Econ_June_09_2018

68 / 80

68 The Economist June 9th 2018
For daily analysis and debate on science and
technology, visit
1
FOUR years ago a woman in her early
30s was hit by a car in London. She
needed emergency surgery to reduce the
pressure on her brain. Her surgeon, Chris
Mansi, remembers the operation going
well. But she died, and Mr Mansi wanted
to know why. He discovered that the pro-
blemhad been a four-hour delay in getting
her from the accident and emergency unit
ofthe hospital where shewas first brought,
to the operating theatre in his own hospi-
tal. That, in turn, was the result of a delay in
identifying, from medical scans of her
head, that she had a large blood clot in her
brain and was in need of immediate treat-
ment. It is to try to avoid repetitions of this
sort of delay that Mr Mansi has helped set
up a firm called The firm’s purpose
is to use machine learning, a form of artifi-
cial intelligence (AI ), to tell those patients
who need urgent attention from those
who may safely wait, by analysing scans
of their brainsmade on admission.
That idea is one amongmyriad projects
now under way with the aim of using
machine learning to transform how doc-
tors deal with patients. Though diverse in
detail, these projects have a common aim.
This is to get the right patient to the right
doctor at the right time.
In s case that is now happening.
In February the firm received approval
from regulators in the United States to sell
type (malignantmelanoma), as successful-
ly as the professionals. That was impres-
sive. But now, as described last month in a
paper in the Annals ofOncology, there is an
AI skin-cancer-detection system that can
dobetter thanmost dermatologists. Holger
Haenssle of the University of Heidelberg,
in Germany, pitted an AI system against 58
dermatologists. The humans were able to
identify86.6%ofskin cancers. The comput-
er found 95%. It also misdiagnosed fewer
benignmoles asmalignancies.
There has been progress in the detec-
tion of breast cancer, too. Last month Khei-
ron Medical Technologies, a firm in Lon-
don, received news that a study it had
commissioned had concluded that its soft-
ware exceeded the officially required per-
formance standard for radiologists screen-
ing for the disease. The firm says it will
submit this study for publication when it
has received European approval to use the
AI —which it expects to happen soon.
This development looks important.
Breast screening has saved many lives, but
it leaves much to be desired. Overdiagno-
sis and overtreatment are common. Con-
versely, tumours are sometimes missed. In
many countries such problems have led to
scans being checked routinely by a second
radiologist, which improves accuracy but
adds to workloads. At a minimum Khei-
ron’s system looks useful for a second
opinion. As it improves, it may be able to
grade women according to their risks of
breast cancer and decide the best time for
their next mammogram.
Efforts to use AI to improve diagnosis
are under way in other parts of medicine,
too. In eye disease, DeepMind, a London-
based subsidiary of Alphabet, Google’s
parent company, has an AI that screens reti-
nal scans for conditions such as glaucoma,
its software for the detection, from brain
scans, of strokes caused by a blockage in a
large blood vessel. The technology is being
introduced into hospitals in America’s
“stroke belt”—the south-eastern part, in
which strokes are unusually common. Er-
langer Health System, in Tennessee, will
turn on its systemnext week.
The potential benefits are great. As Tom
Devlin, a stroke neurologist at Erlanger, ob-
serves, “We know we lose 2m brain cells
every minute the clot is there.” Yet the two
therapies that can transform outcomes—
clot-busting drugs and an operation called
a thrombectomy—are rarely used because,
by the time a stroke is diagnosed and a sur-
gical team assembled, too much of a pa-
tient’s brain has died. s technology
should improve outcomes by identifying
urgent cases, alerting on-call specialists
and sending them the scans directly.
The AIs have it
Another area ripe for AI’s assistance is on-
cology. In February 2017 Andre Esteva of
Stanford University and his colleagues
used a set ofalmost130,000 images to train
some artificial-intelligence software to
classify skin lesions. So trained, and tested
against the opinions of 21 qualified derma-
tologists, the software could identify both
the most common type of skin cancer (ker-
atinocyte carcinoma), and the deadliest
Medicine
From A&E to AI
Artificial intelligencewill improve the speed and precision ofmedical treatments
Science and technology
Also in this section
69 Sucking carbon dioxide from the air
70 The Cambrian explosion