RETINA

BioMathProject 2018: Extracting medically relevant features from the human retina

Background

Retina is the only part of the cardiovascular system which is accessible without any complicated intervention. A medical diagnostic may be possible simply by observing it. Indeed, the retina can reveal potential diseases such as atherosclerosis, hypertension or other pathologies concerning the eye. Assuming that a disease involves more vessels in the retina, a computer extraction of numerical values from images could give a quick overview of the patient's condition.

Goal

In this project, we aim to establish a correlation between retina's characteristic and pathology.

Methods

In order to extract characteristics from retinas, the use of a software that can manipulate images is required. This is why we recommend MATLAB software. The Kolmogorov-Smirnov test is used as a statistical tool to determine if the data are usable and reliable. The results are analysed to find a correlation between numerical characteristics and medical features.

Dataset

Anonymized data from SKIPOGH, a project listing patients related to hypertension.

Supervisor

Mattia Tomasoni

WIKI Wiki of the presentation: [1]