Medical image analysis is the process of extracting meaningful information from medical images for clinical review and medical intervention. To date, there has been considerable effort in applying image processing techniques to assist clinicians with medical diagnosis and intervention. The rapid development of Deep Learning methodologies in medical image analysis has shown the potential of using these technologies to automate the analysis of medical images in the diagnosis of human diseases , reducing likelihood of misdiagnosis while saving on time, labour and cost for clinicians. Medical images capturing the reflection of the interior surface of the eye is called retinal fundus imaging. Such images can be used to provide facts and information about visual disorders such as diabetic retinopathy, glaucoma, age-related macular degeneration, etc.
A growing area of medical research is to evaluate the health of the eye in order to provide a greater insight into the overall health of the human body. While retinal fundus images are used in routine clinical practice to diagnose ophthalmic conditions, evidence suggests that the retinal fundus images contain valuable information regarding the systemic health features of patients . Examples of such diagnoses include brain disease (Alzheimer’s) , and heart-related conditions, such as cardiovascular, hypertension, and stroke .
This PhD proposal aims to develop appropriate deep learning models for use on retinal fundus images for diagnosing both systemic and localised human diseases. These deep learning models will allow for non-invasive, low-cost, and more accessible patient screening during an ophthalmologic examination, with the goal of accessible and earlier disease detection across various patient populations. Images collected by a trained technician, together with the use of an automated deep learning medical image analysis tool, will contribute to improving holistic patient care beyond ophthalmology by allowing patients to be diagnosed with various conditions in a non-invasive manner