There are many active research projects accessing and applying shared ADNI data. Use the search above to find specific research focuses on the active ADNI investigations. This information is requested annually as a requirement for data access.
| Principal Investigator | |
| Principal Investigator's Name: | Wei Qiu |
| Institution: | University of washington |
| Department: | Computer Science and Engineering |
| Country: | |
| Proposed Analysis: | In this project, we are interested in predicting biological age. We aim to use an explainable tree-based machine learning model to detect biomarkers for mortality prediction and age-related disease prediction and evaluate a measure of biological age derived from predicted probability. The traditional called linear models that are often used for predicting mortality makes many assumptions that are often untrue in the real world. In contrast, the tree-based model we use is able to capture more complex relationships between biomarkers and age-related phenotypes. This means it is generally better at making predictions in comparison to linear models. Moreover, in addition to providing accurate prediction scores, explainable tree-based models can help us uncover biological findings by explicitly showing how the complex model makes its predictions. Our preliminary results are on the UK Biobank dataset. We built tree-based mortality prediction and Alzheimer’s disease prediction models using demographics, lifestyle, and clinical features and interpreted it with state-of-art explanation techniques. Moreover, outputting a probability for predictions can be a confusing concept which may make it harder to communicate scientific findings to the general public. As such, we proposed a novel biological age derived from age-related disease or mortality prediction models. Our proposed biological age can successfully differentiate risky samples from healthy samples in the UK Biobank population. Also, we have performed GWAS to find the gene variants that are associated with different biological ages. We would like to request more phenotype data, genetic data and image data from the ADNI group. We would like to use ADNI data to calculate the biological age derived from Alzheimer’s disease prediction model and perform GWAS to find the SNPs that are associated with our biological age. Furthermore, we would like to compare our results with our preliminary results from UK Biobank to validate the generalizability of our biological age. |
| Additional Investigators |
