Combining SIENA and SIENAx for improved quantification of grey and white matter atrophy

Introduction: The quantification of brain atrophy with MRI is vital for understanding the normal aging brain as well as understanding and treating people affected by neurodegenerative diseases such as Alzheimer’s disease and multiple sclerosis. Although several cross-sectional and longitudinal methods exist for the quantification of atrophy [1], SIENA and its cross-sectional counterpart SIENAx [2-3] are currently the most widely used in research and in clinical trials. SIENA has been shown to be robust and accurate (0.15% error) [3], but provides a measure only of whole brain atrophy. SIENAx, on the other hand, provides tissue-specific volumes, but computing change from serial cross-sectional measurements is much less precise (0.5-1% error) [3]. We propose a method to combine the sensitivity of SIENA with the tissue classification feature of SIENAx for a more accurate and precise measurement of tissue-specific atrophy.