Development of a Low-Cost EEG-Controlled Hand Exoskeleton 3D Printed on Textiles

Abstract

Stroke survivors can be affected by motor deficits in the hand. Robotic equipment associated with brain-machine interfaces (BMI) may aid the motor rehabilitation of these patients. BMIs involving orthotic control by motor imagery practices have been successful in restoring stroke patients' movements. However, there is still little acceptance of the robotic devices available, either by patients and clinicians, mainly because of the high costs involved. Motivated by this context, this work aims to design and construct the Hand Exoskeleton for Rehabilitation Objectives (HERO) to recover extension and flexion movements of the fingers. A three-dimensional (3D) printing technique in association with textiles was used to produce a lightweight and wearable device. 3D-printed actuators have also been designed to reduce equipment costs. The actuator transforms the torque of DC motors into linear force transmitted by Bowden cables to move the fingers passively. The exoskeleton was controlled by neuroelectric signal-electroencephalography (EEG). Concept tests were performed to evaluate control performance. A healthy volunteer was submitted to a training session with the exoskeleton, according to the Graz-BCI protocol. Ergonomy was evaluated with a two-dimensional (2D) tracking software and correlation analysis. HERO can be compared to ordinary clothing. The weight over the hand was around 102 g. The participant was able to control the exoskeleton with a classification accuracy of 91.5%. HERO project resulted in a lightweight, simple, portable, ergonomic, and low-cost device. Its use is not restricted to a clinical setting. Thus, users will be able to execute motor training with the HERO at hospitals, rehabilitation clinics, and at home, increasing the rehabilitation intervention time. This may support motor rehabilitation and improve stroke survivors life quality.

Keywords: 3D printing; brain-machine interface; hand exoskeleton; post-stroke; rehabilitation; soft robotics; textiles.

Figure 1

Three-dimensional (3D) design of Hand Exoskeleton for Rehabilitation Objectives (HERO). Isometric view at left and top view at right.

Figure 2

Fabric samples. (A) T1 sample. (B) T2 sample. (C) T3 sample.

Figure 3

Printing tests. (A) All three samples attached to the heated bed. Samples T3 (B), T2 (C), and T1 (D) were analyzed in a stereoscopic microscope.

Figure 4

Final prototype of the exoskeleton's glove.

Figure 5

Cable-driven mechanisms. (A) The absence of a mandrel to wrap the cable can cause it to flex under compression. This mechanism is not dimensionally efficient by virtue of its long length. (B) The Concertina mechanism, which consists of six limiters to prevent cables from bending.

Figure 6

Actuator three-dimensional (3D) model. (1) DC motor; (2) universal couplings for 6 mm (for the lead screw) and 4 mm (for the DC motor); (3) main nut; (4) concertina mechanism; (5) stainless steel M6x1 lead screw; (6) actuator base; (7) bearing mandrel; (8) ball bearing; (9) cable housing entries (*auxiliary view of a model adaptation for use with only one cable); (10) cable housing fixation screw; (11) pin holes to fix micro-switch bases; (12) main nut pins groove; (13) fixation bracket; (14) micro-switch base; (15) honeycomb structure.

Figure 7

Static scheme for cable tension estimation. T is the estimated cable tension, T_req is the required cable tension, and F_f is the Bowden cable friction force.

Figure 8

A trial timeline representation of Graz-BCI paradigm on OpenVibe (OV). The participant is instructed to look at a black screen and wait for a green cross, which indicates the beginning of the trial. After 3 s, a red arrow cue appears randomly to the right or to the left. The left arrow indicates the task to relax (C1), and the right arrow indicates the task to perform right hand MI (C2), without any hand movement. After 1 s, a blue feedback bar appears to indicate MI performance in real-time during 4 s followed by 2–3 s to start the next trial. A BMI session has at least 20 trials for each class.

Figure 9

Average of the actuating force per actuator voltage supply. Error bars presented in the figure represent the standard deviations of each mean.

Figure 10

The Hand Exoskeleton for Rehabilitation Objectives (HERO) attached to the body. The actuators were attached to the right lower limb, on the same side of the exoskeleton glove, only for illustration. Patients should have the actuators attached to their healthy lower limb.

Figure 11

Bits sent to Arduino Uno according to OpenVibe (OV) algorithm outputs in trials with a right arrow [motor imagery (MI) of the right hand]. The green mark represents the moment in which the cross appears on the screen. The yellow mark represents the moment in which the cue to class 2 appears on the screen. The value “0” means the actuators are braked, and the value “1” means the actuators are activated. Upper graph: Trial recorded between 40 and 48 s showing that the feedback can work in a non-continuous mode, according to the decoding received from the MI. Bottom graphic: Trial recorded between 187 and 195 s.

Figure 12

Two-dimensional (2D) trajectories of finger flexion movements. The black curve represents the trajectory of the index fingertip during the execution of an active/natural movement. The red curve represents the passive movement.

Figure 13

Correlation analyses between x and y trajectories. First column: x and y trajectories vs. time for natural and passive movements. Second column: scatter plots with the 45° line (in red) to compare both movements—natural and passive—for x (top) and for y (bottom) trajectories.