Authors:
Rui Bernardes
1
;
2
;
Hugo Ferreira
1
;
Pedro Guimarães
1
and
Pedro Serranho
1
;
3
Affiliations:
1
Coimbra Institute for Biomedical Image and Translational Research, Faculty of Medicine, University of Coimbra, Edifício do ICNAS, Polo 3 Azinhaga de Santa Comba 3000-548 Coimbra, Portugal
;
2
University of Coimbra, CACC - Clinical Academic Center of Coimbra, Faculty of Medicine (FMUC), Coimbra, Portugal
;
3
Department of Science and Technology, Universidade Aberta, Rua da Escola Politécnica, 147, 1269-001 Lisbon, Portugal
Keyword(s):
Optical Coherence Tomography, Retina, Biomarkers, Texture, Convolutional Neural Network, Alzheimer’s Disease, Mouse Model, Diagnosis.
Abstract:
The World Health Organization (WHO) 2015 projections estimated 75.6 million people living with dementia in 2030, an update from the 66 million estimated in 2013. These figures account for all types of dementia, but Alzheimer’s disease stands out as the most common estimated type, representing 60% to 80% of the cases. An increasing number of research groups adopted the approach of using the retina as a window to the brain. Besides being the visible part of the central nervous system, the retina is readily available through non-invasive imaging techniques such as optical coherence tomography (OCT). Moreover, cumulative evidence indicates that neurodegenerative diseases can also affect the retina. In the work reported herein, we imaged the retina of wild-type and the triple-transgenic mouse model of Alzheimer’s disease, at the ages of one-, two-, three-, four-, eight-, twelve- and sixteen-months-old, by OCT and segmented gathered data using a developed convolutional neural network into
distinct layers. Group differences through texture analysis of computed fundus images for five layers of the retina, normative retinal thickness data throughout the observation period of the ageing mice, and findings related to the estimation of the ageing effect of the human genes present in the transgenic group, as well as the classification of individual fundus images through convolutional neural networks, will be presented and thoroughly discussed in the Special Session on ”New Developments in Imaging for Ocular and Neurodegenerative Disorders”.
(More)