The general specifications of these systems are summarized in Table 2; these have been included either because they were, as stated in each report, specifically designed for use in a high electrode count cortical visual prosthesis, retinal

prosthesis, or both. A key requirement for inclusion was that the performance of the inductive link must have been evaluated over a distance of ≥10 mm, which we consider as the likely minimum distance between the external coil and the cortical electrode arrays. Energy dissipated as heat by the implant remains a problem of concern. A high electrode count, continuous stimulation and the possibility of increasing current requirements if the electrode/tissue interface becomes impaired over time, all contribute to the potential for high power requirements and therefore greater temperature increases. Studies of focal and whole-brain heating over short periods (30 min), have shown that temperature rises up to 43 °C TSA HDAC mw can be tolerated without damage (Coffey et al., 2014 and Haveman et al., 2005). In the context of a cortical visual prosthesis,

stimulation is likely to be continuous over a period of hours, therefore the implant must remain at low temperatures to prevent tissue damage. There is little data on the damage to neural tissue resulting from see more chronic, focal cortical hyperthermia, although some information is available from the literature on heating of tissue due to ultrasound exposure. O’Brien et al. (2008) reviewed the literature on thermal effects of ultrasound, including several studies on cat and rabbit brain. From these studies, a conservative temperature–time exposure boundary was produced, which suggests that increases in temperature of 2 °C above 37 can be safely tolerated for lengthy periods (up to 50 h). An important consideration in this context is the normal human brain temperature, which was found in a recent magnetic resonance spectroscopy study to vary regionally between 34.9 °C and 37.1 °C, and to not differ greatly from core body

temperature (Childs et al., stiripentol 2007). The authors commented on some methodological contributions to the measured variation, however it is also known that a temperature gradient exists between cortical and subcortical regions, with cortical temperature typically being lower by up to 1 °C (Mellergard, 1995). Considering the previously-mentioned 39 °C limit (O’Brien et al., 2008), it would seem that the window of thermal safety may vary from one individual to another. Moreover, the stimulation itself may contribute to temperature changes via alterations in oxidative metabolism and cerebral blood flow (Yablonskiy et al., 2000), while increased cerebral blood flow will itself result in greater heat dissipation (Kim et al., 2007); therefore the accurate estimation of the likely temperature increase due to dense, patterned visual cortex stimulation is a complex task. Kim et al.