assisted walking

Vicente Quiles; Laura Ferrero; Eduardo Iáñez; Mario Ortiz; Ángel Gil-Agudo; José M Azorín

doi:10.3389/fnins.2023.1154480

Brain-machine interface based on transfer-learning for detecting the appearance of obstacles during exoskeleton-assisted walking

Front Neurosci. 2023 Mar 14:17:1154480. doi: 10.3389/fnins.2023.1154480. eCollection 2023.

Authors

Vicente Quiles^{1

2}, Laura Ferrero^{1

2

3}, Eduardo Iáñez^{1

2}, Mario Ortiz^{1

2

3}, Ángel Gil-Agudo⁴, José M Azorín^{1

2

3

5}

Affiliations

¹ Brain-Machine Interface Systems Lab, Universidad Miguel Hernández de Elche, Elche, Spain.
² Instituto de Investigación en Ingeniería de Elche - I3E, Universidad Miguel Hernández de Elche, Elche, Spain.
³ The European University of Brain and Technology (NeurotechEU), European Union.
⁴ Biomechanics Unit of the National Paraplegic Hospital, Toledo, Spain.
⁵ ValGRAI: Valencian Graduated School and Research Network of Artificial Intelligence, Valencia, Spain.

Abstract

Introduction: Brain-machine interfaces (BMIs) attempt to establish communication between the user and the device to be controlled. BMIs have great challenges to face in order to design a robust control in the real field of application. The artifacts, high volume of training data, and non-stationarity of the signal of EEG-based interfaces are challenges that classical processing techniques do not solve, showing certain shortcomings in the real-time domain. Recent advances in deep-learning techniques open a window of opportunity to solve some of these problems. In this work, an interface able to detect the evoked potential that occurs when a person intends to stop due to the appearance of an unexpected obstacle has been developed.

Material and methods: First, the interface was tested on a treadmill with five subjects, in which the user stopped when an obstacle appeared (simulated by a laser). The analysis is based on two consecutive convolutional networks: the first one to discern the intention to stop against normal walking and the second one to correct false detections of the previous one.

Results and discussion: The results were superior when using the methodology of the two consecutive networks vs. only the first one in a cross-validation pseudo-online analysis. The false positives per min (FP/min) decreased from 31.8 to 3.9 FP/min and the number of repetitions in which there were no false positives and true positives (TP) improved from 34.9% to 60.3% NOFP/TP. This methodology was tested in a closed-loop experiment with an exoskeleton, in which the brain-machine interface (BMI) detected an obstacle and sent the command to the exoskeleton to stop. This methodology was tested with three healthy subjects, and the online results were 3.8 FP/min and 49.3% NOFP/TP. To make this model feasible for non-able bodied patients with a reduced and manageable time frame, transfer-learning techniques were applied and validated in the previous tests, and were then applied to patients. The results for two incomplete Spinal Cord Injury (iSCI) patients were 37.9% NOFP/TP and 7.7 FP/min.

Keywords: BMI; EEG; closed-loop; exoskeleton; stopping intention; transfer-learning.

Grants and funding

This research is part of grant RTI2018-096677-B-I00, funded by MCIN/AEI/10.13039/501100011033 and by “ERDF A way of making Europe”. This research has been also funded by the Spanish Ministry of Universities through the Aid for the Training of University Teachers FPU19/03165 and by the Miguel Hernández University of Elche (Spain) through the project “Análisis de la actividad cerebral para tareas de asistencia y rehabilitación con exoesqueletos (EXOBRAIN)” from “Convocatoria de Ayudas a la Investigación 2022”.