Retinal images are a goldmine of data. ILLUSTRATUM uses the latest advances in data science, deep learning and digital image processing to analyse images of patient’s retinas stained with the DARC biomarker. The algorithm locates the sick, stressed and apoptosing cells and analyses the number and spacial distribution. This combined with patient data such as age and prior conditions allows the algorithm to:
Accurately diagnose the specific disease (aiming for both specificity and sensitivity to be above 90%).
Predict how the disease will progress in the patient over the coming years.
Help to recommend suitable interventions.
ILLUSTRATUM will be collecting the largest dataset of retinal images in the world. We are building highly scalable infrastructure to support the demands of big data and our growing customer base, with high availability, elastic scaling architecture with distributed processing and convolutional neural networks. We will be collecting relevant medical data as well as the retinal images from the patients to build a vast knowledge base that will sit at the very forefront of neurodegenerative disease research.