Transcript Document

Identifiction of IBD using an
electronic ‘e’ nose
Arasaradnam RP1,2, Ouaret N3, Nwokolo C1, Bardhan KD4,
Covington JA3
1University Hospital Coventry & Warwickshire &
2Clinical Sciences Research Institute, Medical School,
University of Warwick
3School of Engineering, University of Warwick, 4Rotherham NHS Trust
10th BROAD Meeting – February 2012
1
The Senses
5 Senses by Aristotle
 Sight
 Hearing
 Smell
 Taste
 Touch
But also…
 Nociception (pain)
 Equilibrioception (balance)
 Proprioception & kinesthesia
(joint motion and acceleration)
 Sense of time
 Thermoception (temperature
differences)
 Magnetoception (direction)
‘Smell’ @ Warwick
 First research group dedicated to the sense of smell
 First company making artificial olfaction instruments
 First commercial products manufactured here…
 Long history of smell research…
Life in ‘Smell’
Persaud & Dodd Nature 1982
Sniffing diseases..…
Electronic Nose
Human olfactory system
Axel & Buck Nobel Prize 2004
How does it work?
 Array of sensors with different broad sensitivity e.g. Alcohols
 Operate by measuring change in
resistance/capacitance/frequency
ee- e-
e-
ee-
e-
e-
ee-
e-
e-
ee-
e- e
e-
If
a different
compound
causedby
thea air
to change,
the output
Each
sensor changes
its had
resistance
different
amount,
makingpattern
a pattern
Measurement
of baseline resistance
will
different:
of
thebechange
Electronic noses
Gas emissions and
Manufacturing processes
Food and cosmetics
Applications of
Electronic nose
Environmental
pollutants
Homeland security
Medical
applications
Sampling
 Disease alters gut flora -
altered fermentation patterns
which alters the composition
of gases emitted from urine
 Urine headspace measured
using an electronic nose and
FAIMS (field asymmetric ion
mobility spectrometry)
 Novel method – non invasive
 Possible faster and earlier
detection of IBD
Methods
• 48 patients and 14 volunteers; n = 62
• UC (n=20; active n=4), CD (n=20; active n=4) and
14 controls
• Urine analysis with ‘e’ nose and FAIMS
• Analysis was by Principal Component Analysis
(PCA) – not pre-classified and LDA (Linear
Discriminant Analysis) (pre-classified)
E - nose
Warwick Electronic Nose
Sample air is passed over sensors (300 secs)
followed by clean air (300 secs)
Filtered air is passed over
sample (1 L/min), capturing gas/vapours
emanating from the sample
10 ml of sample, placed in a sterilin bottle & heated to 40 oC
Separation of IBD with ‘E’ nose

Arasaradnam et al JMET 2011
Can we detect IBD by measuring Volatile organic
compounds/ gasses?
Arasaradnam et al 2010
FAIMS – Field Asymetric Ion Mobility
Spectrometry
 Simple fast analysis of vapours
and gases
 Detects chemicals in complex
mixtures
 Identifies by mobility (ion
movement through an electric
field)
 Mobility determined by
molecule size and mass
 Chemicals identified and
separated
 Can train to identify anomalies
 Creates a chemical fingerprint
FAIMS
Covington et al 2011
IBD flare v remission
What are we detecting?
Summary of chemical peaks in volunteers and Crohns and UC
patients
Total OA (TOA)
Volunteer (V)
Time
/Patient (Pt) (weeks)
146 ± 6
V
CD
UC
Peaks (n)
Ratio
HA:AA:PG
1 : 1 : 0.05
Total A, K
(TAK)
(%)
7.5 ± 1.5
26 ± 6
(%)
0
164
1 : 1: 0.2
12
10
2
151
1:1:2
20
12
6
145
1 : 1.4 : 0.3
15
8
0
146
1 : 0.8 : 0.05
9
15
3
156
1 : 1.7 : 0.1
20
22
HA=hydrogen azide (HN3); AA=acetic acid (CH3COOH); PG=propylene glycol (C3H6(OH)2);
A=aldehydes; K=ketones; OA=organic acids.
Summary
o Potential alternate diagnosis method
o Able to distinguish between control and IBD patient
groups
o Can also identify specific disease groups ulcerative
colitis and Crohn’s
o Greatest difference is shown between diseased and
non-diseased
o FAIMS can also be used to identify diseased groups
and supports e nose data
Thank you - Questions
 Acknowledgements:
 BROAD Foundation
the future.......?