Stent-based endovascular brain interface: characterization of recording and stimulating properties
Document TypePhD thesis
Access StatusThis item is embargoed and will be available on 2021-08-21. This item is currently available to University of Melbourne staff and students only, login required.
© 2019 Giulia Gerboni
Brain interfaces have the potential to treat traumatic and degenerative neurological conditions, which are major causes of morbidity and disability in modern society. However, despite outstanding progress and numerous promising research outcomes, very few brain interfaces reach a compromise between safety and performance needed to provide a chronic benefit for the patient. The Stentrode is the first ever stent-mounted, multielectrode and chronic brain interface that provides intracranial access to brain activity through the vasculature, with no need for burr-holes or brain penetration. It holds promise for increased safety during and after deployment, while remaining in proximity to neurons. This thesis aimed to investigate the performance of this novel device, the Stentrode, and provide empirical evidence on its ability to reliably access and deliver information to the brain. Firstly, we analysed chronic, endovascular signal quality. Full-field flashes were delivered in awake sheep and recorded with Stentrodes implanted in a dural vessel (superior sagittal sinus). Signal amplitude measures and impedance spectra were collected in the first 30-days post-stent deployment daily in the first two weeks and at alternate days thereafter. Results indicated stabilisation in the third week; however, large, repeatable and cortically widespread visual evoked responses remained significantly larger than background activity throughout the duration of the experiment, suggestions an overall minimal impact on immune response on recording quality. Next, we focussed on feasibility and efficacy of endovascular brain stimulation. The delivery of electrical pulses from within the vasculature has entered clinical practice in the peripheral nervous system however, surprisingly, systematic experimental results on intracranial endovascular stimulation are lacking. We delivered electrical stimuli using chronically implanted Stentrodes, while monitoring brain activation using a large subdural grid in anesthetised sheep. We observed a monotonic growth function for the response amplitude versus stimulation current and identified the current thresholds to elicit brain activation using different stimulation pulse durations that described a typical strength-duration curve. We reported a limited spatial spread of electrically-evoked brain activity that was more anterior and localised compared to centre of activation elicited with flash-VEPs. Finally, we analysed endovascular brain recordings acquired in response to different types of visual stimuli in relation to endovascular electrode placement. Using full-field flashes, we accessed a large amplitude peaks presenting similar latencies with all electrodes, with no dependence on their linear or radial placement. Using localised visual stimuli, linear distance of electrodes from the occipital lobe was inversely proportional to the amplitude of the response. When stimuli were delivered across the visual field, we observed an hemifield-preference with higher amplitudes and shorted latencies of the peaks. However, electrode orientation did not influence the amplitude of the peaks. Overall, these results demonstrate chronic reliable performance of the Stentrode in recording from the brain and efficacy of an endovascular approach to cortical stimulation. In addition, we discuss experimental setups to investigate recording quality based on visual stimuli and present preliminary results on spatial resolution of the Stentrode in relation to localised visual stimuli. Results and methods proposed show promise for improving the applicability of intracranial brain interfaces.
Keywordsstentrode; brain-machine-interfaces; VEPs; visual stimulation; brain stimulation; neuromodulation
- Click on "Export Reference in RIS Format" and choose "open with... Endnote".
- Click on "Export Reference in RIS Format". Login to Refworks, go to References => Import References