EVALUATION COMMITTEE

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

• Professor Stanisa Raspopovic, Professor in Biomedical Engineering, Center for Medical Physics, MedUni Wien (Vienna, Austria)
• Professor Thomas Stieglitz, Professor in Biomedical Microtechnology, Laboratory for Biomedical Microtechnology, Department of Microsystems Engineering-IMTEK, University of Freiburg (Freiburg, Germany)

Chairman: Associate Professor Sabata Gervasio, Associate Professor in Biomedical Engineering and Neuroscience, Laboratory of Neural Engineering and Neurophysiology, Department of Health Science and Technology (HST), Aalborg Universitet (Aalborg, Denmark)

Moderator: Professor Strahinja Dosen, Professor in Biomedical Engineering, Laboratory of Neurorehabilitation Systems, Department of Health Science and Technology (HST), Aalborg Universitet (Aalborg, Denmark)

ABSTRACT

A lower-limb absence is a profoundly impactful event that affects every aspect of daily life. To mitigate the impact of the amputation and allow these people to retrieve their independence, providing in-depth rehabilitation and prescribing more advanced microprocessor-controlled prostheses have been commonly recommended. However, modern-day legs do not provide sensory feedback and people with lower-limb amputation (LLA) keep facing functional, psychological, social, and cognitive challenges which might be a consequence of the absence of proper feedback capacities from the missing limb. Providing artificial sensory feedback has recently emerged as a promising strategy to enhance the functionality and user experience of lower-limb prostheses. However, the diversity of feedback interfaces and evaluation methods presents significant challenges in interpreting and comparing their benefits and limitations. A key issue is the lack of consensus regarding which feedback variables should be prioritized and how they should be encoded, making it difficult to draw generalizable conclusions across studies.

The purpose of this project was therefore to advance the understanding of artificial sensory feedback for lower-limb prosthesis users by developing intuitive, non-invasive feedback strategies and systematically evaluating their effectiveness through holistic and user-centered approaches. This included addressing how different types of feedback are perceived and integrated in tasks relevant to everyday life, and identifying which feedback characteristics align with user needs and preferences. More specifically, we attempt to address four specific objectives:

1. Identify the needs expressed by the people with lower-limb amputation that should be translated into development.

2. Assess the perception capacities of the residual limb using psychophysical evaluations of multichannel tactile stimulation (electrotactile and vibrotactile) during relevant daily life activities.

3. Investigate the impact of a novel feedback scheme developed (OmniFeel) on the functionality, task load, user experience when performing an out-of-the-lab (ecological) task in people with LLA.

4. Compare the impact of different feedback schemes provided from a microprocessor-controlled knee (C-Leg 4, Ottobock) on the functionality, cognitive load, user preferences and feedback needs in people with LLA.

Psychophysical evaluations highlighted the importance of calibrating sensory feedback systems under the specific conditions in which they are intended to be used. Factors such as physical activity (especially gait), and the socket-residual limb interface were found to significantly influence spatial and amplitude perception, with considerable variation between prosthesis users. Additionally, these evaluations showed that the developed vibrotactile feedback could be effectively implemented with vibromotors placed inside the prosthetic socket without further modification, shown by the satisfying levels of comfort, spatial and amplitude perceptions of the feedback. 

Building on these findings, we developed OmniFeel, a novel feedback system designed to convey multi-directional force information from the foot sole. The system proved to be intuitive and easily understandable for users without requiring extended training. A holistic evaluation out-of-the-lab revealed that the benefits of OmniFeel were not uniform across all participants. More specifically, it benefited mostly a participant with a transfemoral amputation, while it did not benefit extensively the participants with transtibial amputation, especially on biomechanical outcomes. Nevertheless, participants generally reported increased confidence, reduced cognitive load, and a preference for having additional sensory feedback to assist with daily activities.

Finally, we developed a feedback interface for a microprocessor-controlled knee (C-Leg 4, Ottobock), incorporating real-time data from embedded sensors such as gait phase, knee angle, and damping characteristics. Through user testing, it became evident that aligning feedback strategies with user priorities is essential. Notably, some variables commonly used in the literature (such as knee angle) were not consistently valued by participants, underscoring the need for more user-centered selection of feedback schemes.

In conclusion, this PhD contributes to the growing field of artificial sensory feedback in lower-limb prostheses, by advancing the understanding of the critical characteristics that novel feedback techniques should embody. The findings from our comprehensive evaluations show that intuitive and meaningful sensory feedback can enhance users’ confidence, cognitive load, perceived control, and interaction with the prosthesis, even when biomechanical improvements are not consistently observed. Importantly, the research reveals a potential mismatch between the technological focus of current feedback strategies and the actual benefits and needs expressed by users. This underscores the necessity of co-design approaches to ensure that future feedback-enabled prostheses are aligned with the lived experiences and preferences of people with lower-limb amputation. These insights carry practical implications for researchers, clinicians, and developers aiming to deliver user-centered innovations. Future research should continue to explore how artificial sensory feedback is integrated into daily activities, particularly across varying ecological settings and populations of people with LLA (level of experience, type of amputation).