8 | NewScientist | 8 September 2018NEWS & TECHNOLOGY
SUBTLE clues in how you type on a
keyboard may be able to reveal early
signs of Parkinson’s. The hope is that
this could help spot the disease before
pronounced hand tremors or serious
changes in the brain have occurred.
To test the approach, hundreds
of volunteers installed a program
on their computers that monitored
their typing over nine months.
Warwick Adams at Charles Sturt
University in Australia then whittled
the sample down to 76 individuals
who were of the appropriate age, not
taking medication, and who either
didn’t have Parkinson’s or whose
condition was of mild severity.
Adams wanted to see if the times
between key presses could be
accurately plotted against a sine
wave of 4-6 hertz – the frequency of
Parkinson’s hand tremors. If these
data points didn’t map well to such
a curve, that would be an indication
that no tremor was present.
Using this technique, the system
was able to correctly identify patients
who had mild Parkinson’s disease
tremor with 78 per cent accuracy
(bioArXiv, doi.org/ctjk).
“The endgame is to develop a
widely available screening test
for both GPs and individuals,” says
Adams, who has Parkinson’s himself,
but not tremors.
Early detection would in
theory allow doctors to prescribe
treatments that can inhibit the
progression of the disease. About
three-quarters of people with
Parkinson’s develop hand tremors.
Adams has previously developed a
method of detecting early Parkinson’s
disease by spotting changes in the
flow of someone’s typing. He thinks
combining the two methods will help
build a diagnostic tool.
However, one difficulty the system
will face is that mild tremors are most
pronounced when resting, not typing,
says Álvaro Sánchez Ferro at the HM
CINAC Comprehensive Neurosciences
Center in Spain. Chris Baraniuk ■Yo u r t y p in g
could reveal
Parkinson’s
Leah CraneTHERE is a weird cold spot on
Europa, and nobody knows what
it is. The first full thermal map
of Jupiter’s icy moon has thrown
up a spot about 300 kilometres
across that seems colder than
its surroundings, and we don’t
have enough data on the area to
figure out why.
Samantha Trumbo at the
California Institute of Technology
and her colleagues used the
Atacama Large Millimeter Array,
a set of more than 60 radio
telescopes in Chile, to measure
the heat radiating from Europa’s
surface in a set of four overlapping
thermal images. It represents the
first full temperature map of the
icy moon.
“We weren’t so interested
in cold spots, we were more
interested in hot spots, because
those could indicate geological
activity,” says Trumbo. “But we
saw this one cold spot repeated
in two of the images.” The spot
is only about 9°C colder than
the surrounding areas, but itis different enough to be
surprising, she says (arxiv.org/
abs/1808.07111).
There are two main potential
explanations for the cold spot:
either it is emitting less heat than
the rest of the surface, or it takes
longer to warm up in Europa’s
morning.
The first could happen if the ice
there has a different composition
to that in the surrounding area,
making it harder for heat to travel
through it. The second could be
a result of a blockier, less grainy
texture. A large block of ice wouldtake longer to warm up, much as
an ice sculpture can last an entire
afternoon but a bowl of ice cubes
will melt within a few minutes.
“If this spot is real, I would
expect it to be a large ice sheet,
maybe younger than the rest
of the surface because it hasn’t
been broken up yet,” saysJessica Noviello at Arizona State
University. “On Europa, I would
be interested in studying
something young to see what
this moon really looked like in its
earlier state and how it changed
over time.”
The model that Trumbo’s team
used to spot this anomaly was
partially based on limited data
from the Galileo spacecraft on the
area’s albedo, or reflectiveness.
Because of that, Noviello and Julie
Rathbun of the Planetary Science
Institute in Arizona doubt that
the cold spot is actually there.
“I looked back over the Galileo
data, and there is no good data
on that area, so we have no idea
what it looks like,” says Rathbun.
Despite the lack of data,
Trumbo insists that there’s
something odd going on. “It is
possible that our albedoes are
not 100 per cent correct – actually
it’s likely,” she says. “But it
would be strange if they were
wrong significantly in only one
location without that spot being
somehow different.”
Figuring out exactly how it is
different will have to wait for
more data. “For now, you can
imagine almost anything,”
Trumbo says. ■Mystery cold spot
found on Europa
NASA/JPL-CALTECH/SETI INSTITUTE
“ The cold spot might emit
less heat than the rest of
the surface or take longer
to warm up in the morning”Part of Jupiter’s icy moon Europa is
even chillier than the rest