제목: Neuroimaging-based Deep Learning Research in Dementia and Normal aging
발표자: 이제연
날짜: 2023년 11월 30일 (목) 오4시
초록: Healthcare is an area with great opportunity for Artificial Intelligence. Particularly, the vast volumes of
neuroimaging data generated in healthcare, coupled with advancements in machine learning techniques,have been augmenting our abilities in analyzing and interpreting neuroimaging data, This, in turn, aids
researchers and clinicians in diagnosing, treating, and monitoring patients with neurological disorders.
This talk will cover the speaker’s recent studies involving the development of neuroimaging-based AI
models for investigating dementia and normal aging. The first study focuses on developing a deep
learning-based brain age prediction model using a large collection of fluorodeoxyglucose positron
emission tomography and structural magnetic resonance imaging and testing how the brain age gap
relates to degenerative syndromes including mild cognitive impairment, Alzheimer’s disease,
frontotemporal dementia and Lewy body dementia. The second study aimed to develop a deep
convolutional neural network model that synthesizes tau-PET brain images from cross-modality imaging
inputs and evaluate clinical implications of the AI-imputed tau PET.