VentRa. Validation study of the ventricle feature estimation and classification tool to differentiate behavioral variant frontotemporal dementia from psychiatric disorders and other degenerative diseases

ABSTRACT Introduction Lateral ventricles are reliable and sensitive indicators of brain atrophy and disease progression in behavioral variant frontotemporal dementia (bvFTD). Here we validate our previously developed automated tool using ventricular features (known as VentRa) for the classification of bvFTD vs a mixed cohort of neurodegenerative, vascular, and psychiatric disorders from a clinically representative independent dataset. Methods Lateral ventricles were segmented for 1110 subjects - 14 bvFTD, 30 other Frontotemporal Dementia (FTD), 70 Lewy Body Disease (LBD), 898 Alzheimer Disease (AD), 62 Vascular Brain Injury (VBI) and 36 Primary Psychiatric Disorder (PPD) from the publicly accessible National Alzheimer’s Coordinating Center dataset to assess the performance of VentRa. Results Using ventricular features to discriminate bvFTD subjects from PPD, VentRa achieved an accuracy of 84%, 71% sensitivity and 89% specificity. VentRa was able to identify bvFTD from a mixed age-matched cohort (i.e., Other FTD, LBD, AD, VBI and PPD) and to correctly classify other disorders as ‘not compatible with bvFTD’ with a specificity of 83%. The specificity against each of the other individual cohorts were 80% for other FTD, 83% for LBD, 83% for AD and 84% for VBI. Discussion VentRa is a robust and generalizable tool with potential usefulness for improving the diagnostic certainty of bvFTD, particularly for the differential diagnosis with PPD.