Neurodegenerative diseases cause tremendous damage to the patients, their families and society. The prevalence of these diseases is expected to sharply increase in the future unless successful treatments are found. There are many drugs currently in development, and better understanding of the disease and the effect of treatment will be necessary to properly assess them. This includes understanding the order, timing, and magnitude of these processes, and how various genetic factors can modulate the disease. Biomarkers derived from imaging data will play a key role in these functions. These measurements will need to be highly precise and accurate in order to capture very small signals years before clinical onset or to capture important links to key genotypes. They must also be robust enough to handle large cohorts of data acquired on a heterogeneous collection of imaging equipment. Developing biomarkers that have these qualities not only requires cutting-edge methods development but also practical and secure approaches to data management as well as rigorous validation in order to satisfy regulatory requirements. In this talk, I will present some of the novel methodology developed in our laboratory to create the next generation of imaging biomarkers for multiple sclerosis and dementias. This includes measurements of atrophy using structural imaging in both the brain and spinal cord, as well as more advanced imaging modalities such as diffusion weighted imaging, amyloid and tau imaging using combined PET-MR scanners, and sodium imaging as a potential marker of neuroinflammation. The informatics platform used to acquire and store imaging data and derived results will also be presented, including incorporation of automated pipelines to provide a seamless solution for image analysis. Combining new biomarkers with robust informatics infrastructure will provide a functional solution for data management and analysis in pivotal clinical research of large-scale natural history studies, phase II and phase III trials. It will also pave the way to the ultimate goal of further evolving these methods from clinical research tools to clinical support of patient specific diagnosis.