Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
Neuroprosthetics are a class of medical devices that aim to restore lost or impaired functions of the nervous system by electrical stimulation or recording of neural tissue. State of the art neural implants suffer today from a mechanical mismatch compared with the soft and curved host tissue, as they constrain mechanically the physiological motion dynamics of the central nervous system. This mismatch causes poor electrode-tissue contact, leading to unspecific stimulation or recording, as well as chronic scarring. This fundamental limitation of conventional systems can be overcome by developing soft neural interfaces, using more compliant materials, which can achieve chronic bio-integration and conform to the static and dynamic mechanics of neural tissue. The development of such soft neural interfaces requires the use of electrical conductors that can elastically deform while maintaining their electrical conductivity.
The main objective of this thesis it to develop a new strategy to engineer elasticity in otherwise rigid materials by structuring them with specific patterns. This happens spontaneously at the microscale on stretchable gold films on silicone that display dense distributions of Y-shaped cracks to favor out-of-plane deformation. This work draws inspiration from these cracks by patterning Y-shaped cuts to engineer reversible elasticity in a multi-layer of metallic and plastic thin films. The geometry of these Y-shaped patterns was first optimized using finite element analysis and macroscopic models. Then, a fabrication process was developed enabling the micro-patterning of polyimide/platinum/polyimide interconnects with microscaled Y-shaped cuts (branches of dimensions âŒ15ÎŒm), which were then encapsulated in silicone. These encapsulated micro-patterned interconnects exhibit a sheet resistance of âŒ15Ω/sq., and remained conductive when elongated by up to 70%, with a resistance increase less than 2.5 times. They were also shown to reversibly stretch at 10% tensile strain for 1 million cycles. This technology allowed patterning of tracks down to 20 ÎŒm in width on a wafer-level scale.
These patterned elastic films were then integrated as interconnects in neural implants, coupled with a previously described stretchable coating on the electrode sites. The main application addressed by this thesis is the auditory brainstem implant (ABI), a neuroprosthesis to stimulate the cochlear nucleus (CN) in the midbrain. Existing ABIs produce highly variable speech perception result and it is hypothesized that current ABIs are too stiff to conform to the curvature of the CN, which might lead to poor electrode contact with the neural structures. A soft human scale ABI was tested in a human cadaveric model to demonstrate that this novel electrode is robust to surgical manipulation and can easily be inserted in the brainstem using a dissolvable hydrosoluble guide. In addition, a scaled-down ABI array was tested in vivo in a mouse model and it was shown that it could reliably recruit the auditory system and remain durable for up to 4 weeks. In addition to ABIs, this technology was also validated for neural recordings at the surface of the minipig cortex, in an acute setting, as well as for chronic spinal cord stimulation in non-human primates.
In the future, this technology could be used to develop personalized soft neural implants that could potentially provide a better chronic biointegration and a more robust interface with neural tissues.