HIV-Associated Neurocognitive Disorders (HAND) are common and can present in older patients with marked similarities to that of age-related neurodegenerative conditions, such as Alzheimer’s disease. The  goal is to identify image phenotypes relevant to HAND to better understand the longer-term impact of HIV and improve diagnosis, monitoring and treatment of aging HIV-infected subjects. Unlike existing brain imaging studies in HIV, which group subjects according to presumed clinical markers, the team will reflect the anatomical heterogeneity of this population by automatically creating maps of images and their proximity relationships. They will then use these maps to separate the population into cohorts with distinct morphometric patterns and analyse the diagnostic power of these patterns.