Page 176 - Computational Retinal Image Analysis

CHAPTER

Statistical analysis and

design in ophthalmology: 10

Toward optimizing

your data

Gabriela Czanner a,b,c , Catey Bunce d,e,f,g
a
Department of Applied Mathematics, Liverpool John Moores University, Liverpool, United Kingdom
b
Clinical Eye Research Centre, Royal Liverpool University Hospital, Liverpool, United Kingdom
c
Department of Eye and Vision Science, University of Liverpool, Liverpool, United Kingdom
d
Faculty of Life Sciences and Medicine, King’s College London, London, United Kingdom
e
Moorfields Eye Hospital NHS Foundation Trust, London, United Kingdom
f
UCL Institute of Ophthalmology, London, United Kingdom
g
London School of Hygiene and Tropical Medicine, London, United Kingdom
1 Introduction
1.1 Data analysis in ophthalmic and vision research

Statistics is the science of learning from data, and of measuring, controlling and
communicating uncertainty.
(Marie Davidian and Thomas A. Louis)
Research in ophthalmology and vision seeks to discover the mechanism by
which pathology manifests, to find ways for efficient diagnosis and monitoring
of diseases, to compare treatments for pathology and make recommendations on
preventative measures. Typical questions might be: What pathological lesions in
the retina are predictive of losing sight in 2 years from now? Why do some eyes
respond well to a therapy and others not? Can we use the functional properties of
the eye such as electroretinography to diagnose stages of diabetic retinopathy?
Answering such questions is the grand ambition of this broad enterprise. While
much is already known forming the basis for our current knowledge, the meth-
ods that produce this knowledge, the scientific method, involves both observa-
tions and running clinical studies producing the data, are evolving, offering real
potential to extend our knowledge. We rely upon data to answer questions but as
new techniques are discovered these data are becoming increasingly complex—
while previous research was conducting using perhaps simple measurements such

Computational Retinal Image Analysis. https://doi.org/10.1016/B978-0-08-102816-2.00010-1 171
© 2019 Elsevier Ltd. All rights reserved.