Investigation of the relation between microbiotic changes and Alzheimer's Disease using machine learning on bile acids

Abstract: Alzheimer's disease (AD) is an increasing problem in modern society, both with regards to public health and cost of care. The causes of AD are not yet fully understood, and there is no cure or inhibiting drug. The aim of this thesis is to investigate the association between the bile acid profile as an indicator of dysbiosis and AD and mild cognitive impairment (MCI) using machine learning algorithms. The hypothesis that bile acid data can be used to predict AD or MCI at the time of diagnosis has been tested, and could not be confirmed. Somewhat better test results were obtained for the transition from normal cognitive function to MCI and from MCI to AD over time. Limitations relevant for this study included the possible uncertainties in the diagnostic patient data as well as in the relationship between bile acids and dysbiosis. The results from transitions in patient's diagnosis could warrant further research on the relationship between the bile acid profile or dysbiosis and changes in cognitive function. We suggest such research is conducted with more sophisticated models.