EVALUATION COMMITTEE

The Faculty Council has appointed the following adjudication committee to evaluate the thesis and the associated lecture: 

• Professor Lorenzo Masia, Ph.D, Chair in BioRobotics & Medical Technology, Institute for Computer Engineering of the University of Heidelberg, Germany
• Associate Professor Sadasivan Puthusserypady Kumaran, Department of Health Technology, Technical University of Denmark 

Chairman:
Associate Professor Shellie Boudreau, Department of Health Science and Technology, Aalborg University

Moderator:
Associate Professor Romulus Lontis, Department of Health Science and Technology, Aalborg University

HOW TO PARTICIPATE

The Ph.D. Defense is organized as a hybrid event you can participate digitally via Zoom or physical presence. 

Please click here to participate via Zoom.

Meeting ID: 688 6464 8789

ABSTRACT

Spinal cord injury (SCI) at the cervical level can lead to paralysis of both arms and legs (tetraplegia) and leave the affected individual severely disabled. Individuals with tetraplegia usually cannot perform activities of daily living (ADLs) independently and need full-time assistance. The lack of independence and privacy decreases the quality of life, causing an increased rate of depression and suicidal thoughts. Upper-limb exoskeletons (ULEs) that mobilize the arm and hand can improve the functionality and help individuals with arm disabilities to perform some ADLs independently and reduce the needed time for assistance. The more severe disability, the higher is the gain from assistive ULEs. However, providing a control interface enabling full and continuous control of a multi-DOF ULE for individuals with complete functional tetraplegia remains a challenge and requires further research and development. Thus, this PhD study aimed to develop a tongue robot interface that provides single modal, full voluntary continuous control of a 5 DOF ULE to empower users with complete functional tetraplegia. More specifically, we developed a new method for designing virtual buttons and a continuous joystick emulation based on the inductive tongue interface (ITCI) and tested them in a study with twelve able-bodied participants. Furthermore, we developed a novel tongue-gesture recognition algorithm. We used these findings to create and compare a gesture-based and a joystick-based control layout for a five-DOF ULE in a study with ten able-bodied individuals. In addition, we evaluated eyes-free use of the control layouts (without visual feedback). Finally, we evaluated tongue control of the five-DOF ULE for empowering individuals with tetraplegia to fully and independently perform highly desired ADLs, including drinking and eating snacks, in a clinical study with ten individuals with tetraplegia. The studies with able-bodied participants showed that a continuous joystick emulation based on the ITCI improved the performance of assistive robot interfaces as compared to virtual buttons by reducing the task time by up to 16% and the number of issued commands by 20%. The tongue gesture recognition algorithm identified a set of six gestures with 94.3% accuracy and 23 gestures with 72.3% accuracy, allowing for a high number of control commands with the ITCI. No significant difference was obtained between the gesture-based and the joystick-based control layouts for controlling the five-DOF ULE. The developed tongue-exoskeleton interface successfully empowered users, even with complete functional tetraplegia, to perform ADLs with the ULE. All ten individuals with tetraplegia who participated in our clinical study successfully controlled the ULE with the ITCI and independently completed the drinking and snacking tasks. The participants rated the intuitiveness of the tongue interface 5.2 on a scale between 1 (low) and 7 (high).