Implantable microsystems and neuro electronic interfaces

The interdisciplinary field of implantable medical microsystems is gaining immense momentum due to the undoubted potential these devices hold in significantly improving the healthcare and also in the basic understanding of several human diseases. The most widely known success stories in this field are those of pacemakers and cochlear implants. However there is a clear need for such devices to mitigate growing number of chronic diseases. For example, it has been convincingly shown that tight regulation of blood glucose reduces the morbidity and mortality associated with diabetes but there are currently no reliable methods for continuous and long term monitoring of glucose concentrations. The only practical and most convenient way to continuously monitor the biomarkers in normal-living conditions is the use of implantable microsensors. Another such example is the development of neural interface devices to record and stimulate the neural signals and assist people with neurological disorders. The Microsystems group at the University of Utah is currently running two relevant programs in developing (a) hydrogel based chronically implantable multi-analyte sensor array and (b) a wireless neural interface device based on the Utah electrode array. A "smart" hydrogel is a three-dimensional polymer network that absorbs large quantities of water in response to some specific environmental trigger such as change of pH, temperature, or concentration of a specific molecule like glucose. In most hydrogel applications studied in the past, hydrogel swelling has been allowed to occur without confinement. However, in our sensing scheme, we confine a smart hydrogel between a semi-rigid porous membrane and the diaphragm of a miniature pressure transducer. In such a scheme, a change in the environmental analyte concentration, as sensed through the pores of the membrane, changes the hydrogel osmotic swelling pressure, thereby changing the mechanical pressure measured by the pressure transducer. The overall goal of this on-going program is to develop an implantable wireless glucose, pH and CO2 microsensor array. These sensors have been designed, fabricated and successfully tested in vitro continuously for 4 weeks and in vivo in wired configuration. The development of neural interfaces to record and stimulate neural signals, to assist individuals with neurological impairments, has been pursued for decades. Significant progress has been made in the past few years in the development of brain-machine interface (BMI). Currently there are four primary techniques available for recording neural signals: electroencephalography, electrocorticography, local field potentials, and single-neuron action potential recordings (single units). Among all these techniques that measure the extracellular potentials of neurons, single-neuron action potential recording using microelectrodes gives highest lateral and temporal isolations and, therefore, the most useful recording data. The use of microelectrodes is an invasive approach and carries with it an inherent risk of infection and surgical complications associated with transdermal interconnections. The reactive response of the tissue around the implanted microelectrode remains to be a major hindrance to chronic recording ability of microelectrodes. Engineering solutions to some of the other common problems of chronic recording are to use wireless transmission of power and signals to the neural interface device to mitigate the risk of infections, and to use biocompatible and hermetic materials to encapsulate the whole device except the recording or stimulating sites. Here we present (a) the integration techniques involved in building a new wireless neural interface device based on the Utah electrode array (UEA), and (b) chronic neural action potential data recorded with a prototype of the wireless transmission system in comparison to a benchmark hard-wired data acquisition system, including principle component analysis (PCA) of action potentials waveforms recorded on selected channels.