The main goal of the MouseAge project is to develop the first visual marker of aging in mice in order to validate potential anti-aging interventions, save animal lives, and greatly speed up the pace of longevity research. The search for reliable biomarkers of aging, in humans and in model organisms, is an ongoing pursuit for scientists performing longevity research. Presently there are no rapid and reliable methods to continually assess rates of aging, morphologically, in mice.
In order to develop the visual biomarker system described in the project, the MouseAge team is employing an Artificial Intelligence (AI) programming technique known as Machine Learning. Machine Learning, whereby a computer system can train itself to become better at a task without explicit programming, has already showed great performance in areas such as human facial recognition, autonomous driving, medical image processing, recommendation engines and many others. While these results are powerful, building up the necessary Deep Learning Neural Networks, or algorithms inspired by the human brain, requires a large dataset of images to use for training: thousands of them.