4,068 posters, 219 videos, 101 topics, 69 sessions, 14,365 authors, 6,128 institutions
ePostersLive by SciGen Technologies S.A. All rights reserved.
Alzheimer's Association International Conference 2026
July 12-15, 2026 | London, United Kingdom, and Online
9353
Machine Learning-based Prediction Of Brain Arteriolosclerosis Severity In The Uk-adrc Autopsy Cohort
Poster Presenter
Authors
Part of Topic
•We applied the SusStaln machine learning algorithm to 21 region-specific B-ASC data from the UK-ADRC autopsy cohort to identify distinct spatial-temporal progression patterns of arteriolosclerosis in an unsupervised data-driven pseudotemporal axis [1].
•Individuals with extremely low or high B-ASC burden (0, 1, 20, or 21 affected regions) were excluded prior to model fitting.
•Regional binary scores (0 = absence and 1 = presence) were converted to probability distributions using a Gaussian kernel (SD = 0.5).
•Models with varying numbers of subtypes K were evaluated using 10-fold cross-validation and the Cross-Validation Information Criterion (CVIC) (Figure 1).
•The optimal four-subtype (i.e., K = 4) was selected according to the lowest CVIC and highest log-likelihood and then participants were assigned to the most probable subtype and disease stage, with stage representing progression severity.
•Sequence uncertainty was quantified using 100,000 MCMC iterations and 25 random start points.
Comments
Email is required but is not shown publicly.
Loading comments...