Developing a machine learning stack model to forecast the progression of mild cognitive impairment to Alzheimer’s dementia, using the Australian Imaging, Biomarker & Lifestyle (AIBL) Study dataset
Our RF-stack model relies on neuropsychological test scores and demographic data to predict the probability of an individual progressing from MCI to AD and accurately forecast the MMSE score after 3 years. This model achieved is comparable if not better prediction accuracy than the existing ML models using neuroimaging data. Further model optimization and validation using data collected from other longitudinal AD cohort studies are required.
Authors
- Colin L. Masters (ORCID: https://orcid.org/0000-0003-3072-7940)
- Liang Jin (ORCID: https://orcid.org/0000-0003-1695-3300)
- Paul Maruff (ORCID: https://orcid.org/0000-0002-6947-9537)
- Yijun Pan (ORCID: https://orcid.org/0000-0002-1442-3333)
- Benjamin Goudey (ORCID: https://orcid.org/0000-0002-2318-985X)
- Yihan Wang (ORCID: https://orcid.org/0000-0002-5353-6250)
- Chenyin Chu
Institutions
- The University of Melbourne (AU)
- Tohoku University (JP)
- Florey Institute of Neuroscience and Mental Health (AU)
- Monash University (AU)
Publication Details
- Journal
- Alzheimer s & Dementia
- Published
- 2024-12-01
- DOI
- https://doi.org/10.1002/alz.084645
- Primary Topic
- Dementia and Cognitive Impairment Research
- Type
- article
- Field-Weighted Citation Impact
- 0.00