Inside News

38 / 64

Clinical Radiology
38 Inside News
Clinical Correlation
Required – A
Radiological Truism
A claim so obvious or self-
evident as to be hardly
worth mentioning
We’ve all read them, perhaps we’ve written
them. Three pages long with sentences
morphing into paragraphs. Finely worded
and artful but obfuscation and hedging
nonetheless. Enhancement in the deltoid;
query infection. The sub-centimetre
nodes in the retroperitoneum; TB is
suggested. The emergency registrar
couldn’t rule out a pulmonary embolus,
well neither could you.
A group at Stanford University has
developed a computer-based review
system utilising natural language
processing, which they used to examine
almost one million radiology reports
over a period of five years. Validated
with a sensitivity and specificity of over
97 per cent and 99 per cent respectively,
the analysis identified uncertain or
hedging statements in 30.9 per cent
of reports. Perhaps unsurprisingly,
trainees were even more likely to make
ambiguous comments.
From hazy mesenteries, to architecturally
distorted apicies and fuzzy posterior
fossae, the aspiring radiologist
regularly encounters findings that resist
characterisation. Yet recognising the
normal and sorting out the relevant
from the inconsequential is at the heart
of imaging interpretation and a key
skill of radiologists. Formulating and
communicating a coherent diagnosis to
help chart a path forward is tougher still.
The smudge of soft tissue in the neck
or the blotch of hypoattenuation in the
brain can be a bewildering trap. Too
often the result is no judgement and
no value-add.
Importantly for patient care, a group
from Beaumont Health in Michigan,
who presented at RSNA in 2015,
showed that follow-up rates for patients
decrease significantly when ‘conditional’
statements (read hedges) are used by
radiologists. By withholding judgement
we may avoid risk and retribution but we
also do our patients a disservice. To do
so regularly cedes the titles of specialist
and doctor.
So to the trainee radiologist; water
infrequently and prune often. Hedges
obstruct the view and should be planted
sparingly.
Clinical correlation required.
Dr Jules Catt
Chair, Radiology Trainees Committee
Emergency
Radiology Special
Interest Group
I am pleased to inform you that the
Executive Committee of the Australian
and New Zealand Emergency Radiology
Group (ANZERG) had its first telephone
conference on 26 February 2016.
We have reached out to various
international emergency radiology
societies as well as the Australasian
College for Emergency Medicine, the
Royal Australasian College of Surgeons,
the Australasian Trauma Society and
other Australian/NZ colleges informing
them of the establishment of our special
interest group. We had a very positive
response from the British Society of
Emergency Radiology who are keen to
collaborate with us.
The committee members will work
towards developing the ANZERG web
page which is on the College website
interest-groups/anzerg. We are also
putting together a registry of ANZERG
members. If you have not yet expressed
your interest, you can do so by writing
to Sarah Hall, Administrator, Member
Support, at the College office on
. Any
suggestions from the wider
membership are welcomed.
Currently there are big variations in the
way first year key conditions are taught.
Drs Perry Cleland and Craig Hacking are
putting together a survey that we will be
sending out to Directors of Training and
Branch Education Officers. The intention
is to gather information on how the
key conditions are taught at various
departments and what assessment
processes are in place to ensure that first
year registrars are ready to go on after-
hours rosters.
Dr Craig Hacking is putting together a
program for the Training Learning Day
at the Annual Scientific Meeting on the
Gold Coast. The ANZERG Committee
is planning other educational sessions
on emergency and trauma radiology
which will be advertised to the broader
group of trainees and Fellows when
finalised. We recognise that almost
every radiologist is faced with acutely
unwell patients within or outside of
the emergency department cohort
of patients and such sessions will be
valuable to our Fellows and trainees.
ANZERG is also proud to contribute to
the Australasian Trauma Society’s regular
member newsletter. In the first article,
Dr Craig Hacking presented the CT
findings of pancreatic trauma. We look
forward to providing regular imaging
educational content with ATS.
A/Prof Dinesh Varma
Chair, ANZERG