A robust and interpretable machine learning approach using multimodal biological data to predict future pathological tau accumulation

Study design and participants

Three separate cohorts were used to generate and test predictive models of regional future tau accumulation. For ADNI, ethical approval was obtained by the ADNI investigators. For BACS, ethical approval was obtained from the institutional review board at Lawrence Berkeley National Laboratory and the University of California, Berkeley. All ADNI and BACS participants provided written informed consent. Primary analysis was performed by investigators not involved in ADNI or BACS recruitment.

Two cohorts were drawn from the ADNI database: ADNI2/GO and ADNI3 (adni.loni.usc.edu). ADNI was launched in 2003 as a public-private partnership, led by Principal Investigator Michael W. Weiner, MD. A major goal of ADNI has been to examine biomarkers including serial magnetic resonance imaging, and positron emission tomography, with clinical and neuropsychological…

Read more…