PosterPDF Available

Equine behaviour analysis in a bilateral lameness model as a precursor to automated detection

Authors:

Abstract

Preliminary research investigating the potential of exploratory and non-exploratory behaviours as a pain indicator in equines. Further work will determine the value of exploratory/non-exploratory behaviours in subtle pain and pain free states. This will inform the suitability of data set for 'training' using machine learning techniques.
Equine behaviour analysis in a bilateral lameness model as a
precursor to automated detection.
Egan, S.1, Brama, P.A.J.2, Kearney C.M.2, McKeown, D.J.3, McGrath, D.1
1: UCD School of Physiotherapy Public Health and Sports Science, 2: UCD School of Veterinary Medicine
3: UCD School of Mechanical & Materials Engineering, University College Dublin, Dublin, Ireland
Exploratory animal behaviour has been
associated with health status. Currently it is
unclear how this is expressed in equines.
Pawing has been linked to orthopaedic pain.
What equine behaviours are pertinent
during the acute 12 hours post bilateral
lameness induction?
Duration of non-exploratory behaviour
(DNEB) calculated from:
Discussion
Preliminary analysis focuses on the 0 12 hour period, 15 behaviours were classified as exploratory/non-
exploratory. Visual inspection indicated moderate to serve lameness, pawing was analysed independently.
5 out of 7 animals demonstrated a consistent trend of increasing DNEB in the 4-8hr period compared to the
0-4hrs, despite differing baseline magnitudes. Pawing indicated a similar pattern in 4 out of 7 horses.
The manual coding process is the initial step to training sensor data using a machine learning approach. The
ultimate objective is to develop a behaviour classification system enabling real-time relevant behaviour
monitoring.
The authors wish to thank the Irish Research Council, Government of Ireland Postgraduate Programme for supporting this research.
Methods
Further analysis is underway to determine the behaviour-relationship with food availability and the recovery
trajectory over 7 days.
This will determine if DNEB & pawing are a useful and sensitive indicator of bilateral lameness for future ‘training’
purposes.
Lipopolysaccharide
(0.25ng) @ 0 hours
Half Recumbency Total Recumbency
Quiet standing Agitated standing
Raised Forelimb (L) Raised Forelimb (R)
Introduction
CCTV monitoring
from 0 -12 hours
00:00:00
00:00:09
00:00:17
00:00:26
00:00:35
00:00:43
00:00:52
00:01:00
0-4hr 4-8hr 8-12hr
Duration (seconds)
Timepoint
Pawing
00:00:00
00:14:24
00:28:48
00:43:12
00:57:36
01:12:00
01:26:24
01:40:48
01:55:12
02:09:36
02:24:00
0-4hr 4-8hr 8-12hr
DNEB
Timepoint
DNEB
H1
H2
H3
H4
H5
H6
H7
Recommendations & Future Directions
Placebo treatment
@ 2hrs (2ml saline)
ResearchGate has not been able to resolve any citations for this publication.
ResearchGate has not been able to resolve any references for this publication.