Page 394 - Handbook of Biomechatronics

388 Lilach Bareket et al.

optimize their performance. Psychophysics tests to evaluate visual function
of the recipients provide an objective assessment of visual function, and
should be provided throughout the rehabilitation phase and beyond. These
tests involves visuotopic mapping of the phosphenes and assessments of light
perception, direction and motion perception, object and shape recognition,
navigational tasks, and various activities of daily living (Dagnelie, 2008).
However, current approaches to assess visual function in sighted people
may not be suitable for blind individuals learning to use a bionic device
and may not provide accurate information to achieve successful rehabilita-
tion. It has been demonstrated with retinal stimulation that experimental
results may be different to this theoretical relationship between VA and elec-
trode size (Lorach et al., 2013). The highest VA demonstrated with the
Alpha IMS device was 20/546, lower than the expected acuity of
20/250, possibly a result of crosstalk between neighboring electrodes, indi-
rect stimulation of passing axons and/or local inhibition from concomitant
stimulation. Local return electrodes surrounding each stimulation element
could provide more focal stimulation (Lovell et al., 2005; Matteucci
et al., 2013; Joucla and Yvert, 2009). In fact, VA as a quantitative measure
may not necessarily corresponds to the spatial resolution produced by the
retinal prosthesis. The traditional tests for estimating VA may need to be
modified when artificial vision is evaluated, taking additional factors into
account such as visual field scanning movements with the head and the time
to complete the visual tests. Head scanning was demonstrated to improve
VA score in studies applying SPV (Cha et al., 1992; Chen et al., 2006),
and it seems to be a natural mechanism that the subjects in the experimental
SPV conditions employed (Caspi and Zivotofsky, 2015). In addition,
patients with low vision may need longer duration to complete the visual
task (Caspi and Zivotofsky, 2015). This additional time may need to be
added into the assessment of visual performance with prosthesis. SPV in nor-
mally sighted individuals is also used as means to test image processing algo-
rithms which transform the visual scene into pixelated forms that can be
elicited electrically into the visual pathways (Zapf et al., 2015; Bourkiza
et al., 2013; Lui et al., 2012; Chen et al., 2009). For example, Dagnelie
et al. (2006) developed a simulation of pixelated vision with cortical pros-
2
thesis and concluded that a 3 3mm prosthesis with 16 16 electrodes
should allow paragraph reading.
A further consideration is the need for consistent and uniform testing
regimes that groups doing visual psychophysics assessment can commonly
adopt. One positive step in this direction is the formation of an international