Description
Biological clocks have enormous potential in health care. For instance, they can serve as motivators for healthier behaviors or objective health indicators for therapeutic or pharmaceutical interventions. For similar reasons, neuroimaging-based brain age measures have received growing interest. Previous research has shown that the chronological age can be accurately predicted through many structural MRI modalities, including T1-weighted and diffusion-weighted imaging. Importantly, the predicted age difference between the brain age (i.e., the estimated age) and actual age has been shown to have better performance in predicting dementia progression in mild cognitive impairment (MCI), when compared with several clinical scales (e.g., CDR) and genetic markers (e.g., APOE genotype).In this talk, I will review two main topics: (1) the common lifestyle factors that influence brain age measures; and (2) the state-of-the-art findings in applying brain age models to dementia diagnosis and prognosis. Following the review, I will report my initial explorations with brain age models, based on the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Specifically, we hypothesize that network-based brain age measures are more valuable for dementia diagnosis and prognosis by virtue of their stronger links with specific cognitive functions. To test this hypothesis, based on receiver-operating characteristic analysis, a range of network brain age models were compared against the global brain age model in terms of the performance in distinguishing among AD, MCI, and controls, as well as between stable and progressive MCI. Finally, I will introduce our ongoing brain age project at the Hong Kong Polytechnic University.
| Period | 8 Aug 2022 |
|---|---|
| Event title | BrainConnects 2022 |
| Event type | Conference |
| Location | Nagoya, JapanShow on map |
| Degree of Recognition | International |
Keywords
- Brain age
- Dementia