From data to dialogue in setting research priorities for Lewy body dementia

Abstract By combining AI-assisted topic modelling and community engagement, this study aimed to advance the research landscape for Lewy body dementia (LBD). Despite being the second most common form of dementia, LBD remains under-recognized and under-researched, with greater care partner burden and poorer quality of life compared to other dementias. A multi-stage prioritization process first employed Latent Dirichlet Allocation to analyse global LBD literature (18,615 abstracts, 2006–2023), revealing an evolution from basic pathogenesis studies to advanced neuroimaging and AI applications post-2017 diagnostic guidelines. The subsequent stages involved consultation with the LBD community in Ireland through surveys, public involvement events, and consensus meetings, which identified crucial gaps in multidisciplinary approaches, prevention strategies, quality of life assessment, and palliative care. The study resulted in fifteen priority areas structured around three pillars: transdisciplinary research, lived experience, and education/information, emphasizing the development of effective diagnostic tools, quality of life frameworks, and treatment approaches.