The purpose of this study is to use statistical and classification models to classify, detect and un

1
Program 2131
Bayesian networks to classify visual field
data A Tucker,1 V Vinciotti,1 XH Liu,1 DF
Garway-Heath2 1Dept Info Systems and Computing,
Brunel University, UK 2Moorfields Eye Hospital,
London, UK




RESULTS
PURPOSE

• 2. A Time Series of VF tests from 24 subjects
  converting from ocular hypertension to glaucoma
• VFs of 255 subjects attending the Ocular
  Hypertension Clinic at Moorfields Eye Hospital
  were examined at 4-monthly intervals (Kamal 2003)
• A subset of 24 patients were taken who had
  converted to glaucoma in their right eye (mean
  age 63.0 SD 12.4)
• All subjects initially had normal VFs and
  developed reproducible glaucomatous VF damage in
  a reliable VF during the course of follow-up
  (conversion)
• Conversion was defined as the development of an
  AGIS score ?1 from initial score of 0, on three
  consecutive reproducible and reliable Humphrey
  24-2 full threshold strategy VFs, with at least
  one location consistently below the threshold for
  normality
• The average number of fields in each patient's
  series was 24, the SD was 8, the maximum was 45,
  and the minimum was 1

ROC curve analysis

• The purpose of this study is to use statistical
  and classification models to classify, detect and
  understand progression in visual fields (VFs)
• We intend to make use of the vast amount of data
  available to build models which avoid inherent
  problems of black box paradigms
• Integration of different types of clinical data
  for the diagnosis and detection of glaucoma
  progression would be helpful to clinicians
• Bayesian network models are ideal for classifying
  VF data whilst facilitating the understanding of
  VF progression
• They are learnable from data
• They model knowledge explicitly (graphical
  structure and probabilities)
• They can incorporate different types of data
  (e.g. clinical and VF)
• A comprehensive comparison of machine-learning
  classifier systems was carried out for glaucoma
  (Goldbaum et al, 2002). Many of these classifiers
  are black box in nature.
• The distribution of point-by-point light
  sensitivity has been explored in normal (Hejl,
  1987) and glaucomatous populations (Weber, 1992).
  Much remains unknown about the behaviour of the
  VF test.
• We know of little research using Bayesian
  networks to understand and classify VF data
  (Tucker, 2003)

• ROC in Figure 3 (left) reveals BNC performs best
  out of the Bayesian methods
• Figure 3 (right) shows that BNC comparable to
  both LR and KNN

Figure 3. The ROC curves generated for the
different classifiers learnt from single VF data
Network Analysis

• Various relationships
• found that relate to
• known anatomical
• information
• Nasal step and arcuate
• paracentral defects
• are influential in
• classification (Figure 4)
• Mean optic nerve
• head angular distance
• (Garway-Heath, 2001) of
• parent and child of a link
• was 15.3 degrees (Figure 5)
• Temporal VF found to be useful for
• classification although conventionally not
  thought important (Figure 4)

Table 1. A Breakdown of the datasets
Figure 4. The direct descendants of the glaucoma
class node
Figure 5. The resultant network structure learnt
from single VF data
Glaucoma Visual Field Classification

Glaucomatous field loss
(1)
Bayesian classifiers 1. Naïve Bayes classifier
(NBC) Assumes feature independence 2.
Tree-augmented Bayes classifier (TAN) Relaxes
independence assumption Tree Structure Learnt
Between Features 3. Bayesian network classifier
(BNC) Bayesian network including class node

• Testing on time-series data
• Classification accuracy was only 66
• However, there are several reasons for such a low
  score. Four characteristics appeared (Figure 6)
• Slow build up in probability of glaucoma prior to
  clinicians decision
• Fluctuations prior to clinicians decision
• Classified as glaucomatous throughout
• Fluctuations throughout

(2)
(3)
Figure 1. Two Sample Visual Fields from a Healthy
Eye (left) and a Glaucoma Sufferer (right)
Figure 2. The Architectures of the different
Bayesian classifiers (1) naïve Bayes (2)
tree-augmented network (3) Bayesian network
Figure 6. Four sample results of testing the
Bayesian network models on the time series
dataset comparing to clinicians conversion
decisions (denoted by a dotted line)
Statistical classifiers 4. Linear regression
(LR) Attempts to classify using straight line
fit to data 5. K nearest neighbour
(KNN) Classifies based upon majority of k
nearest neighbours (calculated using distance
metric)
METHODS

• Datasets
• 1. Single VF tests from 180 subjects
• Subjects included 78 with established early
  glaucomatous VF loss (mean age 57.5 years SD
  12.4) and 102 normal volunteers known not to be
  sufferers (mean age 65.1 years SD 10.1)
• One VF per subject was used for analysis
• Early glaucomatous VF loss was defined as an AGIS
  (Advanced Glaucoma Intervention Study 1994) score
  between 1 and 5, on three consecutive
  reproducible and reliable Humphrey 24-2 full
  threshold strategy VFs, with at least one
  location consistently below the threshold for
  normality in subjects with pre-treatment IOPs gt
  21mmHg
• Normal subjects had VF tests scoring 0 in the
  AGIS classification , IOP lt 21mmHg and no family
  history of glaucoma

DISCUSSION

• A number of Bayesian classifiers have been
  investigated for identifying VF deterioration
  associated with glaucoma, while relationships
  between variables have been explicitly modelled
• The resulting classifiers can be used to help
  understand VF deterioration through network
  structure analysis and comparison with
  clinicians decisions
• Various characteristics typical of early VF
  damage, such as the nasal step, are identified
  within the Bayesian network structures, although
  the finding should be interpreted with some
  caution because the models used in this study may
  be learning clinical classification processes (in
  this case, AGIS)
• The relationship between clinicians decisions
  and the models decisions has shown the potential
  to understand how clinicians come to their
  decisions and possibly use the information to
  improve upon the current VF classification methods

References Goldbaum MH, Sample PA, Chan K et
al.Comparing machine learning classifiers for
diagnosing glaucoma from standard automated
perimetry. Invest Ophthalmol Vis Sci. 2002 43
1 162 169. Heijl A, Lindgren G, Olsson J.
Normal variability of static perimetric
threshold values across the central visual
field. Arch Ophthalmol 1987 105 1544
1549. Weber J, Rau S. The properties of
perimetric thresholds in normal and glaucomatous
eyes. Ger J Ophthalmology 1992 1 79
85. Tucker A, Garway-Heath DF, Liu X. Spatial
operators for evolving dynamic probabilistic
networks from spatio-temporal data. Proceedings
of the Genetic and Evolutionary Computation
Conference. Chicago Springer-Verlag, 2003 pp.
1217. Advanced Glaucoma Intervention Study. 2.
Visual field test scoring and reliability.Ophthalm
ology 1994 101 1445 1455. Kamal D,
Garway-Heath DF, RubenS et al. Results of the
betaxolol versus placebo treatment trial in
ocular hypertension. Graefes Arch Clin Exp
Ophthalmol 2003 241196203. Garway-Heath DF,
Fitzke F, Hitchings RA. Mapping the visual field
to the optic disc. Ophthalmology 2000 107
18091815.