Quantifying Elicited Sensation Intensities in Individuals with Lower Limb Loss via Pressure Matching Task

Introduction:: Individuals with lower limb loss (ILLs) face functional challenges including reduced balance, increased fall risk, and abnormal gait. Neural interface technologies, through direct activation of afferent fibers in peripheral nerves have successfully restored somatosensory feedback to distinct plantar foot regions in unilateral ILLs and improving mobility related functional outcomes^1,2,3 However, users of such systems typically self-report the perceived intensities of elicited sensations using an arbitrary analogue scale. To avoid sensory mismatch between feet during standing or walking, it is crucial to match these sensations with the plantar pressure intensities of the intact foot. In this study, we present a new methodology for quantitatively characterizing elicited sensation intensities in ILLs using a pressure matching task. This approach allows matching perceived sensations on a functionally relevant, individualized scale by having participants match sensations in their missing foot with the pressure experienced by their intact foot.


Materials and Methods:: Three ILLs underwent outpatient surgical procedures to receive high-density nerve cuff electrodes on the sciatic and tibial nerves of their amputated limbs. Two participants two had transtibial (LL1 and LL2) and one had transfemoral limb loss (LL3). Nerve cuff contacts that elicited sensation in the missing heel in response to electrical stimulation were selected for in depth study for each participant. During data collection, participants were seated with their intact limb attached to a dynamometer with an ankle pedal attachment (Biodex System 4). The residual limb and prosthesis were supported in a similar position but not attached to the dynamometer. Both legs were supported and secured at the mid-thigh to minimize upper leg muscle involvement in ankle movements. Each trial involved seven seconds of sensory electrical stimulation, with participants instructed to match the perceived sensation in their missing foot by pressing their intact foot on the ankle pedal three seconds after stimulation onset. Pulse width (PW) values tested were increased in steps of 20 μs and delivered randomly between the minimum and maximum values obtained during a thresholding procedure.¹ Three trials were conducted for each pulse width. After each trial, the participant was also asked to rate the intensity of the perceived sensation verbally on a scale of 1 to 10. Stimulation was controlled by a Performance Real-Time Machine (Speedgoat) to ensure synchronization of stimulation and output torque.


Results, Conclusions, and Discussions:: Preliminary data from one participant was collected with pulse amplitude at 1.3 mA while PW was varied between 80 and 240 us. Figure 1 depicts the self-reported sensation intensities over PW (left) as well as output torque produced by the intact foot (right). While reported intensities show an overall increase as PW is increased, the relationship between self-reported scale and PW is not always consistent. On the other hand, PW and torque had a clearer trendline. Increasing torque was observed until PW of 200 us, followed by a decrease in the torque. However, high standard deviations were also seen in this relationship, especially for a PW of 240 μs, suggesting that more data collection would be necessary. Overall, our results show an increase in torque with increasing stimulus PW indicating matching perceived intensity with the intact foot results in a more consistent characterization of sensation intensity compared to the analogue scale. These preliminary findings suggest an increase in both intensity and output torque across a range of PW with constant pulse amplitude. More data will also be taken with other ILLs to ensure that the trends seen above are replicable.

Acknowledgements (Optional): : This work was supported in part by the Department of Defense under Awards No. W81XWH-18-1-0321 and W81XWH-20-1-0802.


References (Optional): : 1. H. Charkhkar et al (2020), Sci Rep, vol 10, pp. 1-13. 2. H. Charkhkar, et al, J. Neural Eng, vol 15, July 2018, pp. 1–14. 3. Petrini, F.M., Bumbasirevic, M., Valle, G. et al (2019). Nat Med 25, 1356–1363.